Problem-Oriented Policing: Reflections on the First 20 Years Author Michael S. Scott October 2000 This publication was supported through Grant #98CKWXK052 from the Office of Community-Oriented Policing Services, U.S. Department of Justice. The opinions expressed herein are the author's and do not necessarily represent the official position of the U.S. Department of Justice. Suggested Citation Scott, Michael S. Problem-Oriented Policing: Reflections on the First 20 Years. Washington D.C.: U.S. Department of Justice, Office of Community Oriented Policing Services, 2000. Table of Contents Section Page Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii About the Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Summary of the Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 What Is the Purpose of the Report? . . . . . . . . . . . . . . . . . . . . . . . 35 A Brief History of the Spread of Problem-Oriented Policing . . . . . . . . . . . 36 Chapter 1 – Revisiting the Basic Elements of Problem-Oriented Policing . . . . . . . 45 What Is the Distinction Between Problem-Oriented Policing and Problem-Solving? . 45 What Does "Problem" Mean in Problem-Oriented Policing? . . . . . . . . . . . . . 46 How Should Problems Be Defined and Described? . . . . . . . . . . . . . . . . . .48 What Should the Police Be Concerned About in Problem-Oriented Policing? . . . . .49 Focusing on Community Concerns vs. Internal Concerns . . . . . . . . . . . . .49 Finding the Best Response vs. Merely Improving Current Responses and Systems .51 Focusing on Community Problems for Which the Police Should Assume Some Responsibility . . . . . . . . . . . . . . . . . . . . . . 51 What Does a Search for Underlying Conditions, Contributing Factors and Causes Really Mean? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Root Causes vs. Underlying Conditions . . . . . . . . . . . . . . . . . . . . 55 Causation vs. Blameworthiness . . . . . . . . . . . . . . . . . . . . . . . . 56 What Standards of Proof Should Apply in Analyzing Problems? . . . . . . . . . 58 How Should the Police Analyze Problems, and How Well Are They Doing So Now? . . .59 The Value and Limits of Analysis . . . . . . . . . . . . . . . . . . . . . . .59 Inadequate Analysis Resources . . . . . . . . . . . . . . . . . . . . . . . . 60 Problem Analysis Guides . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 The Action Research Model . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Accessing and Analyzing Police Data . . . . . . . . . . . . . . . . . . . . . 64 Searching for Relevant Research and Good Police Practices . . . . . . . . . . 65 Searching for Published Research . . . . . . . . . . . . . . . . . . . . . 65 Searching for Good Practices in Other Police Agencies . . . . . . . . . . .66 Searching for Good Practices Within a Police Agency . . . . . . . . . . . .66 What Does It Mean To Develop an Understanding of the Multiple and Competing Interests at Stake in Problems? . . . . . . . . . . . . . . . . . . . .67 What Does It Mean To Take Inventory of and Critique the Current Responses to Problems? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 How Should the Police Develop and Implement New Responses to Problems? . . . . . 69 Expanding the Range of Response Alternatives . . . . . . . . . . . . . . . . .69 Categorizing Response Alternatives . . . . . . . . . . . . . . . . . . . . . 70 What Does It Mean for the Police to Be Proactive? . . . . . . . . . . . . . .73 Who Should Be Involved in Problem-Oriented Policing, and How? . . . . . . . . . .74 How Should the Effectiveness of Implemented Responses Be Evaluated? . . . . . . .77 Process vs. Outcome Measurement . . . . . . . . . . . . . . . . . . . . . . .77 What Standards of Proof Should Apply in Evaluating Effectiveness? . . . . . .78 What Are the Specific Objectives of Problem-Solving Efforts? . . . . . . . . 80 Chapter 2– Putting Problem-Oriented Policing and Problem-Solving in the Context of the Whole Police Mission . . . . . . . . . . . . . . . . . . . . . . . . . 83 How Does Problem-Solving Fit in With Other Aspects of Police Work? . . . . . . . .83 What Are the Fundamental Objectives of Policing? . . . . . . . . . . . . . . .83 What Are the Various Operational Strategies of Police Work? . . . . . . . . . 85 Preventive Patrol . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 Routine Incident Response . . . . . . . . . . . . . . . . . . . . . . . . .86 Emergency Response . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Criminal Investigation . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Problem-Solving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88 Support Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 At What Levels Is Police Work Done? . . . . . . . . . . . . . . . . . . . . . 90 How Should the Police Integrate the Need To Address Community Problems With the Desire To Improve Administrative and Procedural Processes? . . . . . . . 93 Chapter 3 – Relating Problem-Oriented Policing to Other Movements in Police Reform and Crime Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .97 Team Policing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 Community Policing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .98 Crime Prevention Through Environmental Design . . . . . . . . . . . . . . . . . .100 Situational Crime Prevention . . . . . . . . . . . . . . . . . . . . . . . . . . 101 Crime Analysis and Compstat . . . . . . . . . . . . . . . . . . . . . . . . . . .103 Hot-Spot Policing and Crime Mapping . . . . . . . . . . . . . . . . . . . . . . .104 Broken Windows and Zero Tolerance . . . . . . . . . . . . . . . . . . . . . . . .106 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .107 Chapter 4 – Major Challenges to Advancing Problem-Oriented Policing . . . . . . . . .109 Setting an Agenda for Advancing Problem-Oriented Policing . . . . . . . . . . . .109 Advancing Problem-Oriented Policing Through Training, Research and Practice . . .111 How Will the Principles and Methods of Problem-Oriented Policing Be Taught? .111 How Will the Police Accumulate and Transfer Knowledge About Substantive Community Problems? . . . . . . . . . . . . . . . . . . . . . . . . . . . . .114 How Substantive Knowledge is Shared in the Police Profession . . . . . . .114 Writing Down Problem-Oriented Practice . . . . . . . . . . . . . . . . . .115 Collecting, Synthesizing and Disseminating Research and Practice on Specific Community Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . .116 How Can Problem Analysis Be Improved, and a Systematic Body of Research on Substantive Community Problems Be Developed? . . . . . . . . . . . . . . .117 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .120 Defining Roles for Others in Practicing Problem-Oriented Policing . . . . . . . .121 Are New Alliances Between the Police and the Community Healthy? . . . . . . .121 Are New Alliances Between the Police and Other Government Agencies Healthy? .122 What Should Be the Role of Prosecutors? . . . . . . . . . . . . . . . . . . .123 What Should Be the Role of Local Government Leaders? . . . . . . . . . . . . 126 Should the Police Be Held More Accountable for Reducing Crime, Disorder and Fear? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 Conclusion: How Will We Know if Problem-Oriented Policing Works? . . . . . . . . . . 131 Appendix A: An Analysis of the Best Submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing, 1993-1999 . . . . . . . 135 Appendix B: A Partial List of Problem-Focused Literature . . . . . . . . . . . . . . 165 Appendix C: A Summary of Interviews With Selected Problem-Oriented Policing Practitioners and Researchers . . . . . . . . . . . . . . . . . . . . . . . . . . . .171 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189 Foreword In the last three decades, several concepts have been advanced to structure efforts to improve policing. Among them have been team policing, neighborhood policing, community policing, problem-oriented policing, and, most recently, quality-of-life policing. With much overlap, each concept, as reflected in its name, emphasizes a different need, relegating other commonly advocated reforms to a secondary role, shaped to support that need. This volume traces the efforts to implement problem-oriented policing. The emphasis in problem-oriented policing is on directing attention to the broad range of problems the community expects the police to handle–the problems that constitute the business of the police–and on how police can be more effective in dealing with them. A layperson may think this focus elementary on first being introduced to it. Indeed, laypeople probably assume that police continually focus on the problems they are expected to handle. But within policing, this focus constitutes a radical shift in perspective. Problem-oriented policing recognizes, at the outset, that police are expected to deal with an incredibly broad range of diverse community problems–not simply crime. It recognizes that the ultimate goal of the police is not simply to enforce the law, but to deal with problems effectively–ideally, by preventing them from occurring in the first place. It therefore plunges the police into an in-depth study of the specific problems they confront. It invites consideration of a wide range of alternatives, in addition to criminal law, for responding to each specific problem. Thus, problem-oriented policing draws the police away from the traditional preoccupation with creating an efficient organization; from the heavy investment in standard, generic operating procedures for responding to calls and preventing crime; and from heavy dependence on criminal law as the primary means for getting their job done. It looks to increased knowledge and thinking about the specific problems police confront as the driving force in fashioning police services. The introduction of a new concept to policing is not a neat process, especially in the United States, where approximately 17,000 police agencies operate with a high degree of independence and a record of strong resistance to change. One would be naïve to expect dramatic results in a short time. Indeed, when related to the total field of policing, progress toward achieving the shift in emphasis called for in problem-oriented policing has, over the past two decades, been negligible–a project here, a cluster of problem-solving efforts there. While "problem-oriented policing" has become part of the policing vocabulary, pure examples of its implementation are hard to find; various permutations of the concept are more common. Nevertheless, there have been some indications of significant movement–examples of situations in which officers have identified a specific problem, subjected it to in-depth analysis, and implemented a fresh, novel response that is more effective in dealing with it. When this occurs, one sees the potential of the concept confirmed. And when one assembles these efforts, as occurs at the annual Problem-Oriented Policing Conference or in a publication such as this one, the results appear to be substantial. Michael Scott has taken on the ambitious task, in this report, of describing what has happened in claimed efforts to implement problem-oriented policing over the past two decades, both in the United States and abroad. He is uniquely equipped to have done so, having been directly involved with me in developing the concept in his days as a student; having had a wide range of experiences in training, implementation and research relating to the concept; and having, throughout this period, been a valued colleague. This was an extraordinarily difficult project. Except for mail and telephone surveys, which have proved unsatisfactory in other contexts, and penetrating field inquiries, which are very costly, there is no way to quantify what has occurred–to assess the influence, if any, that the advocacy of problem-oriented policing has had on the minds and operations of police scattered about this vast country and the world. We know that much literaturehas been distributed and numerous training sessions have been conducted, but we know little about the results of these efforts. Despite these limitations, Scott has, by making maximum use of an eclectic collection of sources and some limited field work, succeeded in producing an extraordinarily useful description of what has occurred under the label of problem-oriented policing, appropriately qualifying, at critical points, the sweep of his findings. Given my own effort to follow these same developments, his summation appears both comprehensive and objective. He distinguishes the strong efforts from the weak; identifies the several misunderstandings and distortions of the concept, providing helpful clarifications; reports on the permutations of the concept–both those that have advanced and sharpened the original goals, and those that have detracted from them; and describes the conditions that have facilitated implementation, and the barriers that have been encountered. He has located and made the fullest use of published materials that relate to the topic. His collection of references is the most comprehensive bibliography that has been compiled on problem-oriented policing; his detailed footnotes enrich the manuscript. Throughout the report, he effectively uses specific case studies from the growing collection of problem-oriented policing projects to illustrate his points. In 1990, when I published Problem-Oriented Policing, I wrote, in the introduction, that the concept of problem-oriented policing is open-ended; that it invites criticism, alterations, additions, and subtractions; and that my intent was to stimulate others to contribute to further developing this overall approach to improving policing. Given the vast arena of policing in democratic societies, I had not contemplated how difficult it would be to sort through what has occurred, to "separate the wheat from the chaff." Looking at what has happened in the past 20 years, Scott extracts some of the most significant developments: the extent to which beat-level police officers, with an abundance of latent talent, have grasped the concept and produced remarkable results; the linkage problem-oriented policing has to the parallel development of situational crime prevention, and how the two can enrich each other; and the degree to which implementation efforts reflect the commitment of individuals rather than agencies. Struggling with the difficulty of integrating problem-oriented policing into an agency that is often preoccupied with responding to calls, handling emergencies and investigating crime, Scott himself contributes to advancing the concept by offering some solid suggestions, based on his experience and research, for achieving that integration. And, in the final section of the report, he explores, in some detail, the most pressing questions and issues that have arisen from the efforts to date, and sets forth ways these might best be addressed. For those who are interested in advancing problem-oriented policing and who have read my 1990 work on the subject, this report should be read as a companion volume, updating developments over the past 10 years. It will be of some help to those who are looking for specific guidance in addressing a specific problem. Its greater value, however, will be in the contribution it makes to advancing the fundamental point that improvements in policing–whether in organization, staffing, operations, or even relationships with the community–can best be achieved by focusing more directly on the business of the police–on the varied problems that the community expects the police to handle–and, through study and experimentation, on developing a wider range of new, more specific and more effective ways to deal with them. Herman Goldstein Madison, Wisc. January 2000 Acknowledgments I genuinely appreciate the insights I gained from my visits to the Edmonton, Alberta, Police Service; Savannah, Ga., Police Department; Charlotte-Mecklenburg, N.C., Police Department; Madison, Wisc., Police Department; San Diego Police Department; Reno, Nev., Police Department; Sacramento, Calif., Police Department; Merseyside, England, Police; and Lancashire, England, Constabulary. I also appreciate the insights I gained from speaking at length with the following people: Chris Braiden, Michael Bradshaw, P.J. Duggan, Norm Lipinski, and Apollo Kowalyk (Edmonton); Bob Heimberger (St. Louis); Caroline Nicholl (Washington, D.C.); John Eck and Lorraine Green Mazerolle (University of Cincinnati); Randy Gaber and Luis Yudice (Madison); David Kennedy and Malcolm Sparrow (Harvard University); Nancy McPherson (Seattle); Nancy La Vigne (National Institute of Justice); Ron Clarke (Rutgers University); Rana Sampson (San Diego); Ron Glensor (Reno); Steve Segura (Sacramento); Tim Hope (Keele University, England); Mike Chatterton (University of Manchester, England); Ken Pease (Huddersfield University, England); Michael Barton, Jim Masterman, and Paul Stephenson (Lancashire, England); Brian Gresty, Keith Taylor, Nikki Holland, and Andy Fisher (Liverpool, England); Gloria Laycock (Home Office, England); Dennis Nowicki and Darrel Stephens (Charlotte); Drew Diamond (Tulsa); and Dan Reynolds (Savannah). I am grateful to the following police officials, who permitted me to spend time with them in the course of their regular duties, or who provided me with valuable information: Blair Edl, Jeff Wilks, Steve Camp, Palle Nicollajsen, Tom Farquhar, Kim Moore, Sylvia Church, and Rick Gagnon (Edmonton); Simon Byrne (Liverpool); Jonathan Lewin (Chicago); John Davenport (Madison); Jerry Hoover, Ondra Berry, Kim Gibson, Jacson Mahler, Steve Bigham, Doug Evans, Ken Bunker, and Andy Kachurak (Reno); Guy Swanger and Duane Voss (San Diego); and Richard Zapal and Bill Harvey (Savannah). I appreciate the substantive insights and support for the project provided by Karin Schmerler, Ellen Scrivner and Gil Kerlikowske, and the assistance in the publication process by Megan Tate Murphy and Rita Varano, all of whom worked at the Office of Community-Oriented Policing Services during my fellowship. I further appreciate the helpful comments and suggestion from three anonymous peer reviewers. Finally, I am grateful for the good services and good humor of my editor, Suzanne Fregly, who edited every line of this work–except this one. About the Author Michael Scott is an independent police research and management consultant. He researched and wrote this report as a Visiting Fellow at the U.S. Department of Justice, Office of Community Oriented Policing Services (COPS Office). Scott has helped implement problem-oriented policing from management positions in several police agencies. He served as the chief of police at the Lauderhill, Fla. Police Department where he founded that police agency in 1994 under the principles of problem-oriented policing. He was special assistant to the chief of police at the St. Louis Metropolitan Police Department where he was primarily responsible for overseeing the problem-oriented policing implementation in St. Louis. He was director of administration at the Fort Pierce, Fla., Police Department where he trained officers and guided a reorganization of the department in accordance with the principles of problem-oriented policing. He served as legal assistant to New York City Police Commissioner Benjamin Ward where he helped guide the development of a major community policing effort, the Community Patrol Officer Program. Scott has also worked as a senior researcher at the Police Executive Research Forum (PERF) where he developed and delivered problem-oriented policing training to police agencies around the country and helped organize the first national conference on problem-oriented policing. He has served since 1996 as a judge for the Police Executive Research Forum's Herman Goldstein Award for Excellence in Problem-Oriented Policing and has been a regular presenter at PERF's annual conference on problem-oriented policing. He was the 1996 recipient of PERF's Gary P. Hayes award for leadership in improving police service. He is the author of Managing for Success: A Police Chief's Survival Guide and co-author of Deadly Force: What We Know. A Practitioner's Desk Reference to Police-Involved Shootings in the United States and Challenge to Change: The 21st Century Policing Project, all published by PERF. He is also co-editor of Tackling Crime and Other Public-Safety Problems: Case Studies in Problem-Solving published by the COPS Office. Scott began his policing career as a police officer in the Madison, Wisc. Police Department. He was a research assistant to Herman Goldstein during the first test of problem-oriented policing in the early 1980's in which researchers helped the police explore the problems of drunk driving and repeat sex offenders in Madison. He was again Goldstein's research assistant on a project that culminated in the publication of Goldstein's book, Problem-Oriented Policing. Scott holds a J.D. from the Harvard Law School where he was also an observer of the early Executive Sessions on Community Policing held at the Kennedy School of Government. He also holds a B.A. in Behavioral Science & Law and Sociology from the University of Wisconsin-Madison where he was a student of Herman Goldstein's and where he wrote his undergraduate thesis on the occupational perspectives of police patrol officers. Summary of the Report Introduction What is the Purpose of the Report? This summary report describes how Herman Goldstein's problem- oriented policing framework has been developed and, at times, distorted in the many efforts to make it a standard way of policing. I prepared the report as a Visiting Fellow to the U.S. Department of Justice's Office of Community Oriented Policing Services (COPS Office). I drew upon my personal experience, reviewed relevant literature and problem-oriented policing project reports, visited selected police departments, attended conferences, and talked extensively with Herman Goldstein and others well-versed in problem-oriented policing. A Brief History of the Spread of Problem-Oriented Policing The first formal experimentation with Goldstein's model of problem- oriented policing occurred in Madison, Wisc., in 1981 when Goldstein and his associates worked with the Madison Police Department exploring the community's response to drinking drivers and repeat sex offenders. Around 1982 the police in London and in Surrey, England undertook their own experimentation with the concept. The Baltimore County Police Department formally introduced Goldstein's problem- oriented policing model into its COPE unit's operations in 1983 and the Newport News, Va., Police Department followed suit in 1984. A number of other police agencies began to incorporate at least some of the problem-oriented policing methodology into broader community policing efforts during the 1980s. In 1994, the COPS Office began to link funding for new police officers to the broad concept of community policing of which problem-solving was a key element. The COPS Office was required by law to advance community policing generally, but outside of a few of its competitive funding programs, most of its large funding programs did not require that recipient police agencies engage more specifically in problem-oriented methods. While the link between problem-solving and community policing in this large federal funding program has yielded many benefits, the linkage has also blurred the distinction between problem-oriented policing and community policing. Many police agencies in the United States and Canada, and a growing number in the United Kingdom, Australia, New Zealand, the Netherlands, South Africa, and Scandinavia, report that they are now engaged in problem-oriented policing in some fashion. While there is no easy way to quantify the number of police agencies engaged in problem-oriented policing, much less to gauge the precise nature and quality of those efforts, it is safe to say that far more agencies claim to be engaged in problem-oriented policing today than at any other time. Problem-oriented policing continues to advance across the police field, even while the adoption of problem-oriented policing into particular police agencies seldom happens in a linear fashion. Interest in the concept and commitment to its implementation rises and falls in response to many internal and external factors. Changes in leadership, competing priorities or simply inertia can alter the course of implementation. Accordingly, one might reach different conclusions about the vitality of problem-oriented policing depending on whether one was looking only at selected police agencies or at the police field as a whole. However slow, modest and uneven the movement in problem-oriented policing has been, it is now a central part of at least the language of modern police management. But along with the rise in popularity of problem-oriented policing has come a certain amount of distortion of its original meaning. The next chapter describes how the elements of Goldstein's ideal model of problem-oriented policing have developed in practice. CHAPTER 1: REVISITING THE BASIC ELEMENTS OF PROBLEM-ORIENTED POLICING Revisiting the Basic Elements of Problem-Oriented Policing Herman Goldstein's problem-oriented policing concept is a comprehensive prescription for improving the way in which the police do business. It calls for the police to understand their work in a new light, to recognize that what they are called upon to do is to address a wide range of problems that threaten the safety and security of communities, including, but not limited to what is commonly viewed as serious crime. The concept calls for the police to improve their understanding of the underlying conditions that give rise to community problems and to respond to these problems through a much wider range of methods than they have conventionally used. Behind the seemingly common-sense simplicity of the basic elements of problem-oriented policing lie real challenges for the police, communities and the rest of government to fully understand and implement them. What is the Distinction Between Problem-Oriented Policing and Problem- Solving? In its broadest sense, the term “problem-oriented policing”, as used by Goldstein, describes a comprehensive framework for improving the police's capacity to perform their mission. Problem-oriented policing impacts virtually everything the police do, operationally as well as managerially. The term “problem-solving” is a more limited notion; it describes the mental process that is at the core of problem-oriented policing. What Does "Problem" Mean in Problem-Oriented Policing? The current literature on policing finds the term “problem” popping up everywhere. While it refers to many different matters, Goldstein's use of the term in the context of problem-oriented policing is highly specific. He used the term to convey the notion that one can classify, package and understand police work in a new way, as an aggregation of incidents that share certain common features. The precise understanding of the term “problem” remains much in need of reinforcement. How Should Problems be Defined and Described? How one defines a problem greatly influences how one will address it. One can define or describe problems in a variety of ways. One can describe them in terms of what the offensive behavior is, who the people involved are, when the problem occurs, or where the problem occurs. These various descriptors obviously are not mutually exclusive. The descriptor is merely a shorthand way of describing the entire problem. However one describes a problem in shorthand, one must address the offensive behavior. This is important for several reasons. Without a clear focus on specific forms of offensive behavior, the police run the risk of adopting overbroad or ineffective responses. While it is sometimes convenient to describe problems in terms of a class of people or even one individual, it is dangerous morally, ethically and legally for the police to treat a person or people as the problem itself. Shorthand labels can also mask important distinctions between legitimate and illegitimate behavior. Describing problems as the "drug problem" or even the "narcotics problem" is so broad as to be nearly useless. Characterizing problems with broad labels like "drugs," "violence," "disorder," "neighborhood decline," or "juveniles," without specifying the behavior at issue, often results in a simplistic analysis of the problem and, consequently, to hopelessly inadequate responses. Overly broad definitions of problems also create the risk that the police will be drawn into trying to address aspects of a large problem that are well beyond their capacity or mandate. Any shorthand label for a problem should be followed by a more complete and exacting description of the specific offensive behavior. I have sometimes reminded myself and others of this rule of problem-oriented policing in grammatical terms, by saying, "If you don't have a verb, you don't have a problem." Forcing oneself to include a verb in the description of the problem helps maintain the appropriate focus on problematic behavior. For example, a problem described in shorthand as "transients in a public park" is made more explicit by labeling it "transients sleeping and panhandling in a public park." This simple change draws one's attention to the behavior and not merely to the status of the persons involved. When problem-solvers redefine or reclassify the problem on the basis of preliminary analysis, this leads to conceptually clearer and more manageable initiatives. Asking whether the problem looks any different upon closer analysis remains a vital step in the problem- solving process, but it is too often overlooked. What Should the Police Be Concerned About in Problem-Oriented Policing? Focusing on Community Concerns vs. Internal Concerns Goldstein's starting point for articulating the problem-oriented approach was that police managers should focus on how their agencies address community problems and not merely on how their agencies are administered and organized. Getting police to refocus on community concerns is in itself a significant challenge. Police administrators and officers understandably focus on the organizational, administrative and procedural problems that directly affect their physical safety, career opportunities, financial status, and general occupational contentment. The police are no different in this regard from practitioners in other fields. Ask most medical practitioners today to list their problems, and one can expect to find managed care higher on the list than emphysema or heart disease. Teachers talk more about classroom discipline than how to teach algebra more effectively. Existing case studies in problem-oriented policing demonstrate that the police are capable of using problem- solving methods on substantive community problems. But if the police continue to focus exclusively or primarily on internal organizational problems, even if they apply some problem-solving methods toward their resolution, then problem-oriented policing will have failed on its face. Finding the Best Response vs. Merely Improving Current Responses and Systems Problem-solving inquiries should seek the best response to the substantive problem at hand, and not merely seek to improve current responses and systems. This distinction is a subtle but important one. A fair number of problem-oriented policing projects are essentially efforts to improve a criminal justice or investigative process, devoid of a careful inquiry into whether that process is the most effective means of addressing the problem in the first place. For example, in recent years, a number of police agencies have recognized value in establishing more collaborative working relationships with probation and parole agencies. Accordingly, a number of problem-oriented policing projects have set about finding ways for the police and probation and parole agents to more effectively and efficiently supervise people under conditional release. The underlying logic, of course, is that more effective and efficient supervision will reduce the levels or seriousness of crimes committed by those people. In many instances, however, the assumption that supervision of previously convicted offenders is the best response to the problem goes unexplored and unchallenged. The value of police-probation and parole collaboration becomes stronger if it is first clearly established that improved supervision will result in substantial improvements to the specific community problem. Focusing on Community Problems for Which the Police Should Assume Some Responsibility Goldstein has advocated that the police recognize their role in society as being broader than enforcing the criminal law. At the same time, however, he has argued that the police mandate must not be unlimited. If the police become too involved in every government and quasigovernment function, they risk eroding balances of power in local and even national government. Police agencies run the risk of overextending their expertise and resources–trying to achieve objectives about which they have little or no expertise. By expending resources on newly adopted mandates, they risk devoting too few resources to conventional mandates. The community problems the police should focus on are those that fall within their mandate as it is defined for each agency. In the era of community policing, that mandate has been expanded, partly by the police themselves. Police departments everywhere are initiating programs in which police officers adopt roles of counselors, teachers, coaches, and brokers of charitable works. The most common justification offered for adopting these new roles is that the police can inculcate good moral and civic habits in the community, and as a result, some unspecified measure of offending will be reduced. Too often though, the police adopt these roles for other purposes–to improve their community image or deflect criticism of other, objectionable, police practices. Under a problem-oriented policing approach, the police would recognize how functions like moral education, youth recreation and charity are integral to public safety, but would not see their role as one of providing these services directly, at least not permanently. The key for the police is first, to establish some sense of ownership or responsibility for a community problem, and if the problem falls within the police mandate, either address it themselves, broker ownership to some other entity or, in some instances, merely refuse to accept ownership. The police may join with many divergent entities in studying a problem, but ultimately the responsibilities for various responses should be apportioned among those entities according to their resources and competencies. A good example was provided by the Glendale, Calif., Police Department when in 1997 it helped develop a new program for day laborers that directly responded to legitimate police interests in reducing crime and disorder. The police did not assume responsibility, however, for actually running the program. Similarly, the Fontana, Calif., Police Department in 1998 helped develop a new assistance program for transients that achieved similar objectives without assuming the large responsibility of administering the program. What Does a Search for Underlying Conditions, Contributing Factors and Causes Really Mean? Root Causes vs. Underlying Conditions The search for contributing factors and underlying conditions is sometimes confused with efforts to address the broadest of social and psychological factors that contribute to crime and disorder, factors often referred to as the "root causes" of crime and disorder. Associating problem-oriented policing with a search for "root causes" is misguided. Problem-oriented policing looks for the deepest underlying conditions that are amenable to intervention, balancing what is knowable with what is possible. Many of what are commonly thought of as "root causes" are beyond the police's capacity to change. Causation vs. Blameworthiness Causation and blameworthiness are not the same thing. Problem- oriented responses affix responsibility on those most capable of effecting lasting improvements to the conditions that give rise to the crime and disorder. Those most capable of addressing a problem may not be those most blameworthy for that problem. To many police officers, steeped in the legalistic traditions of assigning blame through the enforcement of the law, the process of spreading out responsibility for responding to problems does not come naturally. Effective problem-solving places a higher priority on improving the overall response to the problem than on assigning blame for the problem. This is why it is so critical that the police develop effective working relationships with those affected by a problem, relationships built in a spirit of mutual trust, to overcome the natural defensiveness that accompanies discussions of causation, blame and responsibility. What Standards of Proof Should Apply in Analyzing Problems? Some police scholars advocate setting high standards of social scientific proof in problem-oriented policing, standards that can best be met by rigorous application of experimental testing conditions. Other scholars have advocated a more flexible standard of proof that takes into account the severity of the problem, the costs of being wrong, the research skills of the problem-solvers, the practicality of various research methods, the body of existing knowledge about the particular type of problem, and so forth. As a practical matter, the standard of proof that ultimately will prevail varies from problem to problem and place to place. Within the broad limits of the law, what stands as an acceptable response to any particular problem depends on what is acceptable to the local community, at least to those members who are paying attention to the problem and who can exercise influence on the particular policymakers. How Should the Police Analyze Problems, and How Well Are They Doing So Now? Problem analysis remains the aspect of the concept most in need of improvement. This is partly due to inadequate resources and weak analysis methods, but it is also due to the different ways in which the police and researchers understand how analysis contributes to addressing problems. The Value and Limits of Analysis In order for the police to commit adequate resources to analyzing problems, they must first fully appreciate how analysis can improve their responses to problems. In order for researchers to help the police with analysis, they must appreciate the practical concerns of and demands upon the police with respect to community problems. (These issues are discussed more fully in chapter 4.) A thorough problem analysis, at a minimum, means fully describing the problem, describing the multiple and often conflicting interests at stake in the problem, calculating the nature and costs of the harm arising from the problem, and taking inventory of and critiquing the current responses to the problem. In the problem-oriented policing model, problem-solvers, whether they be police practitioners or researchers, should be open to doubt about things they thought they knew about the problem and insist upon proving or disproving matters with objective evidence. They must balance the desire to be certain and precise with the practical difficulties in being so. They must recognize what data can and cannot tell them. They should be interested in learning how similar problems have been analyzed and addressed elsewhere while at the same time recognizing how their local situation might be different. They must ask the right questions and not waste effort finding answers to questions of no practical significance. They must balance the need to reflect on problems with the need to act upon them. These are no small challenges and they require that both police practitioners and researchers adjust and adapt the conventional ways in which they analyze problems and decide how to respond to them. Inadequate Analysis Resources Problem analysis can fall short of ideal without adequate time to complete the analysis and the research expertise necessary to do so properly. Research expertise is valuable for setting up an appropriate methodology for conducting the inquiry, ensuring data are complete and reliable, and applying statistical data analyses from which valid conclusions can be drawn. Some problem-solving and analysis guides have gone a long way toward providing street officers with some basic understanding of problem-solving methodologies, but they do not provide the same level of expertise as can trained and experienced researchers. The Action Research Model Goldstein envisioned an action research model in which the researcher is an integral part of a team of people working toward some particular result. The researcher not only collects and analyzes data and draws conclusions, but also proposes interventions along with others trying to intervene in the problem. This research model seeks to balance an outside researcher's independence and objectivity with a pragmatic interest in achieving certain results. Accessing and Analyzing Police Data Computerized record-keeping has been a boon to problem-oriented policing. Data that just a few years ago would have been enormously difficult to retrieve are now available at the touch of a few buttons. Unfortunately, the ease of searching and analyzing large volumes of aggregate coded data too often leads problem-solvers to skip a more detailed analysis of the written narratives in individual police reports. Police report narratives contain many of the more useful insights about problems. Searching for Relevant Research and Good Police Practices An important aspect of problem analysis should be a review of the literature on that problem. That literature might be in published books and articles, or in unpublished reports from within and outside the police agency. In practice, however, literature reviews conducted as part of a problem-solving project are rare. Police practitioners often do not have the benefit of assistance from researchers or do not have access to research libraries. Unfortunately, even if police had more access to research libraries, or if trained researchers were conducting a literature review, their search would not be productive with respect to many types of problems. While there is more relevant research on some community problems than many police officers realize, it is far less than one might expect given how common many problems are and how many public resources are spent trying to address them. Again, compared to the body of literature in most other professions, the amount of published research about common community problems seems miniscule. There simply isn't enough quality research conducted to reliably inform the police about what does and does not work with respect to most crime and disorder problems. The police can also improve their responses to community problems by studying their own and other agency's past efforts to address similar problems. Reports about problem-solving initiatives are a valuable source of knowledge from which to draw, even if those initiatives did not apply rigorous research methods. Unfortunately, most police agencies do not routinely prepare detailed reports on most of their problem-solving initiatives. Some police managers are reluctant to impose what might be perceived as excessive reporting requirements on officers whom they do not want to discourage from engaging in problem-solving. While this is understandable as managers try to coax officers into policing in a different way, a lot of knowledge about how various problems have been handled has been lost. Some police agencies have created computer records, project reports, forms and newsletters to document problem-solving efforts. These have great potential to help officers search for solutions to common problems and to teach officers problem-solving skills through real examples. Somehow, more police-led problem-solving efforts must be documented in writing and police managers must then make these resources accessible and encourage that they be reviewed as a standard step in future problem analysis. Compared to the record-keeping systems and reporting requirements for calls for service, incident reports and criminal investigations, the state of record-keeping and reporting for problem-oriented activities is rather primitive. Ultimately, police agencies must assign the same degree of importance to the official records related to problem- oriented initiatives as they do other official records. What Does It Mean to Develop an Understanding of the Multiple and Competing Interests at Stake in Problems? Many problem-oriented policing initiatives fail to take complete account of all the interests at stake with respect to the problem. This matter of accounting for the various interests is often simplified into a mere inventory of stakeholders. In fact, most stakeholders have multiple and competing interests in a problem. Exploring interests in a problem begins by asking what the social interests are in the problem, and then asking what the government interests are in the problem. Not all social interests should be government interests. Once one identifies the government interests, one can turn to asking what police interests are at stake. If the police conclude they have no interest at stake in the problem, there is little justification for their continued involvement regarding it. There are many social problems in which the police are well-advised not to become embroiled. In exploring the various nonpolice interests at stake in a problem, it is important to go beyond the most visible and obvious interests. There are often hidden commercial interests involved in many problems, as well as latent social prejudices and biases. These interests should at least be brought out in the open, where they can be considered. The careful probing of these interests is among the most enlightening parts of the problem- solving process. What Does It Mean to Take Inventory of and Critique the Current Responses to Problems? Many problem-oriented project reports allude only briefly to the inadequacy of current responses, mainly by making the obvious assertion that a new response is needed. Current responses are often described briefly and generally, and casually discredited as being ineffective. One often reads in problem-solving project reports cursory assessments of current practices such as "the traditional response of handling calls, taking reports and making arrests was not working". But brief and general descriptions like these are not illuminating and, often, not entirely accurate. Individual police officers frequently develop their own innovative responses to problems, responses that are not fully and accurately encompassed in their agency's standard operating procedures. Other agencies and groups may be responding to problems in ways that the police are unaware. Some responses, however traditional, may prove more effective upon closer analysis than they might initially appear. It takes some effort to discern precisely how problems are being handled and to what extent current practice is effective. The flip side of dismissing the value of current conventional responses is, when faced with a problem that is not getting adequate attention, to simply increase the effort put into conventional responses, without carefully considering their strategic value. Many reports on problem-solving projects leap quickly to judgments that greater police presence, more arrests, more certain prosecution, or stiffer penalties are the best response to a problem. Such judgments are often made without examining the effectiveness of existing levels of these interventions. How Should the Police Develop and Implement New Responses to Problems? Expanding the Range of Response Alternatives Goldstein urges the police to greatly expand their range of alternative responses to problems, responses beyond the conventional increased police presence and criminal arrests. A wide range of responses is emerging from reports of problem-oriented policing projects. New responses to chronic problems should be well-considered, following logically from careful problem analysis, not merely a few clever ideas thought up as a hasty reaction. Clever ideas have some value, but without a clear line of reasoning that articulates the basis for the new response, they do not add much to the body of professional knowledge from which other police agencies and communities can draw. Police agencies often copy other agencies' clever or innovative ideas. But without first assessing how they might work in the local situation, these ideas might well prove ineffective. It is also unfortunate when the police launch problem-solving initiatives with a preferred response in mind. The subsequent problem analysis serves more to justify the preferred response than to inform the decision- maker about the nature of the problem. What Does It Mean for the Police to Be Proactive? Problem-oriented policing prefers proactive responses to reactive responses. Proactivity means first, that responses to problems should prevent future harm, and not just address past harm, and second, that the police should speak out about community problems that are not being adequately addressed. Advocating that the police should be more proactive should not be understood as an endorsement of overaggressive police tactics. Goldstein has described a continuum of pressure the police might apply to get other entities to assume or share ownership for community problems. The degree of pressure the police apply should depend on the strength of the evidence they have regarding the nature of the problem and its causes. From least to greatest pressure, the police can do the following to get others to accept ownership or responsibility for problems: • develop educational programs regarding responsibility for the problem • make a straightforward informal request of some entity to assume responsibility for the problem • make a targeted confrontational request of some entity to assume responsibility for the problem • engage another existing organization that has the capacity to help address the problem • press for the creation of a new organization to assume ownership of the problem • shame the delinquent entity by calling public attention to its failure to assume responsibility for the problem • withdraw police services relating to certain aspects of the problem • charge fees for police services related to the problem • press for legislation mandating that entities take measures to prevent the problem • bring a civil action to compel entities to accept responsibility for the problem. Who Should Be Involved in Problem-Oriented Policing, and How? Goldstein has always encouraged line officers' involvement in problem-oriented policing, but he did not anticipate that they would emerge as the leaders in addressing problems. He imagined that command-level police officials and research collaborators would lead most problem-oriented initiatives. In practice, line officers have led many projects, even when the scope of the project has been quite large. In one respect, this provides some evidence of the talent line- level police officers have, talent that police managers do not fully appreciate or exploit. But it may also be that supervisory and command-level officers are not sufficiently engaged in practicing problem-oriented policing. Getting command-level officers involved in, and holding them accountable for, addressing community problems is critical, but there are pitfalls if not done properly. When commanders are held accountable for problem-solving, problems tend to get defined in their terms, and less so in the terms of those most familiar with problems–the community and line officers. The ideal level of police authority for providing leadership in problem-oriented policing projects depends on the scope of the problem being addressed. As a general proposition, supervisors should provide active leadership in localized beat problems; commanders in intermediate-level problems; and top commanders, perhaps including the chief executive, in communitywide problems. In every instance, line officers should be encouraged to be as involved as their time and abilities permit. That higher-ranking police officials seldom actively lead problem- oriented policing initiatives suggests that the problem-solving method of operations has yet to achieve a high level of importance in most police organizations. It tends still to be viewed as something that only beat police officers do. Police chiefs need to pay at least as much personal attention to substantive community problems as they do to administrative and political concerns. Some command officers, to the extent they are supportive of problem-oriented policing, see their role as administrative manager, ensuring that systems are in place and resources available for line-level problem-solving. This is fine as far as it goes, but without more personal and direct command-level leadership, few large and complex community problems are likely to be taken on in a sophisticated, problem-oriented way. Line-level officers lack the requisite resources in most instances to conduct the sort of analysis and effect the sort of responses necessary to bring about substantial improvements in communitywide problems. Given the abundance of communitywide problems in every jurisdiction, supervisors and command-level officers need to become more personally engaged in problem-oriented policing. How Should the Effectiveness of Implemented Response Strategies Be Evaluated? Process vs. Outcome Measurement Perhaps the single greatest source of confusion relating to the evaluation of problem-oriented policing initiatives surrounds the distinction between the measurement of processes and the measurement of outcomes. The measurement of processes is the documentation of the actions taken in implementing responses, and an assessment of whether the responses were actually implemented as intended. The measurement of outcomes is the assessment of the ultimate impact the responses had on the problem, as defined (i.e., Did the problem improve, worsen or remain the same? Were the outcome objectives achieved?). These two different types of evaluation are often confused. Most commonly, evaluators limit their inquiry to determining how well and to what degree the police and others actually implemented their plan of action. While this information is vitally important, it cannot be substituted for some inquiry about what effect the plan of action, however well- implemented, had on the problem. Ideally, a problem-oriented policing project will include measurement of both processes and outcomes. What Standards of Proof Should Apply in Evaluating Effectiveness? Goldstein acknowledges the many difficulties in establishing precise and certain conclusions in the complex world of human behavior where policing occurs, and accordingly, he is willing to settle for less than the most rigorous tests of effectiveness in most instances. Insisting on rigorous standards, however justified theoretically, would likely stifle much experimentation with the problem-oriented concept. How precise and certain one has to be in problem-oriented policing depends greatly on the consequences of being wrong. What Are the Specific Objectives of Problem-Solving Efforts? The Newport News study first delineated a set of generic legitimate objectives in problem-solving. It grouped those objectives into five categories: 1. totally eliminate a problem; 2. substantially reduce a problem; 3. reduce the harm created by a problem; 4. deal with a problem better (e.g., treat people more humanely, reduce costs or increase effectiveness); and 5. remove the problem from police consideration. The fifth objective, removing problems from police consideration, differs from the first four in that it does not directly address the question of whether the problem, as experienced in the community, will be improved by removing it from police consideration. Taken to the extreme, the police could claim success in problem-oriented policing merely by working to absolve themselves of responsibility for problems. If shifting responsibility for addressing a problem to another entity results in more effective handling of the problem, then the objective is legitimate. If such a shift results merely in some efficiency gains for the police, then it may have some merit, but one cannot consider it an effective resolution. Often neglected in evaluations are indicators of the prevalence of the problem, the net harm caused by the problem, the possible displacement of the problem, the possible unintended benefits of the response, and an accounting of the total costs arising out of the problem and responses to it. CHAPTER 2: PUTTING PROBLEM-ORIENTED POLICING AND PROBLEM-SOLVING IN THE CONTEXT OF THE WHOLE POLICE MISSION How Does Problem-Solving Fit in With Other Aspects of Police Work? Where does problem-solving leave the conventional tasks and methods for responding to calls for service or investigating crimes? How should police administrators who endorse problem-oriented policing reconcile the demands on their agencies to continue performing conventional police tasks with the new demands to engage in substantive problem-solving? Answering these questions requires returning to some first principles of policing. Goldstein argued that to understand policing properly, one has to distinguish between the objectives the police are trying to achieve and the methods they use to achieve them. Accordingly, investigating crimes and enforcing laws, long thought of as basic policing objectives, are not objectives in and of themselves, but rather methods for achieving other, more broadly stated, objectives. Problem-oriented policing, then, is concerned with expanding on and improving the methods the police use to achieve their more fundamental objectives. What Are the Fundamental Objectives of Policing? The fundamental objectives of policing are the ultimate purposes for which police agencies have been created. Goldstein characterized the fundamental objectives of the police as follows: 1. to prevent and control conduct threatening to life and property (including serious crime); 2. to aid crime victims and protect people in danger of physical harm; 3. to protect constitutional guarantees, such as the right to free speech and assembly; 4. to facilitate the movement of people and vehicles; 5. to assist those who cannot care for themselves, including the intoxicated, the addicted, the mentally ill, the physically disabled, the elderly, and the young; 6. to resolve conflict between individuals, between groups, or between citizens and their government; 7. to identify problems that have the potential for becoming more serious for individuals, the police or the government; and 8. to create and maintain a feeling of security in the community. The ultimate aim of problem-oriented policing is to continually make the police better at accomplishing each of the above objectives–to better prevent crime, to better assist victims, to make communities feel safer, and so forth. Everything the police do, whether using conventional or innovative methods, should be in pursuit of one or more of these fundamental objectives. Problem-oriented policing makes sense to those who share these fundamental beliefs about the police's role and who see policing as a complex and sensitive function, but less so to those who don't. What Are the Various Operational Strategies of Police Work? It is also possible to understand policing in terms of the various methods or strategies used to achieve these objectives. The police employ innumerable specific tactics, but one can better understand these in terms of a few core operational strategies. There are five core operational strategies–preventive patrol, routine incident response, emergency response, criminal investigation, and problem-solving–and one ancillary operational strategy, support services. The first four operational strategies constitute the ways police have conventionally done their work, at least since the 1930s. Problem-solving is a new operational strategy, introduced in Goldstein's problem-oriented policing concept. Each operational strategy of police work has unique and distinct features. Each represents a particular process or method for approaching situations the police encounter. Each is taught to police officers (problem-solving, only recently), and officers are taught when each is appropriate. Each has a distinct general procedural framework that guides officers in doing their work within that operational strategy. Each has a distinct general goal or objective. Each entails a unique way of defining a unit of work, and distinct general performance standards and indicators. Each has its own accountability, reporting and record-keeping systems. Preventive patrol remains the predominant operational strategy of policing in terms of time spent, all research questioning its effectiveness notwithstanding. Most reactive police business is handled using routine incident responses which entail the methodical collection of information about a situation, and classification of the situation. Police use emergency responses far less frequently than routine incident responses, yet they are probably the most critical to the police agency's success, because human life is most directly at stake. Criminal investigations, while constituting a smaller proportion of police work than most people imagine, dominates the public's (and the police's own) perception of police work. Support services (like providing copies of police reports, taking fingerprints for noninvestigative purposes, distributing or teaching generic crime prevention information, and operating youth activity programs) serve primarily to promote and enhance police legitimacy in the eyes of the public by providing nonconfrontational, nonadversarial and noncontroversial services to the public. Problem-solving is the least well-developed operational strategy. Like the other operational strategies, problem-solving has a distinct framework for guiding action. Problem-solving methodology in policing is known familiarly by such acronyms as SARA or CAPRA. It entails problem identification, analysis, response, and evaluation. The general objective of problem-solving is to reduce harm caused by patterns of chronic offensive behavior. The unit of work in problem-solving is known as a "problem," a "problem-solving project" or a "POP project." Performance indicators are significant reductions in harm that are plausibly caused by some specific intended intervention, reductions that hold for some reasonable period of time. Standards of proof have not been sufficiently developed, but the current standards are adapted from the social sciences. Problem-solving also involves some specialized training, and systems for reporting and accounting for problem-solving are being developed. For most of the history of policing, problem-solving has not been recognized as a distinct operational strategy of police work. Even since the advent of problem-oriented policing, most police agencies still have not elevated problem-solving to the level of the other operational strategies, failing to develop the formal systems needed to sustain it. At What Levels is Police Work Done? One can also understand police work in terms of the various levels at which police operate. That is, policing in any given jurisdiction occurs on several scales, ranging from highly localized to intermediate levels to a communitywide level. Each operational strategy can be applied at each operating level. For example, criminal investigation occurs at the localized level during the investigation of a single crime with a single victim. It also occurs at the communitywide level, where the policies and practices for investigating an entire class of crimes, and potentially affecting the entire community, are determined. The same pattern holds for the problem-solving operational strategy, which ranges from highly localized problem-solving (e.g., one drug house, or even one person) to the intermediate level (e.g., a prostitution strip), to the communitywide level (e.g., juvenile homicides throughout a city). Nearly all police work can be understood within this general conceptual framework of objectives, operational strategies and operating levels. The framework helps explain what the police are trying to achieve, how they are trying to achieve it, and on what scale they are operating. The ultimate goal of police reform is to enable the police to better achieve the full range of their objectives, effectively, efficiently and in a manner consistent with basic principles of justice. To do so, the police must be able to perform well in each operational strategy of police work, and at each operating level. This requires that the police develop an organizational capacity to employ the appropriate operational strategy of police work with the appropriate level of resources. It means having a refined understanding of what particular objectives the police are trying to achieve. It means being able to make smooth transitions between and among the various operational strategies of police work, and up and down the operating levels. Making the links between and among the cells of this matrix is challenging and demands sophisticated police work and management–knowing, for example, when a pattern of routine incidents indicates a larger underlying problem that might lead to worse disruption of community life if not addressed, and then using the right level of resources and the right processes to address the situation. A good police department is one in which all operational and administrative systems are aligned and prepared to respond to the community's needs. Where policing often goes wrong is in failures to recognize and balance competing objectives, failures to recognize that a different operational strategy is required for a situation, and failures to use the right level of resources for a particular situation. Precisely because the dynamics of social conflict change so quickly, police organizations are seriously challenged to become highly sensitized to these changes and to respond appropriately. In its broadest sense, problem-oriented policing is a framework designed to help police meet this challenge. How Should the Police Integrate the Need to Address Community Problems With the Desire to Improve Administrative and Procedural Processes? Problem-solving methods can be applied to community problems as well as to internal administrative and procedural problems, but the mere application of a problem-solving process does not automatically render the undertaking a form of problem-oriented policing in Goldstein's terms. For example, a police department supply clerk could use a problem-solving process to work out difficulties ordering uniforms, but this would not make uniform acquisition part of problem-oriented policing. The "problems" to which Goldstein refers in problem-oriented policing are matters directly relating to the public's safety and security, not to the police agency's inner workings. The police can apply problem-solving to the process of investigating crimes or responding to emergencies, but if this results only in making these processes more efficient, without creating some overall improvements to the public's safety and security, it does not constitute problem-oriented policing. Problem-oriented policing entails making tangible improvements to the public's safety and security, and increasing police effectiveness, not merely making police processes less burdensome to the police and/or the public. While it is legitimate and proper to apply problem-solving methods to administrative issues or to promote procedural efficiency, no amount of efficiency-driven problem-solving can substitute for the more important and more challenging application of problem-solving to community crime, disorder and fear. Similarly, making the organizational and administrative changes necessary to support problem-oriented policing is not the same as practicing problem-oriented policing. Only systematic and well- analyzed improvements in policies and practices–those made to increase public safety and security–constitute the essence of problem- oriented policing. All else, however important, is ancillary. It is difficult to overstate the extent to which administrative and political matters can consume the time and attention of the decision- makers most responsible for public safety, including police administrators, other government agency administrators and legislators. Even when there is a deliberate move to adopt a problem orientation to policing or local government, the business of managing organizational change often crowds out the business of addressing actual community problems, at least among top decision-makers. It may turn out that the practice of problem-oriented policing should precede the realignment of the police organization. Without a clear understanding of what the final product is–the successful conclusion of problem-oriented policing initiatives that demonstrably improve public safety–the process of realignment seems uncertain and threatening. Organizational change in police agencies should flow from the experiences of addressing community problems, in somewhat the same way that assembly-line processes in automobile manufacturing plants should flow from the design of the automobile. In short, form should follow function. CHAPTER 3: RELATING PROBLEM-ORIENTED POLICING TO OTHER MOVEMENTS IN POLICE REFORM AND CRIME PREVENTION Various schools of thought on modern police reform, as well as several parallel or complementary movements and theories, have significance for the problem-oriented policing movement. All these movements in the realm where policing, crime prevention and research intersect, have influenced, and been influenced by, problem- oriented policing. Some of these movements complement problem- oriented policing and are variations on its themes, emphasizing one or another element. Other movements compete with problem-oriented policing for acceptance as a general model for improving policing. Team Policing Team policing, a loose collection of ideas about how the police might more effectively serve the public, is, in hindsight, the precursor to contemporary community policing methods. Decentralization of authority, which was central to team policing's underlying theories, proved threatening to many police executives, and did not survive as well as geographic decentralization. Team policing might have been a bit ahead of its time–too much, too soon, to be sustained–but many of its premises were and remain sound, and it had sufficient appeal both to the community and to rank-and-file police officers. Indeed, several core features of team policing, like stability of geographic assignment, unity of command, interaction between police and community, geographic decentralization of police operations, despecialization of police services, greater responsiveness to community concerns, some decentralization of internal decision- making, and at least some shared decision-making with the community, are in place in many of today's police agencies. Community Policing It is beyond the scope of this writing to explore all the distinctions between and similarities of community policing and problem-oriented policing, except to summarize a few distinctions Goldstein has made. Problem-oriented policing primarily emphasizes the substantive societal problems the police are held principally responsible for addressing; community policing primarily emphasizes having the police engage the community in the policing process. How the police and the community engage one another under a problem-oriented approach should depend on the specific problem they are trying to address, rather than being defined in a broad and abstract sense. Carefully analyzing problems before developing new response strategies is given greater weight and importance under problem- oriented policing than under community policing. Community policing emphasizes that the police share more decision-making authority with the community; problem-oriented policing seeks to preserve more ultimate decision-making authority for the police, even while encouraging the police to solicit input from outside the department. Community policing expands the police's role to advance large and ambitious social objectives, like promoting peaceful coexistence, enhancing neighborhood quality of life, promoting racial and ethnic harmony, and strengthening democratic community governance; problem-oriented policing is more cautious, emphasizing that the police are more limited in their capacity to achieve these goals than many people imagine, and guards against unrealistic expectations of the police. From the perspective of those committed to problem-oriented policing as a framework for police reform, the community policing movement has been a mixed blessing. On the positive side, the general idea of community policing has been enormously popular with the general public and, consequently, with elected officials. More specifically, the promise to the public of more access to the police, more police presence in the community, and greater police responsiveness to community concerns largely accounts for community policing's popular appeal. This popularity has translated into substantial financial and authoritative support for a wide range of programs, policies, training, and research, some of which has also benefited the problem-oriented policing movement. Community policing's emphasis on improving the general relationship of the police to the community at large, to minority communities and to organized community groups has undoubtedly helped the police be more effective in their efforts to address particular community problems in a problem-oriented framework. This is no small achievement of the community policing movement. On the negative side, the most politically popular features of community policing have not been the behind-the-scenes analyses of community problems, but the more visible programs that put police officers in all kinds of unconventional settings–on foot and bicycles, in classrooms, in community meetings, at youth recreation functions, etc.–and that have police officers providing unconventional services to the public, like entertaining and educating youth. The attraction to these aspects of community policing has drawn some financial and authoritative support away from the analytical aspects of problem- oriented policing. The popularity of community policing has helped problem-oriented policing gain a degree of attention it might otherwise not have so quickly, but has reduced it to the level of a simplified analytical process for guiding police activities. The challenge for problem-oriented policing advocates will be to maintain support for the further development of the concept's less visible, but more critical, elements. Crime Prevention Through Environmental Design Crime prevention through environmental design (CPTED), while existing as an independent method for analyzing and addressing crime problems closely tied to a geographic setting, has supported the movement toward problem-oriented policing. Conversely, problem- oriented policing has reinforced the concept of CPTED. It has allowed police officers and others who make design decisions to view crime control from an entirely new perspective other than law enforcement. Once exposed to the CPTED principles and methods, many police officers find themselves more open to understanding problem-oriented policing's broader implications. Situational Crime Prevention Situational crime prevention, a relatively new branch of criminology, is perhaps the single most important intellectual development that reinforces and informs the problem-oriented policing movement. The two concepts developed somewhat independently, and then began to influence one another. Situational crime prevention shifts the focus away from deterrence and rehabilitation-based efforts to change offenders' underlying attitudes and behaviors, and toward more situation-specific methods of convincing offenders that committing a particular crime in a particular place at a particular time is not worthwhile. In one respect, problem-oriented policing is the broader concept, not limited to crime problems, but also concerned with the full range of social disorder problems the police must address. In another respect, situational crime prevention is the broader concept, not limited to police actions, but concerned with the actions of any entity capable of preventing crime. Problem-oriented policing has at times been criticized for lacking a criminological theory for its foundation. This criticism presumes that a theory for improving police service must first set forth a theory for preventing crime. This, however, is a far more ambitious, and perhaps unrealistic, goal to which problem-oriented policing never aspired. Problem-oriented policing is best understood as a framework for organizing the police and their activities so that the police are better positioned to learn how to prevent crime and disorder, and to apply that knowledge. It has no explicit preference for one criminological theory over others. It seeks to leave the police open to understanding various criminological theories, and experimenting with practical applications of those theories to determine what works best under what circumstances. If those theories were ultimately proven wrong, it is unlikely that problem-oriented policing advocates would similarly conclude that the problem-oriented approach was also wrong. It would merely add to the knowledge base from which police practitioners could draw to guide their strategic decisions. Crime Analysis and Compstat Crime analysis, as it has conventionally been practiced, is quite different from problem analysis. One of the most prominent and popularized crime analysis methods is patterned after the New York City Police Department's Compstat method. In essence, Compstat is a crime analysis method by which computerized crime statistics are analyzed and presented to operational commanders, who are then responsible for developing operational tactics to respond to emerging crime patterns. The degree to which this basic method is consistent with problem-oriented policing depends entirely on the details of how it is practiced. Problem-oriented policing calls for a broad inquiry into many types of community problems demanding police attention, not just reported Part I offenses. It also calls for analyzing multiple sources of information to develop a fuller understanding of each problem. Where a Compstat-style method results in commanders' selecting from among a limited and conventional set of responses to address problems, such as extra patrol or increased enforcement, it also departs radically from a problem-oriented methodology. Problem- oriented policing calls for a broad and uninhibited search for responses to particular problems, placing special emphasis on responses that minimize the need for the police to use force and large- scale arrest campaigns. A Compstat-style method can foster a hostile atmosphere, more like an inquisition than an inquiry; in this sense, it also differs from problem-oriented policing. Problem-oriented policing, while stressing accountability, also places a high priority on the free exchange of ideas, an exchange that is difficult to achieve in a tension-filled and rigidly hierarchical setting. Finally, problem-oriented policing puts a high premium on communication, consultation and collaboration with entities outside the police department at all stages of the planning process. Ideally, a Compstat-style method would be entirely consistent with problem-oriented policing. As one way to identify specific problems, a computer-generated pattern of crimes would be only the beginning of a more in-depth and broader analysis of the nature of the problems, their underlying conditions and the limits of current responses. For many police agencies, Compstat methods represent a significant advancement in the use of crime data to inform operational decisions. Problem-oriented policing, however, is a considerably more sophisticated and involved approach to handling police business. Crime Mapping and Hot-Spot Policing Crime mapping, now almost a specialized field in itself, can support problem-oriented policing. Crime mapping is enabling police practitioners and researchers to think about crime and disorder and their relationship to other geographic phenomena in ways that were previously unimagined or impractical. Problem-oriented policing specifically calls for, among other things, an analysis of police incidents in terms of location as a potentially useful way to aggregate incidents into clusters. A spatial incident pattern can help stimulate a better understanding of the underlying causes of certain community problems. Crime mapping alone seldom suffices as problem analysis, but it is a potentially useful analytical tool. Hot-spot policing, in essence, requires that the police concentrate their attention and resources on places where and times when there is a significantly high volume of demand for police services. At this basic level of understanding, the idea is compatible with problem-oriented policing. But crime mapping and hot-spot policing are not comprehensive approaches to policing, as is problem-oriented policing. Many problems the police must contend with do not lend themselves to spatial concentrations, and thus will not show up on any hot-spot maps and much of the information the police need to get a complete and accurate picture of community problems is not readily captured in data that are mapped. To the extent that those who use computerized maps to analyze problems become fascinated by the technology itself, there is a risk that the reliability of the data underlying the maps will be taken for granted. In fact, a lot of police data relating to the location of crimes and incidents are ripe for misinterpretation. Broken Windows and Zero Tolerance The "broken windows" theory of crime and disorder asserts that by having the police and community address the many minor community incivilities and signs of neglect, more serious crimes and disorder will be prevented. This idea has spawned as a consequence, intended or not, an idea popularly referred to as "zero tolerance." Zero tolerance prescribes that the police will restrict or eliminate the use of discretion in enforcement, that they will enforce laws as strictly as possible within their means. The way in which the broken windows theory and the zero tolerance strategy have developed in practice, they have little in common with problem-oriented policing. In so many respects, the very notion of zero tolerance is antithetical to problem-oriented policing. The police, of necessity and largely for good cause, exercise enormous discretion in choosing which laws to enforce, when, where, and how. Problem-oriented policing builds on that premise, drawing into enforcement decisions even greater input from the community, prosecutors and other government officials. Optimally, the refined use of the police's arrest powers and the exploration of the many alternatives to arrest will result in less reliance on criminal sanctions to address crime and disorder. Problem-oriented policing does allow that brief periods of concentrated law enforcement might be entirely appropriate to intervene in and disrupt a pattern of crime or disorder, but rejects the wholesale adoption of anything like "zero tolerance law enforcement" as a standing remedy for most community problems. CHAPTER 4: MAJOR CHALLENGES TO ADVANCING PROBLEM-ORIENTED POLICING The development of problem-oriented policing in the past 20 years is encouraging even though quite limited. Perhaps this is to be expected given that the police profession, certainly as compared to most other professions, is relatively young and still in an early developmental stage. It is still developing systems, standards and methods for accumulating and applying research knowledge to practice. Police leaders and the government officials they report to must better appreciate the value that research adds to their decision-making about how to address complex problems of crime, disorder and fear. They must overcome the pressures on them that demand immediate action to complex problems, and resist adopting simplistic responses to them. Problem-oriented policing's full potential will not be achieved in a climate of haste and impatience. Advancing Problem-Oriented Policing Through Training, Research and Practice How Will the Principles and Methods of Problem-Oriented Policing be Taught? Training in the principles and methods of problem-oriented policing for the many different audiences who might benefit from it remains sporadic and of varying quality. PERF continues to offer training in problem-oriented policing, some of which is now offered under the auspices of the Community Policing Consortium. Some of the regional community policing institutes that were provided start-up funding by the COPS Office provide some training in problem- solving, but they had considerable latitude to design their own training curricula and courses. From a problem-oriented policing perspective, it is unfortunate that the institutes' training in problem-oriented policing was not mandated and standardized. Much of the balance of national training programs in problem-oriented policing is provided by small training and consulting firms and a few colleges. The number of training experts is remarkably small. Many police agencies and professional training organizations have not yet fully adopted problem-oriented policing into their organizational missions. Most inhouse training in problem-oriented policing, including that offered as part of preservice academies, is limited to one or two days of instruction. Such limited instruction, offered in discrete blocks of time, can familiarize participants with only the basic concepts; it can hardly be expected to make them proficient in practicing problem- oriented policing. Ideally, training in problem-oriented policing will move beyond simply covering the mechanics of problem-solving to a more advanced treatment of the state of knowledge about common community problems the police confront. Such training would not be limited to teaching enforcement procedures, investigative methods, or laws and policies, but would cover the nature and known causes of the problem, and proven methods of effective prevention, intervention and reduction. There is a need for national training programs to provide police officials, including chief executives, middle managers and analysts, with intensive guidance in applying problem-oriented policing methods to difficult substantive community problems. Moreover, little has been done to convey the concept to audiences other than police practitioners and researchers. Among the target audiences whose particular interests have not been adequately addressed are judges, prosecutors, elected officials, other government agency leaders, and community organization leaders. How Will the Police Accumulate and Transfer Knowledge About Substantive Community Problems? How Substantive Knowledge is Shared in the Police Profession Knowledge in policing is passed on more by listening and talking to other practitioners than by reading published literature. However much this oral tradition strengthens the police's social bonds, it inhibits the transfer of reliable, accurate knowledge. Whereas researchers are expected to be familiar with the relevant literature on a particular subject, there is no similar expectation in policing. Also, there remain far too few opportunities for police officials to spend an extended period of time outside their own organizations in learning environments, a practice deemed essential in many other professions. There are several notable exceptions in which a key police official's sabbatical resulted in problem-oriented policing being introduced upon their return. Writing Down Problem-Oriented Practice Problem-oriented policing has suffered from a lack of quality writing about project work. Without written evidence, the transfer of knowledge about problem-oriented police work is limited to the storytelling of the particular officers involved. Once they lose interest in telling their stories, the knowledge dissipates. The efforts to chronicle good problem-oriented practice at the national level have been beneficial, but modest. These few efforts represent a much smaller investment than Goldstein had in mind, and few of the case studies entail rigorous research methods. Practitioners must be encouraged to continue using problem-oriented approaches to community problems, and to maintain records of their actions. But self-reporting, without some independent verification, lacks reliability. Researchers, whether in-house or external, must be encouraged to do the more formal writing about problem-oriented projects, writing that serves two audiences: researchers and practitioners. If the profession desires and values good written reports of problem-oriented policing, then it must use people with substantial research and writing skills to produce them. Collecting, Synthesizing and Disseminating Research and Practice on Specific Community Problems Those publications that have attempted to capture the state of research and practice with regard to specific types of community problems have not been organized into a centralized reference system. Many of the recent conferences intended in part to bridge the gap between researchers and police practitioners, and to focus on research lessons that would be of interest to practitioners do not present much that directly relates to the police response to community crime and disorder problems. How Can Problem Analysis Be Improved, and a Systematic Body of Research on Substantive Community Problems Be Developed? There still is no coherent research agenda that would lead to a comprehensive and current body of knowledge about specific types of community problems and/or common types of responses to them. A standard literature search on any particular problem would lead the researcher to a host of different professional journals, books and technical reports, many of which would provide only a theoretical perspective, rather than a practical perspective from which one might adopt proven interventions or fashion new ones. The amount of potentially useful information is no doubt much greater than most police officials realize, but because it has not been systematically compiled and annotated for use by practitioners, it remains largely unavailable to the police. The police are not engaging in much policy-level problem analysis themselves. Police research and planning units should shift their focus to studying their agency's response to large-scale community problems. They should expand beyond conventional methods such as identifying spatial patterns of crime through mapping. Police agencies without such in-house expertise or resources should collaborate with outside researchers. Police researchers must have the skills necessary to conduct advanced problem analysis or, at a minimum, be able to make intelligent use of what literature exists on substantive problems. The sort of police practitioner-researcher collaboration envisioned for problem-oriented policing has not occurred more often as a result of difficulties on both sides. For their part, some police officials are impatient with extensive research, preferring to work on smaller-scale problems with rudimentary research than to wait for more sophisticated research to shed new light on larger problems. Researchers, for their part, sometimes find it difficult to make the transition from pure social science research methods to the action research called for in problem-oriented policing. Criminology and related criminal justice sciences have been slow or reluctant to substantively engage in problem-oriented policing. There are few academic researchers with practical experience in problem-oriented policing, so some police agencies would be hard-pressed to find the right kind of research assistance, even if they sought it. For their part, the police have viewed criminology as abstract and, accordingly, have not sought to incorporate the lessons of criminology into their practices. The experiences of the past two decades suggest that the best avenue for systematically advancing knowledge is one that requires contributions from both practitioners and researchers. Whether improvements in the research community will generate greater interest among the police in using research to address community problems, or whether a greater police demand for such research will spur researchers to action is not clear. One thing is clear: The quality and quantity of the underlying research and the writing about problem- oriented projects need substantial improvement, even while the current, more modest efforts should be recognized and encouraged. Defining Roles for Others in Practicing Problem-Oriented Policing Are New Alliances Between the Police and the Community Healthy? Problem-oriented policing stresses police collaboration with the community to address problems. Under certain conditions, however, these new collaborations between police and community present significant challenges in a constitutional democracy. At times, the "majority rules" philosophy of the community and the conservative traits of the police combine to support police practices that the courts find threatening to the constitutional order. Goldstein imagined that the processes used in problem-oriented policing, in which the police carefully develop responses based on thorough research, and subject those responses to review and input from many perspectives, would reduce the possibility that the courts would challenge and strike down police actions. The mere application of a problem-solving process to community problems, however, does not guarantee that all the interests of a constitutional democracy will be protected. Are New Alliances Between the Police and Other Government Agencies Healthy? The police and other agencies can often accomplish more working together than they can working independently, but new alliances between the police and other government agencies hold potential for overreaching. Each agency, police included, must maintain some independence to protect against overzealousness and abuses of authority. Partnerships should not be abandoned because of the possibility of overreaching, or even because of occasional incidents of overreaching, but administrators and oversight bodies should remain aware of the risks. What Should Be the Role of Prosecutors? Historically, prosecutors have related to the police almost exclusively in terms of the criminal investigation function. Prosecutors exert a powerful influence on police practices, despite the reality that only a small percentage of police work culminates in criminal prosecution. Prosecutors jealously guard against any diversion of police resources away from criminal investigation. There have been some efforts to reconsider prosecutors' role in the larger enterprise of promoting public safety. A few local prosecutors' offices around the United States have experimented in what has come to be known as community prosecution. Typically, in community prosecution, prosecutors are assigned to geographic areas and are responsible for learning more about their area's public safety concerns, and prosecuting all or most of the crimes that arise out of that area. If community prosecution, however, is limited to prosecuting criminal cases along geographic lines, it is not a significant departure from conventional practice, and does not necessarily reinforce problem- oriented policing. If prosecutors actually reconsider their function as one of solving community crime, disorder and fear problems, rather than just prosecuting individual cases, they reinforce problem-oriented policing. Without prosecutors, a valuable perspective on crime problems is missing from many police-led initiatives. Prosecutors are better-aware of how cases are processed through the court system and, accordingly, are more aware of the relative effectiveness of existing means for disposing of cases. Prosecutors also are more aware of the range of legal responses that might be used to address a particular problem, as well as some of the risks of alternative approaches. Prosecutors have access to court data and to judges, and research skills the police often lack. When prosecutors are open-minded and take a broad perspective on their role, they can greatly facilitate problem-oriented policing. The absence of prosecutors from the problem-solving process conveys a powerful signal to the police that problem-solving is not valued as highly as criminal investigation. This can discourage the police from investing more fully in problem-solving. The emerging movement toward community prosecution is a positive development toward advancing problem-oriented policing, but it is far from complete. This new orientation toward prosecution remains rare among prosecutors' offices, and it will require every bit as much effort to reorient prosecutors to their work as it is taking to reorient police officers to theirs. It will require some changes in how law schools train students, especially those aspiring to become government lawyers. Currently, conventional legal training offers little that would prepare a prosecutor for problem-oriented prosecution. What Should Be the Role of Local Government Leaders? If prosecutors have had limited involvement in problem-oriented policing, local government leaders have probably had even less so. It is not enough that local government leaders generally endorse community policing. They must invest time and energy in understanding problem-oriented policing's full implications. For the most part, local government leaders still attribute primary responsibility for public safety to the police, fire and ambulance services, despite growing evidence that crime, disorder and fear are greatly influenced by land-use planning, economic development, business regulation, code enforcement, architecture, public housing management, and traffic engineering. The responsibility for public safety should be more evenly distributed among local government agencies. Were this the case, local government leaders would play a primary role in coordinating and guiding problem-oriented initiatives to reduce crime, disorder and fear. They must invest in research and analysis, and information technology–investments that, while not guaranteed to pay off immediately, are highly likely to pay off in the long term. Without leadership to create new expectations that departments analyze and collaborate on public safety problems, it is not likely to happen. Should the Police Be Held More Accountable for Reducing Crime, Disorder, and Fear? After two decades of experimentation with problem-oriented policing, we are not really much closer to answering the question of whether the police should be held more accountable for reducing crime, disorder and fear, and if so, what approach would best achieve this. Goldstein has long argued that problem-oriented policing is an approach that recognizes the limits of police authority and the limits of police practices alone to bring about significant changes in public safety. When the police and the community accept that the police are not omnipotent, the police can solicit and receive the active support of the community and other government agencies to more effectively address the problems of crime, disorder and fear. Problem-oriented policing has demonstrated an internal logic, has been successfully applied at the project level, and remains a promising approach for the foreseeable future. There is growing reason to believe that collaborations of police, governments, businesses, and communities, committed to carefully analyzing community problems and developing tailored responses, can bring about significant changes to public safety levels. Beyond that, claims about the police's capacity to single-handedly reduce crime, disorder and fear at the community or higher level are simply not warranted. The greatest promise of problem-oriented policing may be that it is the approach most likely to maintain the delicate balance between freedom and order, and minimize the likelihood that police actions will undermine their legitimacy in society. This is so largely because the problem-oriented approach rejects the very excessive reliance on the enforcement of criminal law, and the use of force that accompanies it, that so often leads to abuse and consequent erosion of public trust in the police. Achieving that much, while incrementally and systematically improving our understanding about how police and communities can effectively reduce crime, disorder and fear, is a considerable improvement from past approaches to policing. Conclusion: How Will We Know if Problem-Oriented Policing Works? The ultimate test of problem-oriented policing is whether it proves successful in enhancing police service. Asking whether problem- oriented policing works, and asking whether the problem-oriented policing movement has been successful, are separate matters. The first question is a search for proof that the problem-solving methodology reduces crime and disorder, makes communities safer, and does so better than any other approach to policing. The second question is a search for proof that problem-oriented policing has become the standard approach to policing. Whether problem-oriented policing works depends, of course, on what one believes to be the objectives of the police. Successful policing, in the broadest sense, is policing that achieves its multiple objectives. Because these objectives sometimes compete with one another, there can be no such thing as maximally effective policing, only optimally effective policing, whereby the police have balanced their objectives. To paraphrase Morgan Stanley Dean Witter's marketing slogan, one can only determine problem-oriented policing's success "one problem at a time", at least at the microlevel. That is, one should assess police effectiveness with respect to each discrete social problem the police are at least partially responsible for addressing. Because problems of crime, disorder and fear arise and abate through a complex interaction of social norms, laws and technology, there really can be no end point to policing. As one class of problems abates, new classes of problems arise. Indeed, police work is always described in the present participle–"policing"–and never in the past tense. A community is never considered to have been policed. Thus, while it is appropriate to judge problem-oriented policing by the degree to which it is effective in addressing society's current problems, one should also judge it by the degree to which it prepares the police to identify and respond to future problems. The problem-oriented policing movement can be said to have succeeded once police agencies have integrated the problem-solving operational strategy of police work into their operations at least as completely as they have the other operational strategies of preventive patrol, routine incident response, emergency response, and criminal investigation. It will have succeeded too once the imbalance between policing's "means" and "ends" has been altered to better reflect a direct concern on the part of police administrators and researchers with the substantive aspects of police business. As is probably true in all fields, the development of an important idea, or of several important ideas simultaneously, is not neat and clean. There is no central policymaking entity, at least not in American policing. Scholars and practitioners alike shift through time in their understanding and support of the various ideas. The ideas themselves are shaped by factors other than pure theory or tested practice: by political and popular interest, available funding and the desire to achieve distinction. While the uneven and sometimes contradictory way these various movements push and pull the police profession frustrates those who are committed to one idea or another, in the long run, this is for the best. It is best for society as a whole, and best for the problem-oriented policing movement. The diversity of ideas and the highly decentralized way they are implemented have ultimately led to refinement of the best of them. Were it even possible for the development of problem-oriented policing to be centralized and made more consistent, it would likely weaken the idea. A single wrong turn in centralized policymaking results in many wrong turns in police practice. There are risks to promoting homogeneity in the implementation of problem-oriented policing, whether through the requirements of federal funding programs or through other means. An idea such as problem-oriented policing, which has yet to be fully developed, needs diversity to grow. And so it is that problem-oriented policing competes in the messy marketplace of ideas about how to improve policing. Problem-oriented policing must pass the rigorous tests of academic scrutiny and criticism to prevail as a path for improving policing. To be tested properly, it must be implemented with at least basic fidelity to the fundamental principles laid out by Herman Goldstein. Goldstein never intended that problem-oriented policing, at least as he articulated it, be understood as a finished or definitive product. Indeed, as the police scholar Jean-Paul Brodeur wrote: "[I]t would seem as difficult as it is futile to measure with precision the extent to which the new strategy has been implemented. Such a measurement implies freezing a paradigm that is characterized by its openendedness." Problem-oriented policing has come a long way in 20 years, from the chalkboards and classrooms of the University of Wisconsin, to the squad rooms, community meeting halls and conference rooms where modern policing is played out. It has achieved a degree of professional interest, and some measure of public and political interest, that must be heartening to Herman Goldstein and those who believe in his idea. The development of problem-oriented policing, however, is far from complete. Ironically, the popularity of the idea puts it at risk of burning out, and that would be unfortunate. It is precisely because problem-oriented policing is so deeply rooted in what Goldstein calls the basic arrangements for policing in a free and open society–the most fundamental challenges for establishing domestic tranquility and order–that police, community and government officials can ill afford to rest comfortably on the progress made to date. Introduction What is the Purpose of This Report? My aim in writing this report is to describe how Herman Goldstein's problem-oriented policing framework, as I understand it, has been developed and, at times, distorted in the many efforts to make it a standard way of policing. I will not attempt to argue the merits of Goldstein's problem-oriented policing concept in this writing other than to say that I believe in its merits.1 I try not to declare certain practices "right" or "wrong;" I don't have the wisdom for that. Moreover, it is nearly impossible to know all that is occurring in the name of problem-oriented policing. The field of policing is much too large, diverse and decentralized, so my frame of reference will necessarily be limited. I have sought to discuss those trends and practices that are generally accessible and therefore observable. My approach to research for this report was to combine personal experience, a review of relevant literature (see References and Appendix B) and problem-oriented policing project reports (see Appendix A), site visits to selected police departments, attendance at conferences, extensive discussions with Herman Goldstein, and interviews of others well-versed in problem-oriented policing (see Appendix C). Problem-oriented policing is still in its relative infancy. It has not withstood the long tests of time or sufficient critical evaluation. The concept itself could conceivably be proven misdirected or fail to be properly implemented. I intend to take stock of the problem-oriented policing movement, clarify its original principles, encourage promising developments, and, perhaps, correct some distortions. The specific objectives of this report are 1. to clarify the core elements of Goldstein's ideal model of problem-oriented policing; 2. to describe distortions to various core elements in the practice of problem-oriented policing; 3. to place the concepts of "problem-oriented policing" and "problem-solving" within the context of total police service; 4. to describe the strongest aspects and greatest deficiencies of the move toward problem-oriented policing; 5. to assess the overall progress made by police agencies, governments and research institutions in advancing problem- oriented policing; and 6. to propose directions for the future development of problem- oriented policing. 1Goldstein has published extensively on the subject. Readers interested in his original materials are encouraged to read Goldstein (1977, 1979, 1987a, 1987b, 1990a, 1991, 1993c, 1996a). I have drawn on his published works, unpublished speeches, notes, training materials, and personal comments to enhance my understanding of the problem-oriented policing concept. A comprehensive listing of Goldstein's writings on problem-oriented policing is provided in the references section. A Brief History of the Spread of Problem-Oriented Policing Tracing the development of problem-oriented policing, as with any ideological movement, is difficult. It is hard to say who thought of what, and when, and precisely when a particular idea was translated into action. One often hears in police training sessions on community or problem-oriented policing, "Oh, we've been doing community policing [or problem-oriented policing] for years; we just didn't call it that." Whatever grains of truth there are in such assertions,2 most police agencies can trace the formal introduction of a concept like problem-oriented policing to a particular time in their history. 2There is little historical evidence that the police have done the kind of analysis of community problems and deliberate policy formulation envisioned in problem-oriented policing, despite some claims to the contrary. Wrote one reviewer of Goldstein's Problem- Oriented Policing, "The use by police departments of systematic studies to analyze and respond to recurring crime problems is as old as the billy club and the paddy wagon…" (Diulio 1990). The first formal experimentation with Goldstein's model of problem- oriented policing occurred in Madison, Wisc., in 1981 when Goldstein and his associates worked with the Madison Police Department exploring the community's response to drinking drivers and repeat sex offenders (Goldstein 1980, 1990a; Goldstein and Susmilch 1981, 1982a, 1982b, 1982c). That isn't to say that the Madison Police Department was the first to systematically study a community problem, but it was the first to formally apply Goldstein's model. Around 1982, with support from Gary Hayes, the then-executive director of the Police Executive Research Forum (PERF), the London Metropolitan Police undertook their own experimentation with the concept (Hoare et al. 1984), as did the Surrey, England, Police Force (Leigh, Read and Tilley 1996). Again with Hayes' encouragement, the Baltimore County Police Department formally introduced Goldstein's problem-oriented policing model into its COPE unit's operations in 1983 (Taft 1986). The Newport News, Va., Police Department followed suit in 1984 (Eck and Spelman 1987). The efforts in both Baltimore County and Newport News benefited from Goldstein’s personal involvement and guidance. Both departments’ initiatives had some outside funding, and thus resulted in excellent, detailed written reports that more widely communicated the problem-oriented policing concept to police practitioners and researchers.3 The published report on the Newport News project provided perhaps the most comprehensive treatment of problem-oriented policing at that time, and remains an important and influential document. 3The Baltimore County project received funding from the Florence V. Burden Foundation to prepare a descriptive report and prescriptive training materials. The U.S. Department of Justice's National Institute of Justice funded the Newport News project. A number of other police agencies began to incorporate at least some of the problem-oriented policing methodology into broader community policing efforts during the 1980s. Among the more prominent efforts cited by Goldstein (1990a) were those in New York City (the Community Patrol Officer Program); Edmonton, Alberta (downtown foot beats); Flint, Mich.; Los Angeles (the Community Mobilization Project in the Wilshire district); Houston (referred to as neighborhood-oriented policing); Oxnard, Calif.; Savannah, Ga.; Evanston, Ill.; Tulsa, Okla.; Beloit, Wisc.; and Halton, Ontario. Through the 1980s, PERF helped a number of agencies to replicate various elements of the problem-oriented policing model developed in Baltimore County and Newport News. Among those agencies were the Tampa, St. Petersburg and Clearwater police departments in Florida. PERF's project to apply problem-oriented methods to drug problems in San Diego, Tampa, Atlanta, Philadelphia, and Tulsa, funded by the Bureau of Justice Assistance from 1987 to 1990, proved significant in furthering the problem-oriented policing movement. While some of these sites achieved only modest substantive and organizational success, the San Diego Police Department used this research project as a catalyst to make a substantial investment in problem-oriented policing. The first few national conferences on problem-oriented policing were conceived and partially funded out of this project. The national conferences (the first of which drew about 200 participants in 1990, and currently are drawing about 1,250 participants) have become a major means by which the concept of problem-oriented policing has spread, especially among police practitioners. The publication in 1990 of Goldstein's Problem-Oriented Policing spurred some interest among police agencies, but probably had a greater impact on the police research audience.4 Police agencies, which acquire knowledge and skills differently than research institutions, were offered some modest, but important, training opportunities in problem-oriented policing starting in 1989. While on staff at PERF, I designed a two-day training seminar in the basic principles and methods of problem-oriented policing, and offered it nationwide.5 PERF continued and expanded the training after I left.6 A number of police agencies and police officials today can trace their engagement in problem-oriented policing to those training sessions conducted in the late 1980s through the early 1990s. Even today, some of the training materials developed for those early programs surface in modified form in police training. Through about 1994, PERF provided a significant amount of the limited problem-oriented policing training available, at least in the United States. Since then, much of the training in community policing and problem-solving has been offered under the auspices of the Community Policing Consortium7 and more recently, the Regional Community Policing Institutes.8 Many of the agencies that received PERF's training have not significantly incorporated problem-oriented policing into their operations, but a few have, and have made significant contributions to the concept's development, as a result. 4A number of reviews of Problem-Oriented Policing have been published. Among them are Bayley (1991), Sherman (1991), Das (1992), Vaughn (1992), Mastrofski and Uchida (1993), and Diulio (1990). See, also, Brodeur (1998b). 5Among the sites that received this early training were Alexandria, Va.; Aurora, Ill.; Boca Raton, Fla.; Gaston County, N.C.; Hillsborough County, Fla.; Hurst, Texas; Macon, Ga.; Milwaukee; Monroe County, Fla.; New York City Transit Police; Reno, Nev.; St. Paul, Minn.; Suffolk County, N.Y.; and Wilson, N.C. 6PERF's training was subsequently coordinated in succession by Rana Sampson, Susie Mowry, Ron Glensor and John Lusardi. Lusardi and Sampson subsequently founded private firms that provide training in problem-oriented policing. 7The Community Policing Consortium, created in 1993 with funding from the Office of Community Oriented Policing Services, is a joint enterprise of PERF, the International Association of Chiefs of Police, the Police Foundation, the National Organization of Black Law Enforcement Executives, and the National Sheriff's Association. 8Approximately 30 Regional Community Policing Institutes were established in 1997 through funding by the Office of Community Oriented Policing Services to provide training in community policing and problem-solving. Police agencies often resist long-term change, but remarkably, respond to many short-term programmatic innovations. The short history of problem-oriented policing bears this out. At most of the agencies mentioned above, one strong person encouraged the idea of, and certainly advocated, experimenting with problem-oriented policing. Many of the problem-oriented initiatives generally associated with a particular agency prove, upon closer inspection, to be attributable to one or a few individuals. Usually these are high-ranking personnel, although sometimes the lone champion9 of problem-oriented policing is at the line or supervisory level. When the high-ranking champions leave the agencies, as they inevitably do, the push to engage in problem-oriented policing typically wanes, as well. Many of the police agencies listed in Table 1 experienced the departure of at least one principal champion of problem-oriented policing. These people are generally recognized as having been the driving force behind their agencies' efforts to adopt a problem-oriented policing approach, however long- or short-lived. 9By champion, I refer to someone who makes deliberate study of problem-oriented policing by reading the literature, attending conferences, doing or supervising problem-oriented work, maintaining contacts with others outside the organization about the concept, importing new problem-oriented ideas into the organization, and keeping the concept alive in the organization's consciousness. Table 1 Police Agencies, and the Problem-Oriented Policing Champions Who Work or Have Worked There -Police Agency -Problem-Oriented Policing Champion -Baltimore County Police Department -Chief Neal Behan (retired), Maj. Philip Huber (left) -Charlotte-Mecklenburg, N.C., Police Department -Chief Dennis Nowicki (retired); Chief Darrel Stephens -Chicago Police Department -Deputy Chief Charles Ramsey (left), Director Barbara McDonald -Edmonton Police Service -Superintendent Chris Braiden (retired) -Fort Pierce, Fla., Police Department -Chief Gil Kerlikowske (left), Director of Administration Michael Scott (left) -Joliet, Ill., Police Department -Chief Dennis Nowicki (left), Capt. William Fitzgerald -Lauderhill, Fla., Police Department -Chief Michael Scott (left); Deputy Chief Michele Reilly (left) -London Metropolitan Police -Police Commissioner Sir Kenneth Newman (retired) -Madison Police Department -Chief David Couper (retired), Lt. Randall Gaber, Sgt. Joe Balles -Merseyside, England, Police Constabulary -Chief Inspector Brian Gresty (retired) -Newport News Police Department -Chief Darrel Stephens (left) -Peel, Canada, Regional Police -Chief Robert Lunney (retired) -Philadelphia Police Department -Inspector Ed McLaughlin (left) -Reno Police Department -Chief Robert Bradshaw (left), Deputy Chief Ron Glensor -San Diego Police Department -Chief Jerry Sanders (retired), Nancy McPherson (left) -Seattle Police Department -Chief Norm Stamper (retired), Nancy McPherson (left) -Police Agency -Problem-Oriented Policing Champion -St. Louis Metropolitan Police Department -Police Board President David Robbins (left), Chief Clarence Harmon (retired), Special Assistant to the Chief of Police Michael Scott (left), Sgt. Robert Heimberger -St. Petersburg Police Department -Chief Darrel Stephens (left to become city administrator, now chief in Charlotte-Mecklenburg Police Department) -Thames Valley, England, Police Constabulary -Chief Constable Charles Pollard, Chief Inspector Caroline Nicholl (left) -Tulsa Police Department -Chief Drew Diamond (left), Maj. Carolyn Kusler (left) In 1994, the U.S. Department of Justice created the Office of Community-Oriented Policing Services, known more informally as the COPS Office. The Justice Department created this agency primarily to oversee the federal funding of 100,000 new U.S. police officer positions. The COPS Office linked the position funding to the broad concept of community policing, of which problem-solving was a key element.10 Thus, Goldstein's problem-oriented policing model was linked to an enormous federal funding scheme. The COPS Office was required by law to advance community policing generally, but outside of a few of its competitive funding programs, most of its large funding programs did not require that recipient police agencies engage more specifically in problem-oriented methods. The largest COPS Office program to directly fund problem-solving projects, as opposed to just funding community police officer positions, is the Problem- Solving Partnerships (PSP) Program. Four hundred seventy police agencies received funding to help them identify, analyze, address, and evaluate a specific substantive community crime or disorder problem.11 Similarly, the School-Based Partnership Grant Program funds problem-solving methods in schools.12 While the link between problem- solving and community policing in this large federal funding program has yielded many benefits, the linkage has also blurred the distinction between problem-oriented policing and community policing. 10A national process evaluation of the COPS Office reported that while community policing takes many forms among the grant recipient agencies, some form of problem-solving is occurring in most agencies (Roth and Ryan 2000). 11The Police Executive Research Forum (2000) evaluated the PSP Program, and Michigan State University Professor Tim Bynum is evaluating problem analyses conducted by 16 selected grantees. Twice during my research, I reviewed reports PSP grantees submitted to the COPS Office. I found that most of the reports provided too little information for a person unfamiliar with the particular problem to form a good understanding of it. (I do not know to what degree the quality of the reports reflected the quality of the actual work done on the problem.) By my own estimate, one generally confirmed by the other COPS Office staff members reviewing the reports, only 15 to 20 percent of the PSP reports were of good quality. 12The School-Based Partnership Program is also funded for a program evaluation to assess the impact problem-solving methods and responses had on identified problems. As of 2000, many police agencies in the United States and Canada,13 and a growing number in the United Kingdom,14 Australia, New Zealand, the Netherlands,15 South Africa, and Scandinavia, report that they are engaged in problem-oriented policing in some fashion. A few agencies have expressly made problem-oriented policing the focal point of their long-range strategic plans.16 A review of police agency websites, increasingly becoming a standard method of communication, finds dozens of agencies specifically citing the adoption of problem- oriented policing methods in their operations. There is no easy way to quantify the number of police agencies engaged in problem-oriented policing, much less to gauge the precise nature and quality of those efforts. However, it is safe to say that far more agencies claim to be engaged in problem-oriented policing today than at any other time. 13See Leighton (1994) for a description of the development of community and problem- oriented policing in Canada. See, also, Saville and Rossmo (1995) for a description of early problem-oriented policing efforts in British Columbia. 14See Leigh, Read and Tilley (1996, 1998) and Read and Tilley (2000) for descriptions of the spread of problem-oriented policing in the United Kingdom. They conclude that "problemoriented policing is being widely considered in British police services. It appears to be an idea whose time has come" (1998:54). See, also, Bennett (1994), and Tilley (1999). Herman Goldstein has helped advance the problem- oriented policing concept in the United Kingdom through various consultations and speeches (Goldstein 1995a, 1996a). 15See Willemse (1994) for a description of the development of crime prevention principles in the Netherlands. 16The Colorado Springs, Colo., Police Department described its commitment to problem-oriented policing in a document titled Total Problem-Oriented Policing. The Lauderhill, Fla., Police Department, where the author served as the first police chief, made problem- oriented policing the organizing concept during the creation of the new agency. Problem-oriented policing continues to advance across the police field, even while the adoption of problem-oriented policing into particular police agencies seldom happens in a linear fashion. Interest in the concept and commitment to its implementation rises and falls in response to many internal and external factors. Changes in leadership, competing priorities or simply inertia can alter the course of implementation. Accordingly, one might reach different conclusions about the vitality of problem-oriented policing depending on whether one was looking only at selected police agencies or at the police field as a whole. At the risk of overlooking some exemplary efforts (and perhaps giving more credit than is due elsewhere), below is a list of the police agencies that, at one time or another, have been prominently associated with problem-oriented policing. Their prominence may have resulted from a particular person in the agency, involvement in a research effort, or publication of the agency's efforts in the professional literature. Inclusion on this list is not a testament to the depth or quality of the agency's commitment to problem-oriented policing. Table 2 Police Agencies Prominently Associated With Problem-Oriented Policing17 1. Edmonton Police Service18 Alberta 2. Phoenix Police Department Arizona 3. Tempe Police Department Arizona 4. Fresno Police Department19 California 5. Hayward Police Department California 6. Huntington Beach Police Department California 7. Los Angeles Police Department20 California 8. Oxnard Police Department California 9. Sacramento Police Department California 10. San Diego Police Department21 California 11. Santa Ana Police Department22 California 12. Santa Barbara Police Department23 California 13. Arvada Police Department Colorado 14. Aurora Police Department24 Colorado 15. Colorado Springs Police Department25 Colorado 16. Longmont Police Department Colorado 17. Cleveland Police26 England 18. Lancashire Police27 England 17Sources for this list include the files and personal knowledge of the author and Herman Goldstein; Community Policing and Problem- Solving: Strategies and Practices, by Kenneth J. Peak and Ronald W. Glensor (1996); and various other publications, some of which are footnoted. 18See Hawkins (1998), Koller (1990), Hornick et al. (1990), and Weisel and Eck (1994). 19See West (1995). 20See Goldstein (1990a) and Greene (1998a). 21See Police Executive Research Forum (1989), Bureau of Justice Assistance (1993a), and Capowich and Roehl (1994). 22See Jesilow et al. (1998). 23See Barrett (1996), and Weisel and Eck (1994). 24See Williams and Sloan (1990). 25See Kramer and McElderry (n.d.). 26See Leigh, Read and Tilley (1998). 27See Kirby (1997). 19. Leicestershire Police Force28 England 20. London Metropolitan Police29 England 21. Merseyside Police Force30 England 22. Surrey Police Force England 23. Thames Valley Police Force England 24. Clearwater Police Department Florida 25. Delray Beach Police Department Florida 26. Fort Pierce Police Department Florida 27. Hillsborough County Sheriff's Office Florida 28. Lauderhill Police Department Florida 29. St. Petersburg Police Department Florida 30. Tampa Police Department31 Florida 31. Atlanta Police Department32 Georgia 32. Savannah Police Department33 Georgia 33. Aurora Police Department Illinois 34. Chicago Police Department34 Illinois 35. Evanston Police Department Illinois 36. Joliet Police Department Illinois 37. Wichita Police Department Kansas 38. Baltimore County Police Department35 Maryland 39. Ann Arbor Police Department Michigan 40. Flint Police Department36 Michigan 41. Kansas City Police Department Missouri 42. St. Louis Metropolitan Police Department Missouri 43. Charlotte-Mecklenburg Police Department North Carolina 44. Lincoln Police Department Nebraska 45. Reno Police Department Nevada 46. New York City Police Department37 New York 47. Suffolk County Police Department New York 48. Tulsa Police Department38 Oklahoma 49. Halton Regional Police Ontario 50. Peel Regional Police Ontario 51. Portland Police Department Oregon 52. Abington Township Police Department Pennsylvania 53. Philadelphia Police Department39 Pennsylvania 54. Stockholm Police Sweden 55. Austin Police Department Texas 56. Houston Police Department40 Texas 57. Newport News Police Department41 Virginia 58. Seattle Police Department Washington 59. Spokane Police Department Washington 60. Beloit Police Department Wisconsin 61. La Crosse Police Department Wisconsin 62. Madison Police Department42 Wisconsin 28See Leigh, Read and Tilley (1996, 1998). 29See Hoare, Stewart and Purcell (1984), and Goldstein (1990a). 30See Gresty et al. (1997), and Berry (1999). 31See Police Executive Research Forum (1989) and Bureau of Justice Assistance (1993a). 32See Police Executive Research Forum (1989) and Bureau of Justice Assistance (1993a). 33See Young (1998), and Weisel and Eck (1994). 34See Skogan (1998), Skogan, et al. (1999) and Hartnett and Skogan (1999). 35See Taft (1986), Higdon and Huber (1987), and Goldstein (1990a). 36See Goldstein (1990a). 37See Goldstein (1990a), and McElroy, Cosgrove and Sadd (1993). 38See Police Executive Research Forum (1989) and Bureau of Justice Assistance (1993a). 39See Police Executive Research Forum (1989), Bureau of Justice Assistance (1993a), Weisel and Eck (1994) and Berry (1996). 40See Goldstein (1990a). 41See Eck and Spelman (1987), Weisel and Eck (1994), Babcock (1996) and Goldstein (1990a). 42See Goldstein (1990a), Couper and Lobitz (1991) and Wycoff and Skogan (1993). A number of the agencies listed have, for all appearances, abandoned their efforts to implement problem-oriented policing and are no longer strongly associated with the concept. In a few cases, this abandonment reflected a conscious policy decision. More commonly, the momentum for problem-oriented policing subsided due to lost interest, competing priorities or inertia. A number of other agencies not listed here have only recently sought to implement problem- oriented policing, and their efforts have not yet been widely recognized. As so many police officers, researchers and observers have noticed, when one goes to visit agencies renowned for their problem- oriented policing efforts, these visits often prove disappointing. Frequently, what once was there is no longer, or what is there proves less substantial than what one expected from distant reports. However slow, modest and uneven the movement in problem-oriented policing has been, it is now a central part of at least the language of modern police management. If language can influence culture, as surely it can, problem-oriented policing is making inroads to the professional culture of police management and operations. The term problem-oriented policing itself now has an acronym–POP. The "International Conference on Problem-Oriented Policing" is more commonly referred to as the "POP Conference." Police personnel commonly refer to substantive problem-solving initiatives as "POP projects." And to capture those "POP projects," PERF has created a computerized compilation of them named "POPNet" and a private consulting firm is marketing a computer software program called "POP Track."43 The movement has its own T-shirts and coffee mugs. The concept has even appeared in police mystery fiction,44 as well as in British television police dramas.45 So, too, with the term problem- solving. Rare in police management circles 20 years ago, it is now standard in the police lexicon. Some have called problem-oriented policing problem-solving policing (Moore 1998). 43POP Track is a software program designed to allow police agencies to track problem-oriented policing projects and resources. It was developed and is marketed by Law Enforcement Assistance Network, a private police consulting firm headed by several police officials with experience in implementing problem-oriented policing and community policing. 44See Par Four, by Elizabeth Gunn (1998). 45An episode of the BBC program "Cops" apparently had a fictional police chief inspector invoking problem-oriented policing in an address to a community group. A critic, writing for the Manchester Guardian, apparently did not recognize problem-oriented policing, declaring it an imaginary term. Along with this rise in the popularity of terms associated with problem-oriented policing has come a certain amount of distortion of its original meaning. A lot of what is presented as a "POP project" incorporates no careful study of an underlying problem, but reflects merely a programmatic effort that may or may not affect a particular community crime or disorder problem. Problem-oriented policing is sometimes described as an operational strategy itself, as in "we applied problem-oriented policing tactics to the problem." In fact, there are no distinctly problem-oriented tactics.46 Problem-solving, specifically referring to an analytical process in the context of problem-oriented policing, is also used frequently to refer to methods for addressing administrative or personnel matters, or, more generally, to any mental process involving some degree of reflection before action. Accordingly, one should not assume that any use of the terms problems, problem-solving or problem-oriented in the policing context is directly connected to Goldstein's original conception of problem-oriented policing. 46See, for example, Klein (1998), and O'Connor, Shelley and Grant (1998), who discuss problem- oriented policing's impact on, respectively, gangs and domestic violence. Some elements of Goldstein's ideal model of problem-oriented policing have been grafted onto other conceptualizations of police work and police reform. First-generation programs, projects and experiments in community policing, dating back to the late 1960s, did not explicitly incorporate a structured analytical methodology into the new forms of police work. Only in around the mid-1980s did community policing advocates begin to incorporate some elements of problem-oriented policing into that framework. Problem-oriented policing and community policing were both explored during the Executive Sessions on Community Policing held at Harvard's Kennedy School of Government that began in 1985. During this time, and partly as a result of the discussions in these sessions, problem- oriented policing began to be incorporated into the concept of community policing, an integration that has never been complete, or completely warranted.47 Failing to acknowledge all aspects of the problem-oriented model beyond beat-level problem-solving, some community policing advocates repackaged problem-oriented policing as a tactic, tool or method one might use within a community policing philosophy. This repackaging failed to recognize some critical differences between community and problem-oriented policing, namely, that they have different primary goals and, consequently, some different methods. Moreover, it reduced Goldstein's involved, intensive and rigorous communitywide problem analysis to more informal street-, beat- or neighborhood-level problem-solving. Other advocates of police reform introduced yet additional labels, like "neighborhood policing," that are not theoretically distinguishable from community policing.48 47Goldstein himself at one time linked problem-oriented policing to community policing by characterizing the various reform concepts as falling under the larger umbrella of community policing (1987a; see, also, Brodeur 1998b), a link he has since come to reconsider and regret (1995b). 48For a more in-depth discussion of the various perspectives on community policing, see Police Executive Research Forum (1996:2-8). All these efforts toward concept clarification produced a variety of labels and a lot of confusion among police practitioners, much of which lingers today. To some degree, all the debate and efforts to synthesize and harmonize the different concepts reflected an intellectual battle for the high ground in police reform, with advocates' claiming each concept to be the overarching framework under which all others would be subsumed. Goldstein, at the University of Wisconsin, and PERF, in the 1980s, represented one school of thought–the problem-oriented policing school. Robert Trojanowicz and his colleagues at Michigan State University represented the community policing school of thought. Mark Moore and George Kelling, at Harvard University, sought to synthesize community policing and problem-oriented policing, drawing heavily on theories about organizational strategy. Kelling's own brand of police reform, now popularized as the "broken windows theory," would later come to be seen as another school of thought. Lee Brown, then-police chief in Houston, represented the neighborhood policing school of thought. David Couper, then-police chief in Madison, was a prominent proponent of yet another school of thought–the total quality management school. Lawrence Sherman (1998) proposed "evidencebased policing" as an alternate construct for improving police service.49 The distinctions between and among problem-oriented policing and other police reform movements, including community policing, are discussed at greater length in chapter 3. 49Sherman attempts to draw a distinction between evidence-based and problem-oriented policing based on the scientific measurement standards he asserts are central to evidence- based policing and largely ignored by problem- oriented policing. The distinction is more one of degree than of kind. See chapter 1, the sections titled "What Standards of Proof Should Apply in Analyzing Problems" and "How Should the Effectiveness of Implemented Responses Be Evaluated?" for further discussion of standards of proof in problem-oriented policing. It would itself be a great oversimplification and distortion to suggest that these schools of thought and their respective adherents were in diametrical disagreement with one another. In fact, they agreed a lot, at least about the need for improvement in policing and for improved relations between the police and the public. One can more accurately understand these schools of thought as different ways of conceptualizing some common themes. The respective schools of thought had different themes as their organizing principle and gave differing priorities to the aspects of reform they held in common. All of their views, now considered mainstream, represented the radical or reform view of policing only 20 years ago. Chapter 1 Revisiting the Basic Elements of Problem-Oriented Policing In this chapter, I will revisit the core elements of Goldstein's problem-oriented approach to policing. In doing so, I will present some of my own insights into each element's meaning, and comment on observable trends in each element's application in the current practice of problem-oriented policing. What Is the Distinction Between Problem-Oriented Policing and Problem Solving? The terms problem-oriented policing and problem-solving are often used interchangeably in the literature that has built up around Goldstein's concept. This masks an important distinction between the two. In its broadest sense, the term problem-oriented policing, as used by Goldstein, describes a comprehensive framework for improving the police's capacity to perform their mission. Problem-oriented policing impacts virtually everything the police do, operationally as well as managerially. The term problem-solving, which came into more prominent use by other scholars in the mid-1980s, more specifically describes the mental process that is at the core of problem-oriented policing. Problem- solving models such as SARA50 were created to express this mental process. Thus, "problem-solving" is a more limited notion than "problem-oriented policing." Goldstein himself has been especially careful to avoid using the term problem-solving too freely, precisely because, as he argues, many, if not most, of the problems the police confront are too complex for anything approaching a final solution. Reducing harm, alleviating suffering and/or providing some measure of relief from problems are ambitious enough aims for the police. Other scholars, such as Ron Clarke, have also distinguished between problem-oriented policing and problem-solving, but have done so by drawing distinctions based on the scope of the initiative. In Clarke's view, the more routine activities of beat-level police officers to address recurring problems involving a single location or person constitute "problem-solving."51 Clarke contrasts this with more ambitious initiatives by police agencies to carefully study entire classes of problems and to make more systemic improvements in the response to those problems (1997). The precise distinction between problem- oriented policing and problem-solving is less important than the fact that problem-oriented policing is considerably more ambitious and far- reaching than routine and generic forms of problem-solving. 50SARA is an acronym for scanning, analysis, response, and assessment, developed as a problem-solving model by Police Executive Research Forum researchers during the Newport News problem-oriented policing project. Variations include CAPRA - an acronym for clients, acquire/analyze information, partnerships, response, and assessment of action taken - developed by the Royal Canadian Mounted Police and PROCTOR - an acronym for problem, cause, tactic (or treatment), output, and result - described in a report on problem- solving in the U.K. by Read and Tilley (2000). 51Other scholars mistakenly view this informal, beat-level problem-solving as the sum and substance of problem-oriented policing. One textbook puts it this way: "In other words, problem-oriented policing is simply an attempt to formally articulate the principles and processes that veteran police officers have long recognized as the essence of high-quality policing" (Fyfe et al. 1997:393). Widespread use of the term problem-solving has created another kind of confusion. Many people not familiar with the problem-oriented policing literature understand the entire notion of police problem- solving differently than do those familiar with problem-oriented policing. Two examples typify the confusion. In the first, I was a guest on a Canadian radio talk show, the subject of which was problem- oriented policing. The host pressed me to explain to his audience why the police should seek to solve the personal problems of criminals rather than arrest criminals. In the second, during a recent meeting of community policing experts and domestic violence prevention experts, the domestic violence experts questioned why the police would seek to problem-solve domestic disputes rather than arrest batterers (Ohlhausen Research Inc. 1999). Both inquiries reveal how some people, recently exposed to the concept and terminology of problem- oriented policing, understand problem-solving to be a form of mediation in which police give at least as much attention to offenders' personal interests as they do to victims'. It is little wonder that someone with this understanding of problem-solving would find it unappealing. What Does “Problem” Mean in Problem-Oriented Policing? The dictionary defines problem as “any question or matter involving doubt, uncertainty or difficulty”, or a “question to be considered, solved or answered.” Given this broad definition of the term, it is easy to see how its use in the context of problem-oriented policing can get distorted. The current literature on policing finds the term “problem” popping up everywhere. It describes difficult employees, administrative concerns, complicated crimes or incidents–in short, any matter involving difficulty. There are references to problem-solving courts52 and problem-solving in correctional facilities. Teaching young children and students methods of resolving conflicts is often referred to as problem-solving.53 This is not to suggest that the phrase is always being used inappropriately, only that Goldstein's use of the term in the context of problem-oriented policing is highly specific. He used the term to convey the notion that one can classify, package and understand police work in a new way, as an aggregation of incidents that share certain common features. 52Some of the labels associated with problem- solving in courts are therapeutic jurisprudence, restorative justice and community justice (Rottman and Casey 1999). 53Some references of this type make no claims to being related to problem-oriented policing (e.g., an Office of Juvenile Justice and Delinquency Prevention article, "Preventing Violence the Problem-Solving Way," refers to developing interpersonal cognitive problem- solving skills in children). Others explicitly seek to connect this sort of problem-solving to the concept of problem-oriented policing [e.g., in "Reducing Fear in the Schools: Managing Conflict With Student Problem-Solving" (Kenney and Watson 1998), published in Problem- Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work (O'Connor, Shelley and Grant 1998), the authors described a program in which police officers taught problem-solving skills to high school students]. The policing language when Goldstein first proposed his problem- oriented approach lacked any other term to capture what he was trying to describe. Police work was understood primarily in terms of crimes, incidents, events, and calls for service, and occasionally, as a series of various classes of these things. Because, as Goldstein observed, police agencies were designed primarily to respond to isolated events, they lacked the terminology to communicate about aggregated events. Goldstein might have chosen other nouns to convey his notion of a problem, such as "cluster," "aggregate" or "class," or adjectives like "chronic," "recurrent" or "habitual." Such terms might have better captured the essence of his idea, that much of what the police do is connected, not isolated–connected to past events, to similar events, to seemingly unrelated social phenomena–and that these events will continue to occur without more profound interventions. Goldstein wrestled with the question of giving his concept a label at all, for a variety of reasons, one of which was the risk of oversimplification and distortion of the basic meaning of the concept. Some of his concerns have proven well-founded. Many uses of the term “problem” in policing today seem to be artificial attempts to graft the concern onto the framework of problem-oriented policing, or merely to capitalize on the growing popularity of the terminology.54 One gets the sense that nearly everything published in recent years remotely related to policing invokes the terminology of problem- solving, no matter how loosely connected it is to Goldstein's concept of problem-oriented policing. Some people have proposed the use of the term solution-oriented policing instead of problem-oriented policing, presumably to put a more positive spin on the concept. Whatever purpose this might serve in terms of marketing the concept, it draws attention away from the most central feature of Goldstein's concept, the idea of aggregated incidents that constitute part of police business. The term problem-oriented policing has survived in the literature for over 20 years now, long enough for a reasonably common understanding of it to have emerged, and so there is little to be gained at this stage by altering the terminology. The precise understanding of the term “problem”, however, remains much in need of reinforcement. 54A recent compilation of articles drawn from the various periodicals published by the Community Policing Consortium was titled Problem-Solving. The collection does include a section on specific police projects addressing community problems, but it also contains articles on such matters as jail management, grant writing and court security. In some training courses in problem-oriented policing, and in the corresponding materials developed over the past decade, a precise definition of the term “problem” has been put forth. With slight variations, this definition has been something like the following: A problem is two or more incidents, similar in nature, that [are] of concern to the police and to the public.55 In a general sense, this definition is consistent with Goldstein's model. It is misleading, however, in that it suggests a mathematical precision to defining problems, never intended by Goldstein. There is nothing magical about the number two.56 Even a single incident might prompt a broader investigation into an underlying problem if the incident is of sufficient consequence to the community and there is a high likelihood of future similar incidents.57 Nor does the mere occurrence of multiple similar incidents automatically constitute a problem. The number of incidents must be both substantial and significant to warrant defining the pattern as a problem; that is, the number of incidents should exceed some norm or expected occurrence rate. The common feature among the incidents, moreover, must bear some significance. Exploring the pattern of burglaries committed by males, for example, is not likely to be productive, given that there is nothing remarkable about burglars being male. 55The training curriculum in community problem-solving distributed by the Community Policing Consortium uses this definition. 56The notion of "two or more incidents" in the definition of a problem may derive from a misreading of the definition of a problem proposed by Eck and Spelman in Problem- Solving: Problem-Oriented Policing in Newport News. There, a problem is defined as "a group of incidents occurring in a community, that are similar in one or more ways [emphasis added], and that are of concern to the police and the public" (p. 42). 57In some police agencies where managers pressure officers to solve lots of problems, officers sometimes classify merely interesting or complex incidents as problems. To be sure, the quality of the police service delivered matters more than how it is classified or labeled, but defining every clever, creative or preventive response to an incident drains from the intended meaning of "problem." “There has been a tendency to simplify and reduce the problem-solving concept, and to focus on particular innovations rather than the systems and managerial behaviors that produced them. This tendency is by no means unique to the police.” – Malcolm Sparrow How Should Problems Be Defined and Described? How one defines a problem greatly influences how one will address it. One can define or describe problems in a variety of ways. One can describe them in terms of what the offensive behavior is (e.g., playing loud music), who the people involved are (e.g., vehicle owners playing loud music from high-powered car stereos), when the problem occurs (e.g., late at night), or where the problem occurs (e.g., in a park in a residential area). These various descriptors obviously are not mutually exclusive. The descriptor is merely a shorthand way of describing the entire problem. However one describes a problem in shorthand, one must address the offensive behavior. This is important for several reasons. Without a clear focus on specific forms of offensive behavior, the police run the risk of adopting broad, ineffective responses. While it is sometimes convenient to describe problems in terms of a class of people or even one individual, it is dangerous morally, ethically and legally for the police to treat a person or people as the problem itself. A common example is the police response to a variety of problems surrounding transients (also referred to as "street people" or "the homeless"). The police often speak in shorthand about the "transient problem," because the underlying behavioral problems are common, yet too numerous to articulate briefly. If this shorthand, however, leads the police to view the mere presence of transients as the problem and, consequently, to address the problem by removing the transients, they risk committing serious violations to the transients' rights and to their own professional obligations. The shorthand description, "transients in a public park," is made more explicit by labeling it "transients sleeping and panhandling in a public park." This simple change draws one's attention to the behavior and not merely the status of the persons involved. Shorthand labels can also mask important distinctions between legitimate and illegitimate behavior. Describing problems as the "drug problem" or even the "narcotics problem" is so broad as to be nearly useless. Any shorthand label for a problem should be followed by a more complete and exacting description of the specific offensive behavior. I have sometimes reminded myself and others of this rule of problem-oriented policing in grammatical terms, by saying, "If you don't have a verb, you don't have a problem." Forcing oneself to include a verb in the description of the problem helps maintain the appropriate focus on problematic behavior. Many problem-oriented policing initiatives are defined too broadly (Clarke 1998). Characterizing problems with broad labels like "drugs," "violence," "disorder," "neighborhood decline," or "juveniles," without specifying the behavior at issue, often results in a simplistic analysis of the problem and, consequently, to hopelessly inadequate responses. Overly broad definitions of problems also create the risk that the police will be drawn into trying to address aspects of a large problem that are well beyond their capacity or mandate. For example, the police are unlikely to have the organizational expertise or capacity to stimulate economic redevelopment of a neighborhood in decline, though they can make substantial contributions in a larger partnership effort. There is evidence that when social issues are defined too broadly, the psychological capacity of those attempting to address the issues is diminished (Weick 1984). When problem-solvers consciously seek to redefine or reclassify the problem on the basis of preliminary analysis, this process often leads to conceptually clearer and more manageable initiatives. Asking whether the problem looks any different upon closer analysis remains a vital step in the problem- solving process, but it is too often overlooked. What Should the Police Be Concerned About in Problem- Oriented Policing? The principle of problem-oriented policing that asserts that its focus should be on community problems has led to some confusion in practice. This principle actually incorporates several distinct ideas. The first is that the police should primarily be focused on community problems, as distinct from organizational, administrative and managerial problems of police and other agencies. The second is that the problem-solving inquiry should seek the best response to the substantive problem at hand, and not merely seek to perfect existing systems and processes for addressing that problem. The third is that the community problems the police should focus on are those that fall within their mandate. I will discuss each idea in turn. “Problem-oriented policing is a larger concept than mere problem-solving. It has tremendous ramifications for the structure of police organizations.” – Rana Sampson Focusing on Community Concerns vs. Internal Concerns Goldstein's starting point for articulating the problem-oriented approach was that police managers should focus on how their agencies address community problems and not merely on how their agencies are administered and organized. Getting police to refocus on community concerns is in itself a significant challenge. When one asks police practitioners to list problems confronting their agencies, without specifying whether those problems should be substantive community problems or internal problems, they prioritize the internal problems. Those problems are the most immediately apparent to the practitioners, and affect them in a most personal way. They affect their physical safety, career opportunities, financial status, and general occupational contentment. The police are no different in this regard from practitioners in other fields. Ask most medical practitioners today to list their problems, and one can expect to find managed care higher on the list than emphysema or heart disease. Teachers talk more about classroom discipline than how to teach algebra more effectively.57 Typically, however, a simple reminder to police practitioners to focus on substantive community problems will readily get them engaged in discussions of the problems of crime and disorder. Getting this to occur routinely is far more difficult. 57Other fields such as medicine and education have a substantially larger body of research regarding the substance of the work than do the police, so even if one could get the police to talk more about substantive community problems, there would be less research to inform those discussions than is the case in other fields. Existing case studies in problem-oriented policing demonstrate that the police are capable of using problem-solving methods on substantive community problems. But, if the police continue to focus exclusively or primarily on internal organizational problems, even if they apply some problem-solving methods toward their resolution, then problem-oriented policing will have failed on its face. Ironically, the development of the concept of problem-oriented policing has suffered somewhat from the means-over-ends syndrome Goldstein described regarding the management of police agencies. One can more readily find literature on how to implement problem- oriented policing within a police agency than find literature on how police have applied problem-oriented methods to specific community problems. For years, the planners of the annual Problem-Oriented Policing Conference held in San Diego debated how much to emphasize workshops and presentations on the implementation of problem-oriented policing vs. workshops and presentations on police responses to substantive community problems. In the early years of the conference, the implementation workshops drew greater interest from conference attendees, and conference planners tried to balance those expressed interests with a desire to get attendees interested in substantive problems. Despite efforts over the years to balance substance and process, the issue continues to arise, and implementation always seems to be favored over substance. In some sense, this serves as partial confirmation of Goldstein's original premise that the police are highly susceptible to the means-over-ends syndrome. Finding the Best Response vs. Merely Improving Current Responses and Systems Problem-solving inquiries should seek the best response to the substantive problem at hand, and not merely seek to improve current responses and systems. This distinction is a subtle but important one. A fair number of problem-oriented policing projects, including those submitted for the Herman Goldstein Excellence in Problem-Oriented Policing Award, are essentially efforts to improve a criminal justice or investigative process, devoid of a careful inquiry into whether that process is the most effective means of addressing the problem in the first place. For example, in recent years, a number of police agencies have recognized value in establishing more collaborative working relationships with probation and parole agencies. Accordingly, a number of problem-oriented policing projects have set about finding ways for the police and probation and parole agents to more effectively and efficiently supervise people under conditional release. The underlying logic, of course, is that more effective and efficient supervision will reduce the levels or seriousness of crimes committed by those people. In many instances, however, the assumption that supervision of previously convicted offenders is the best response to the problem goes unexplored and unchallenged. As a matter of general practice, improved collaboration between the police and probation and parole agents is good strategy and should be encouraged. The value of that collaboration, however, becomes stronger if it is first clearly established that improved supervision will result in substantial improvements to the specific community problem. Some police practitioners find this explanation frustrating. Having satisfied the threshold requirement to focus on a substantive community problem rather than an internal organizational problem, they don't readily see how they can become fixated on a particular process that they intuitively believe will adequately address the crime or disorder problem. Again, the efforts to improve investigative, prosecutorial, adjudicative, or correctional processes using a problem- oriented approach is commendable, but falls short of the ideal if it is not first demonstrated that these processes are best able to reduce the harm the problem causes. The best problem-oriented policing efforts are those that remain focused on the end objective–some form of harm reduction–and develop or improve processes only as a means toward that end. Focusing on Community Problems for Which the Police Should Assume Some Responsibility The community problems the police should focus on are those that fall within their mandate as it is defined for each agency. Here, too, is a great source of confusion. What is, precisely, the police mandate is a hard question, to be sure. In the era of community policing, that mandate has been expanded considerably, partly by the police themselves. Police departments everywhere are initiating programs in which police officers adopt roles of counselors, teachers, coaches, and brokers of charitable works. The police adopt these unconventional roles for various purposes. The best practice is when police officers adopt these roles for a limited time, and in the specific context of addressing a specific community problem (e.g., a police officer organizes a youth activity to provide youth an alternative to delinquency, and then turns the program over to someone else). Too often, the police adopt these roles for other purposes–merely to improve their community image or deflect criticism of other, objectionable, police practices. The most common justification offered for adopting these new roles is that the police can inculcate good moral and civic habits in the community, and as a result, some unspecified measure of offending will be reduced. The most apparent example of this expanding police mandate is the D.A.R.E. program, in which police officers become part-time teachers in the moral education of grade school students. D.A.R.E. has become an industry within an industry, and in spite of evaluations that conclude the program does not achieve its original objectives to reduce illicit drug use by young people (Rosenbaum et al. 1994, Sherman et al. 1997), it remains enormously popular with schoolchildren, parents, some educators, elected officials, and D.A.R.E. officers. However commendable these efforts may be in moral terms, they are all subject to stricter scrutiny in a problem-oriented policing context. Goldstein has advocated that the police recognize their role in society as being broader than enforcing the criminal law. At the same time, however, he has argued that the police mandate must not be unlimited. If the police become too involved in every government and quasigovernment function, they risk eroding balances of power in local and even national government.58 The police's moral authority, derived from their powers to arrest and to use force, can easily be misused to advance particular moral or political viewpoints. As with the military, there is a sound political rationale for keeping the police out of certain realms of social decision-making. Police agencies run the risk of overextending their expertise and resources–trying to achieve objectives about which they have little or no expertise. By expending resources on newly adopted mandates, they risk devoting too few resources to conventional mandates. 58David Bayley, while commending the problem-oriented approach, expressed concern that "problem-oriented policing could transform the Anglo-Saxon notion of a restricted and specialized police role into the Continental one of an omnicompetent police" (1991). See, also, Vaughn (1992), who argues that the American public desires a restricted and reactive police and that a "truly proactive police structure may be impossible to implement in our current democratic system of government" (p. 352). Ideally, each police agency should develop a clear and firm understanding of its mandate. This will and should vary from agency to agency, and should be shaped by community desires. Few police agencies or communities have expressly advocated that the police assume responsibility for morally educating children, entertaining children through recreation or brokering charity. Those responsibilities are assigned elsewhere. Under a problem-oriented policing scheme, the police would recognize how functions like moral education, youth recreation and charity are integral to public safety, but would not see their role as one of providing these services directly, at least not permanently. The key for the police is first, to establish some sense of ownership or responsibility for a community problem, and if the problem falls within the police mandate, either address it themselves, broker ownership to some other entity or, in some instances, merely refuse to accept ownership. If the police assume ownership, they then must establish some clear nexus between the problem, its causes and the proposed response. In so doing, the police may well conclude that a youth recreation program, for example, is precisely what is needed to address a particular problem. But that may not be enough to justify having the police provide that service. Once the police establish the nexus between the problem, its causes and the proposed response, they must then decide whether they should take responsibility for implementing the response. Returning to the previous example, the police might determine that a youth recreation program would be a viable response to a problem of after-school residential burglaries, but that would not necessarily mean that they should organize or run that program. I will explore this question of ownership of problems further later in this chapter in the section titled "How Should the Police Develop and Implement New Responses to Problems?" The best problem-oriented policing initiatives that call for responses outside the conventional police mandate are those that clearly identify the legitimate police interests in a particular community problem; establish a causal nexus between the crime, disorder or safety problem and the gap in services; and limit the police role in delivering the new services to that of catalysts, advisors or referring agency. A good example was provided by the Glendale, Calif., Police Department when in 1997 it helped develop a new program for day laborers that directly responded to legitimate police interests in reducing crime and disorder. The police did not assume responsibility, however, for actually running the program. Similarly, the Fontana, Calif., Police Department in 1998 helped develop a new assistance program for transients that achieved similar objectives without assuming the large responsibility of administering the program. In problem-oriented policing, it isn't the nature of the response that determines its quality; it is the link that is drawn between the response and legitimate police interests. The police may join with many divergent entities in studying a problem, but ultimately the responsibilities for various responses should be apportioned among those entities according to their resources and competencies. It is important that the police distinguish between taking an interest in seeing a particular problem handled more effectively and taking responsibility for implementing new responses. In some instances, the one may lead naturally to the other–when, for instance, the police take an interest in improving the response to commercial robberies, and then take responsibility for implementing new responses to the problem. In other instances, the police will properly take an interest in a problem that comes to their attention, but refuse to assume responsibility for responding to that problem. For instance, the police are properly interested in the health and welfare of homeless people, and may want to see improvements in the services provided to them, but might not accept responsibility for distributing food and clothing to those in need. The demands of providing social services can easily overwhelm police officers and their agencies. The failure to establish limits to the police mandate and to apportion responsibilities for addressing problems appropriately has led to some backlash among the police. Some police agencies that have become heavily immersed in problem-oriented policing now feel the need to remind themselves and others that their core responsibilities are to address crime, and not to do social work. While this sort of backlash runs the risk of taking the police back into an artificially narrow understanding of their role, it is understandable given some officers' and agencies' tendencies to try to expand their mandates into popular programs and tasks that have little demonstrable impact on crime, disorder and safety. The backlash may actually reflect continuing frustration on the part of some police that, having identified a community problem that does not fall squarely within their mandate, they get stuck with the responsibility to respond to the problem merely because other entities refuse to do so. What Does a Search For Underlying Conditions, Contributing Factors and Causes Really Mean? Implicit in any effort to solve a problem is the effort to understand why the problem exists. Goldstein explicitly calls for "an in-depth probe of all of the characteristics of a problem and the factors that contribute to it…" (1990a:36). The search for underlying conditions, contributing factors and causes has raised a number of practical issues for the police in their efforts to employ problem-oriented policing. For each problem they explore, they must consider how deeply they should look to understand why it exists, how certain they should be in their understanding, what might be done to effectively address the problem, and who should assume responsibility for taking action. In doing so, they (and those who must authorize and support their decisions) must make critical judgments about how best to study the problem. The direction and depth of the inquiry will shape how police and the community respond to the problems. For example, in exploring problems associated with youth gangs, one could probe deeply into the underlying sociological and psychological motivations of gang membership which might lead to interventions designed to change the underlying social conditions in which youth gangs exist. In the alternative, one could limit the probe to understanding the ways youth gangs operate which might lead to more practical interventions designed to reduce the harm to the community caused by youth gangs. Problem-oriented policing calls for judgments in these matters that are neither too ambitious and leading to inquiries into the unknown or unknowable, nor too simplistic and leading to inquiries that only scratch the surface of complex problems. Root Causes vs. Underlying Conditions The search for contributing factors, underlying conditions and causes is sometimes confused with efforts to address the broadest of social and psychological factors that contribute to crime and disorder, factors often referred to as the "root causes" of crime and disorder. Criminologists have long debated the root causes of crime and disorder, some looking to individuals' psychological motivations such as greed, jealousy, anger, or mental defects; others to sociological conditions such as economic deprivation and racism. A new school of thought has expanded the notion of causation of crime and disorder to include opportunity as a causative factor. Scholars like Ronald Clarke and Marcus Felson (1993) argue that the ease or difficulty of committing crimes and evading detection impacts crime and disorder in a real sense. They assert that opportunity is more amenable to intervention than either individuals' psychological predispositions to commit crimes or broad sociological factors that influence crime. In this regard, the search for root causes is not necessarily or exclusively a search into the soul of the human condition, but includes a search into the more mundane ways humans arrange their physical world. Associating problem-oriented policing with a search for "root causes" is misguided in two important respects. First, it suggests that effective responses can be found only by addressing the most fundamental factors or human conditions that give rise to problems. Second, it implies a degree of certainty about causation that is seldom achievable. However helpful a deep probe into people's social or psychological conditions might be in understanding crime and disorder, it seldom proves practical in achieving more immediate police objectives. Goldstein's problem-oriented approach is compatible with Clarke and Felson's theories of crime opportunity because it seeks both to understand and to effectively intervene. It looks for the deepest underlying conditions that are amenable to intervention, balancing what is knowable with what is possible. If one thought of the contributing causes of crime and disorder as falling on a continuum ranging from root causes to underlying conditions to marginal contributing factors, and the capacity of the police or others to intervene and modify these factors as falling on another continuum ranging from greatest capacity to intervene to least capacity to intervene, then one could conceive of appropriate problem-oriented interventions as ones that optimized the capacity for intervention in significant contributing factors. Generally speaking, it is easier to intervene in factors that contribute least to crime and disorder, and most difficult to intervene in factors that contribute most to crime and disorder. For example, hate is a root cause of some types of assault, yet the police and others have little capacity to intervene to modify the hatred that fuels the violence. Weapon availability is a less strong contributing factor to the assault than is the underlying hatred, but the police and others have a greater capacity to intervene to modify weapon availability.60 A number of other factors, like threat of punishment, likelihood of detection, presence of people to interrupt an assault, intoxication, agitating events, etc., contribute in varying degrees to the assault. Each factor also varies in the degree to which the police and others can intervene to modify it. Problem-oriented policing calls for finding that combination of capacity for intervention and strength of causation that offers the most promise for reducing the likelihood, frequency or severity of the assault in the present. Thus, the search for causation in problem-oriented policing is practical rather than theoretical. The primary goal is to reduce future harm, and not so much to establish blame or redress past harm.61 60Research and practice in the United Kingdom have demonstrated that reducing the use of certain types of breakable glassware in pubs can significantly reduce the severity of injuries to victims of barroom assaults (Sherman 1990). 61An editorial in the Savannah Morning News (1999) on gun control legislation highlights the debate on the causes of violence. The editorial read in part: "The real question isn't how kids kill. It's why kids kill. Until Congress is willing to look at what's inside a kid's mind, as opposed to the pistol or shotgun in his hands, the problem is far from resolved." Whatever truth this holds from a philosophical perspective, from a practical police perspective, probing the minds of violent youth is less likely to reduce violence than controlling the instruments they use to commit it. Causation vs. Blameworthiness Causation and blameworthiness are complex questions in any context. When the police respond to isolated incidents of crime and disorder, they may or may not be concerned with establishing the causes of these incidents. In many instances, the police are either not at all concerned with causality, limiting their objectives to restoring peace and order, or their interest in causality is limited. If the police define the incident as a crime, then of course they will set about establishing causality and blameworthiness in a legal sense (X assaulted Y and should be punished); but they may not be interested in establishing causality in the broader sense, in which they seek to understand the conditions and dynamics that gave rise to the incident (X assaulted Y partly as a consequence of crowded conditions in the bar). Problem- oriented policing implies a concern with causation in the broader sense. Causation and blameworthiness are not synonymous, however. To say that X caused Y does not automatically mean that X is blameworthy.61 In problem-oriented policing, the search for causation is also broad because the primary interest is in preventing recurrences of the harm and equitably distributing the responsibility for preventing the harm. This broad notion of causation in a problem-oriented policing approach can make the approach controversial. The entity most capable of bringing about an improvement to the problem may not be the entity generally deemed most blameworthy for that problem. Applying the notion of blameworthiness inherent in criminal law typically implicates those people who actually commit crimes or are nuisances. Problem-oriented responses affix responsibility on those entities most capable of effecting lasting improvements to the conditions that give rise to the crime and disorder. The notions of assuming responsibility for prevention and assuming blame for the problem can become confused. To many police officers, steeped in the legalistic traditions of assigning blame through the enforcement of the law, the process of spreading out responsibility for responding to problems does not come naturally. Their training has taught them to look for the people or entities most responsible for causing the harm, and compelling them to account for that harm. 62The search for causation in the context of problem-oriented policing carries with it similar dilemmas as does the search for causation in the context of criminal law. It raises questions about the distinction between causes in fact and proximate causes. Establishing a cause in fact requires a showing that the harm would not have been suffered but for the act in question. If the harm would have occurred anyway due to other acts or factors, then the act in question is not considered the cause in fact. Establishing proximate causes requires a judgment that the act in question is sufficiently closely connected to the harm to be deemed blameworthy. Criminal law is more demanding than civil law on the question of causation; that is, it is concerned primarily with establishing blame and consequently administering punishment to those people most responsible for causing harm. Civil law is more concerned with restoring the victims of harm to wholeness, and thus allows for somewhat more expansive notions of causation and responsibility. Criminal law resolves these questions through case law and statutes that slowly evolve around socially acceptable notions of culpability. In the problem-oriented policing context, these questions are resolved on an ad hoc basis, depending heavily on the judgment of those leading the problem-solving initiative, and the judgment of the affected parties as expressed through a variety of political and administrative processes. An example that stands out from my own experiences in problem- oriented policing training is the now well-known Gainesville, Fla., convenience store robbery problem (Clifton 1987). This case study is often used in problem-oriented policing training to demonstrate how a problem-oriented approach can prove more effective at preventing serious crime than can a strictly criminal-law approach. The essence of the Gainesville response was to assign some increased responsibility to convenience store owners to add staff as a robbery prevention measure. This increased the owners' short-term costs. The presentation of this response in police training sessions around the United States evoked strong but mixed responses. Some officers endorsed the response as reasonable and effective. Others objected to the very idea that convenience store owners, who clearly were not the ones actually robbing their own stores, should bear any additional responsibility for others' criminal conduct, regardless of how effective the measure might prove. Effective problem-solving requires that the police, and all parties with a stake in the problem, place a higher priority on improving the overall response to the problem than on assigning blame for the problem. This is much easier said than done, of course. The two ideas of causation and blameworthiness get intertwined easily, and people then equate accepting responsibility for changing their practices with accepting the blame for causing the problem. This is why it is so critical that the police develop effective working relationships with those affected by a problem, relationships built in a spirit of mutual trust, to overcome the natural defensiveness that accompanies discussions of causation, blame and responsibility. What Standards of Proof Should Apply in Analyzing Problems? The search for causation also raises issues regarding the standard of proof to be applied.63 Policing is both a social and a legal enterprise, but neither the strict standards of social science research nor the standards of the law are entirely suited for settling questions of causation and responsibility in the context of problem-oriented policing. 63The rules of evidence govern the search for causation in criminal and civil law, and the standards of proof are well-established (proof "beyond a reasonable doubt" in criminal law, and a "preponderance of the evidence" in civil law). Some police scholars, like Lawrence Sherman (1998), advocate setting high standards of social scientific proof in problem-oriented policing, standards that can best be met by rigorous application of experimental testing conditions. Other scholars, like Ron Clarke, have advocated a more flexible standard of proof, a standard Clarke has occasionally referred to as "good-enough proof." By "good-enough proof," Clarke means that the standard of proof sufficient to support a particular new response to a problem should take into account the severity of the problem, the costs of being wrong, the research skills of the problem-solvers, the practicality of various research methods, the body of existing knowledge about the particular type of problem, and so forth. Goldstein and Susmilch (1981) also advocated more flexible and adaptive standards of proof, standards that ideally will be raised as the problem-oriented methodology becomes more advanced and the body of research on what causes various problems, and what works in controlling those problems, grows.64 Professor Nick Tilley (a police scholar at Nottingham Trent University in England who is emerging as Herman Goldstein's counterpart in British policing), and Pawson have argued that quasiexperimental methods are poorly suited for establishing causality in such a complex enterprise as policing. They, too, favor a form of evaluation they call "scientific realist," an approach that takes into careful account the specific context in which the policing is occurring and the precise mechanisms the police use to effect change, and that builds knowledge on the basis of a series of effective and ineffective responses (Pawson and Tilley 1994; see, also, Tilley 1993). 64Goldstein and Susmilch offered an important caution regarding these flexible research standards when they wrote: "These adjustments will frequently require relaxing social science standards of proof… Relaxing such standards of proof is both complex and hazardous. There is a very thin line between the eclectic research we propose here and shabby or bad research" (1981:95). As a practical matter, the standard of proof that ultimately will prevail varies from problem to problem and place to place. Unlike criminal and civil law, problem-oriented policing is not primarily governed by tribunals that apply uniform rules and laws. (Occasionally, however, the courts do serve as arbiters of problem-oriented decisions, as occurred in the Gainesville situation, in which a federal court ruled in favor of the proposed problem-oriented response.) Within the broad limits of the law, what stands as an acceptable response to any particular problem depends on what is acceptable to the local community, at least those members who are paying attention to the problem and can exercise influence on the particular policymakers. How Should the Police Analyze Problems, and How Well Are They Doing So Now? By most accounts from those who observe problem-oriented policing carefully, problem analysis remains the aspect of the concept most in need of improvement.65 Goldstein's ideal model of problem-oriented policing calls for analysis that is systematic, thorough, insightful, discriminating, and honest; that is, the analysis should provide the most comprehensive understanding of the problem possible. This ideal is rarely achieved in the practice of problem-oriented policing for several reasons. Practitioners and even some researchers do not always fully appreciate the value of analysis in responding to problems or understand precisely what information should be analyzed; the resources available for analysis, including appropriate research expertise, are often inadequate; and good analytic systems are often lacking. 65See Capowich and Roehl (1994) and Police Executive Research Forum (2000). Michigan State University Professor Tim Bynum is preparing a written problem-analysis guide for police. He has noted from his review of many problem-solving projects that this stage of the process is typically weak. The forthcoming analysis guide was funded by the COPS Office and is due to be published and disseminated sometime in 2001. The Value and Limits of Analysis In order for the police to commit adequate resources to analyzing problems, they must first fully appreciate how analysis can improve their responses to problems. In order for researchers to help the police with analysis, they must appreciate the practical concerns of and demands upon the police with respect to community problems. (These issues are discussed more fully in chapter 4.) A thorough problem analysis, at a minimum, means fully describing the problem, describing the multiple and often conflicting interests at stake in the problem, calculating the nature and costs of the harm arising from the problem, and taking inventory of and critiquing the current responses to the problem. In the problem-oriented policing model, problem-solvers, whether they be police practitioners or researchers, should be open to doubt about things they thought they knew about the problem and insist upon proving or disproving matters with objective evidence. They must balance the desire to be certain and precise with the practical difficulties in being so. They must recognize what data can and cannot tell them. They should be interested in learning how similar problems have been analyzed and addressed elsewhere while at the same time recognizing how their local situation might be different. They must ask the right questions and not waste effort finding answers to questions of no practical significance. They must balance the need to reflect on problems with the need to act upon them. These are no small challenges and they require that both police practitioners and researchers adjust and adapt the conventional ways in which they analyze problems and decide how to respond to them. Inadequate Analysis Resources Problem analysis can fall short of ideal without adequate time to complete the analysis and the research expertise necessary to do so properly. In Goldstein's original conceptualization of problem- oriented policing, a typical inquiry into a problem would either be headed up by or have substantial involvement of trained researchers. While he recognized the value of more modest problem-solving initiatives undertaken by street police officers, his ideal has always been for more substantial inquiries of larger-scale problems. In practice, far more street-level problem-solving of localized problems has occurred than have higher-level inquiries into communitywide problems. In most such instances, street police officers have had little in the way of research support for their analysis beyond being provided with requested tallies of data from crime analysts or information managers. Missing is expert guidance on setting up an appropriate methodology for conducting the inquiry, assistance in ensuring the data are complete and reliable, and assistance in applying statistical data analyses from which valid conclusions can be drawn.66 Spending more time and resources on problem analysis would improve most problem-oriented policing initiatives. Good police- researcher collaborations are important in this regard. 66For a good discussion of common deficiencies in problem analysis and how police agencies can improve their analytical capabilities, see Clarke (1998). Problem Analysis Guides In place of the sort of expert guidance in research that would be ideal are some rudimentary guides for problem-solving and analysis. Many departments engaged in problem-oriented policing have developed local customized guides and forms intended to facilitate and promote problem analysis. Most police agencies engaged in problem-oriented policing teach and offer written guides in problem-solving processes, most commonly the SARA model and variations thereof. Some problem-solving and analysis guides have gone a long way toward providing street officers with some basic understanding of problem- solving methodologies, but they are not substitutes for the expertise trained and experienced researchers provide. A few police agencies have developed, or are developing, more advanced computerized programs that allow officers to search for relevant information and make sense of it. The Leicestershire Police Force in England has developed a visually attractive and easy-to-use computer program that identifies incident patterns for officers that they might address in a problem-oriented way. The Chicago Police Department, supported by funding from the COPS Office, modified and enhanced a problem identification and analysis system, known as Information Collection for Automated Mapping (ICAM).67 Also funded by the COPS Office, a consortium of Massachusetts police agencies is working with University of Cincinnati Professor Lorraine Green Mazerolle and a computer company to develop and implement a computer program called the "Problem-Solver" that will similarly facilitate the search for useful data to analyze problems (Green Mazerolle and Haas n.d.).68 The Charlotte-Mecklenburg Police Department is undertaking a large project to develop a comprehensive and integrated computer system that will support and facilitate problem-solving, as well. Similarly, University of Cincinnati Professor John Eck is developing a generic problem-solving manual that incorporates routine activity theory and situational crime prevention. Eck's manual, as yet unpublished, is unique in that he is designing it to permit users to either work forward from problem analysis to responses or backward from desired responses to problem analysis. In either direction, it leads users into a detailed problem analysis. Eck's manual uses the conceptual framework of what is now commonly referred to as the "crime triangle" or "problem analysis triangle." The crime triangle is derived from routine activity theory and posits that all crimes (and, by extension, all problems) require victims, offenders and locations (Felson and Clarke 1998). Figure 1 Crime Triangle Image of Triangle with Sides Labeled: Offender, Victim, Location 67The ICAM2 system was designed with considerable input from line police officers, so that it would reflect both their information needs and their technical expertise. The system allows officers to search and query the department's records system and other community data systems in a variety of ways, and provides the data in graphics, statistics or maps. Details about the ICAM programs can be found in Buslik and Maltz (1998) and in a variety of internally produced Chicago Police Department documents. 68The Police Foundation is conducting an experiment to measure the impact a computerized mapping feature of this information system has on problem-solving activities. Eck then adds to his model the notions of "handlers" to describe people capable of controlling victims, offenders and locations; and "tools" to describe various instruments used to commit, prevent or facilitate problem behavior. Finally, he incorporates the notions of "streets" and "routes" to encourage analysis of the movement patterns that correlate with problem behavior. In some training in which the crime triangle is introduced, trainees are taught that to effectively impact a problem, at least two sides of the triangle must be addressed.69 As with the "two or more incidents" definition of a problem, there is no theoretical foundation for this rule. It is merely a means of getting new problem-solvers to think beyond simplistic responses. 69The community problem-solving curriculum distributed by the Community Policing Consortium makes this assertion. As is true with all guides, they can serve to either expand thinking or limit it. For a user short on conceptual skills or lacking in innovation, a guide can greatly expand the scope of his or her analysis. For both the newcomer and a more expert user, a guide might actually inhibit the scope of analysis by suggesting artificial limits. Both processes of identifying problems and analyzing them require some analytical methods and, accordingly, are often confused in practice. Many of the computer programs designed to help the police spot patterns of incidents that might constitute problems, like mapping and database programs, are more limited in the extent to which they can help the police fully understand the nature and causes of problems. They are most useful for alerting the police to the existence of potential problems, but they do not suffice for a complete analysis. For example, a computer mapping program might help the police detect an emerging pattern of commercial burglaries. It might even go so far as to pinpoint the more specific problem of burglaries of self-storage facilities. But knowing the spatial and temporal patterns of this problem alone is insufficient to guide the police in developing a new response to the problem. That will require another level of inquiry, one that will require looking at the facilities' physical layout, understanding the various management practices of the companies that operate the facilities, interviewing known offenders, finding out what kinds of property are stolen, etc. The data systems and research methods useful for identifying problems and for analyzing them in greater depth may overlap, but they are often quite different. The Action Research Model The sort of research model that Goldstein envisioned, and that he adopted in the early application of problem-oriented policing in Madison, is known as action research. In action research, the researcher is an integral part of a team of people working toward some particular result. The researcher not only collects and analyzes data and draws conclusions, but also proposes interventions along with others trying to intervene in the problem. This research model seeks to balance an outside researcher's independence and objectivity with a pragmatic interest in achieving certain results. In the beat-level practice of problem-oriented policing, the principal researcher is usually the police officer trying to intervene in the problem; that is, police officers become their own researchers. There have been only a few more ambitious collaborative initiatives of the sort Goldstein piloted, but they have been notable.70 Perhaps the best recent example is the collaborative work of the Boston Youth Gang Task Force and Harvard University researchers. Their work, recognized as the best submission for the 1998 Herman Goldstein Award for Excellence in Problem-Oriented Policing, blended sophisticated research capabilities with intimate knowledge of the particular crime problem. The effort resulted in significant reductions in homicides of young people in Boston. More recently, the Charlotte-Mecklenburg Police Department has invited Herman Goldstein and Ron Clarke to serve as external advisors on several large-scale problem-oriented policing initiatives the agency is undertaking. In addition to advising on the particular problems, Goldstein and Clarke are helping the department enhance its internal research and analysis capabilities to better support problem-oriented policing. A couple of federally funded initiatives are seeking to advance this type of action research on substantive community problems. The COPS Office's Problem-Solving Partnership Program provided funding to over 400 jurisdictions to apply problem-oriented action research techniques to selected crime and disorder problems. Each grantee was required to spend some funding on external research assistance, both to aid in problem analysis and to evaluate intervention outcomes. The results are mixed, with some good research collaborations and some nominal ones.71 The quality of the final reports will reveal more about the potential for action research in U.S. police agencies. The National Institute of Justice (NIJ) has recently funded a program called Strategic Approaches to Community Safety Initiatives (SACSI) in five cities.72 In those cities, the local U.S. attorney is to convene a group of relevant experts to identify a pressing public safety problem and, along with a researcher, develop a new response strategy. These initiatives are intended to be of a scope and sophistication comparable to the Boston effort described above and are also intended to train researchers in problem-oriented policing research methods. It remains to be seen to what degree these efforts are consistent with a problem- oriented policing approach, but at least, as designed, they incorporate the basic principles of the approach. 70The collaboration between the University of Missouri-St. Louis and the St. Louis Metropolitan Police Department to reduce levels of youth firearm violence (Rosenfeld and Decker 1998) is an outstanding example of an ambitious and sophisticated problem-oriented initiative. Although the project was not completely evaluated due to failures in police implementation, the published findings reflect a solid demonstration of many principles of problem-oriented policing. See, also, Homel (1998) for police-research collaborations on alcohol-related problems in Australia. 71PERF's (2000) evaluation of the Problem- Solving Partnership program concluded optimistically that the problem-solving model is viable, but noted many deficiencies in the grantee agencies' ability to follow the model and manage the problem-solving and grant process. It specifically noted common deficiencies in problem-solving training, data access and project staffing. 72The program has been funded in Memphis, Tenn.; New Haven, Conn.; Indianapolis; Winston-Salem, N.C.; and Portland, Ore. NIJ and the COPS Office have also funded a program called Locally Initiated Research Partnerships in Policing (McEwen 1999) to promote police agency and outside research collaborations. There is no requirement that the research focus be substantive, however, and a review of summaries of the 41 projects funded (McEwen and Pandey 1998) reveals that only about 15 of the projects study the police response to specific community problems (e.g., domestic violence, homicide, carjackings, gun violence, drug dealing, and workplace violence). The other projects are focused on such concerns as the implementation of community policing, public attitudes about the police, and crime mapping technology. It is to be hoped that the 15 projects that do focus on the police response to substantive community problems will reinforce the value of this type of research.73 73The 15 projects that are most promising from a problem-oriented research perspective are (1) Berkeley, Calif., Police Department/East Bay Public Safety Corridor Partnership, National Council on Crime and Delinquency (domestic violence); (2) Lexington, Ky., Division of Police/Eastern Kentucky University (displacement of drug dealing); (3) Boston Police Department/Harvard University, Northeastern University (unspecified problems); (4) multiple small Western police departments/LINC (violence against women and girls); (5) Bay City, Mich., Police Department/Saginaw Valley State University (nonviolent juvenile crime); (6) Hagerstown, Md., Police Department/Shippensburg University (neighborhood-based crime and fear of crime); (7) Framingham, Mass., Police Department/Social Science Research and Evaluation Inc. (domestic violence); (8) Buffalo, N.Y., Police Department/State University of New York at Buffalo (unspecified problems); (9) Oakland, Calif., Police Department/University of California at Berkeley (gun violence and problems at a local festival); (10) Forest Park, Ohio, Police Division/University of Cincinnati (unspecified problems); (11) Prince George's County, Md., Police Department/University of Maryland (homicides, carjackings, gun-related calls); (12) multiple small police departments in Alabama/University of South Alabama (unspecified problems); (13) Charlottesville, Va., Police Department, County of Albermarle Police Department, University of Virginia Police Department/University of Virginia (workplace violence); (14) Seattle Police Department/University of Washington (domestic violence); and (15) Arlington County, Va., Police Department/Urban Institute (unspecified problems). For years, before the advent of problem-oriented policing, the British Home Office has engaged in action research projects with British police forces, producing a body of reports on various problems. These reports, published under various series titles, are top quality and usually are the product of just the sort of research collaboration Goldstein has advocated. To date, the Home Office initiatives and reports are superior to any similar undertakings in North America. The Compstat method developed by the New York City Police Department, and now being emulated in many other agencies, extracts crime data from computers and subjects the data to scrutiny by panels of top-level police commanders and analysts, who then work with local commanders to interpret the data and develop appropriate responses. These inquiries vary in tone and style, but their primary purpose is to motivate police commanders to address crime problems, and to hold them accountable for doing so. They are not principally designed for careful problem analysis, though they promote some analysis. (I discuss the Compstat method and its relationship to problem-oriented policing more fully in chapter 3.) A different analysis method, the Problem Analysis Advisory Committee, has been pioneered by the Newport News Police Department and emulated elsewhere. It is primarily an analytical resource for those undertaking problem-solving projects. The problem-solvers query participating experts to encourage greater depth in the probing of a problem, and to guide their own analysis. The San Diego Police Department has used this method extensively. Accessing and Analyzing Police Data The widespread application of computers to police record-keeping has, for the most part, been a boon to the practice of problem- oriented policing. Data that just a few years ago would have been enormously difficult to retrieve are now available at the touch of a few buttons. From a research standpoint, the samples that can readily be amassed and analyzed are much larger than was practical in a paper record system. Unfortunately, the ease of searching and analyzing large volumes of aggregate coded data too often leads problem- solvers to skip a more detailed analysis of the written narratives in individual police reports. Goldstein has long claimed that the narratives contain many of the more useful insights about problems. Searching for Relevant Research and Good Police Practices Goldstein envisioned that an important aspect of problem analysis would be a review of the literature on that problem. That literature might be in published books and articles, or in unpublished reports from within and outside the police agency. In practice, however, literature reviews conducted as part of problem-solving projects are rare. Police practitioners often do not have the benefit of assistance from researchers or do not have access to research libraries.74 Recent research by Northwestern University Professor Alexander Weiss confirms that police agencies acquire and exchange information about practices more through personal contacts than by reading the literature (1997, 1998). How police agencies and officers communicate and share professional knowledge is a complex cultural matter. 74Where libraries exist at all in police agencies, they have typically been developed and designed primarily for occasional use by police recruits writing papers, attorneys doing legal research, or officers preparing for promotional exams. Tellingly, in some departments the library is located closer to the police academy than to the research and planning section. In chapter 5, I discuss further the police-research communication gap and propose some ways to close it. Searching for Published Research Unfortunately, even if police had more access to research libraries, or if trained researchers were conducting a literature review, it is not at all clear that their search would be that productive with respect to many types of problems. While there is more relevant research on some community problems than many police officers realize, it is far less than one might expect given how common many problems are and how many public resources are spent trying to address them. There simply isn't enough quality research conducted to reliably inform the police about what does and does not work with respect to most crime and disorder problems. Outside of a few specialized areas that have received substantial research interest, the body of applied research on crime and disorder problems is not large. Again, compared to the body of literature in most other professions, the amount of published research about common community problems seems miniscule. The recent series of publications titled Crime Prevention Studies, and two volumes titled Situational Crime Prevention: Successful Case Studies that developed out of the situational crime prevention model, come the closest to building a body of literature that is relevant to and useful for practicing problem-oriented policing. At present, there are twelve volumes in the Crime Prevention Studies series, each composed of 15 or so articles (Clarke 1993, 1994a, 1994b, 1996; Eck and Weisburd 1995; Homel 1996, 1997; Weisburd and McEwen 1998; Green Mazerolle and Roehl 1998; Painter and Tilley 1999; Natarajan and Hough 2000; Farrell and Pease 2001), and two editions of the Situational Crime Prevention volume (Clarke 1992, 1997b). Searching for Good Practices in Other Police Agencies Other sources of relevant information, like written reports of problem-solving initiatives, are less accessible. Some of the annual submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing are in a computer database called POPNet, operated by the Police Executive Research Forum and accessible through the Internet. These are summaries only, however, and do not convey the complete understanding of the projects that might come from reading a full narrative report. Moreover, the entries are not subjected to much quality control, so their value is further limited.75 Jane's Information Group, the British company best known for publishing information about military hardware, is developing a counterpart to POPNet. Known as COPcase, this web-based system (copcase.janes.com) provides detailed reports on effective police problem-solving initiatives. While this system is still in its infancy, it shows considerable promise. The Police Executive Research Forum published reports on the award-winning projects from 1998 (Solé Brito and Allan 1999) and, with support from NIJ and the COPS Office, have published reports on the award-winning projects from 1999 (National Institute of Justice 2000). 75Since POPNet's inception in November 1996, through February 2000, 84 police agencies have submitted a total of 107 cases in 41 problem- type categories. The most frequent problem types are drugs (12), juveniles (nine), gangs (eight), and community decline (seven). Frequent contributors have been the San Diego Police Department (13) and the Edmonton Police Service (eight). Many of the entries are summaries from the Herman Goldstein Award for Excellence in Problem-Oriented Policing program. The brief narratives make it difficult to get a good understanding of the projects. Some of the problem-type categories are not substantive–e.g., "lack of communication" and "community dissatisfaction." A few are redundant–"burglary" and "breaking and entering," and "traffic" and "traffic safety." A few have no entries, including "theft" and "graffiti," both common problems. Some are overly broad (e.g., "calls for service," "juveniles," "disturbance," and "public safety"). Searching for Good Practices Within a Police Agency The police can also improve their responses to community problems by studying their own and other agency's past efforts to address similar problems. Reports about problem-solving initiatives are a valuable source of knowledge from which to draw, even if those initiatives did not apply rigorous research methods. Unfortunately, most police agencies do not routinely prepare detailed reports on most of their problem-solving initiatives. Some police managers are reluctant to impose what might be perceived as excessive reporting requirements on officers whom they do not want to discourage from engaging in problem-solving. While this is understandable as managers try to coax officers into policing in a different way, a lot of knowledge about how various problems have been handled has been lost. Some police agencies have created internal computerized databases to store information about problem-solving projects.76 Some agencies also maintain paper files for each project and have an official numbering system to keep track of those files. To ensure at least minimal documentation, some agencies require officers to complete problem-solving project forms. Most forms capture basic data about the problem's nature, the analysis and responses, and some impact measures. A few agencies have chronicled problem-solving projects through in-house newsletters.77 These computer records, project reports, forms and newsletters have great potential to help officers search for solutions to common problems and to teach officers problem-solving skills through real examples. The best of these efforts are commendable, but still quite modest and limited. Somehow, more police-led problem-solving efforts must be documented in writing and police managers must then make these resources accessible and encourage that they be reviewed as a standard step in future problem analysis.78 76Most often, however, these databases are not linked to the department's main computer network, and typically, one person maintains both the databases and the project reports, so they are not widely accessible. The Edmonton Police Service and the Savannah Police Department maintained their POP project files in database programs that proved difficult to use, and thus became neglected. I found 74 POP projects in the centralized database of the Edmonton, Alberta, Police Service; 180 in one division of the Lancashire, England, Constabulary; 135 in the Savannah, Ga., Police Department; 158 in the Reno Police Department; and 74 for 1999 in one patrol sector of the Sacramento, Calif., Police Department. Different agencies have different policies about documenting POP projects, and place varying levels of priority on entering information into centralized databases. Thus, computerized POP project files are a crude way of gauging the amount of problem-solving occurring in an agency. Certainly, compared with the enormous investment and commitment to documenting and storing data on police calls for service, incident reports and criminal investigations, the state of record-keeping for problem-oriented activities is primitive. (In the main, police records are used to either establish the elements of crimes for possible prosecution or account for police officers' time and actions; they are not typically designed or used as a source of information for addressing problems.) In my visits to some agencies, it was not uncommon to find that if a record system for problem-oriented policing projects existed, few people had the knowledge or capacity to access it. Some departments' files had fallen into disuse after a few years, and consequently contained little current data. In no agencies did I find that the department's central records unit, responsible for maintaining most other official records, had any responsibility for maintaining problem-oriented policing project files. Ultimately, police agencies must assign the same degree of importance to the official records related to problem-oriented initiatives as they do other official records. What Does It Mean To Develop an Understanding of the Multiple and Competing Interests at Stake in Problems? Many problem-oriented policing initiatives fail to take complete account of all the interests at stake with respect to the problem. This matter of accounting for the various interests is often simplified into a mere inventory of stakeholders. By interests in a problem, Goldstein meant the various reasons why the police or the community either is, or is not, concerned about a particular problem. Thus, any one stakeholder is not limited to having a single interest in the problem. Indeed, most stakeholders have multiple and competing interests in a problem.79 77The St. Louis Metropolitan Police Department; St. Petersburg, Fla., Police Department; Fresno, Calif., Police Department; Reno Police Department; and Stockholm, Sweden, Police Department at one time or another have produced regular newsletters exclusively dedicated to chronicling in-house problem-solving efforts. 78Herman Goldstein recounted an experience while working with the Charlotte-Mecklenburg, N.C., Police Department that illustrates the difficulty in sharing knowledge within the police field. No fewer than five different Charlotte- Mecklenburg police officers were addressing problems related to motels at the same time. They learned of one another's efforts only through Goldstein. Moreover, they were unaware that the department's Crime Prevention Unit maintained a file on the prevention of similar problems, and unaware that the Portland, Ore., Police Bureau had produced a comprehensive manual on police responses to motel problems, a manual that had been published for national distribution. One enterprising officer eventually researched the topic via computer access to the National Criminal Justice Reference Service. 79There is a distinction between the collective interests that all members of society presumably share, and the more narrow interests of particular individuals or groups in advancing a particular agenda. For example, society and government have broad collective moral, health, commercial, and public safety interests in controlling prostitution. Within the context of a specific prostitution problem, the interests might be narrower–a motel has an interest in the revenue prostitution generates, but a competing concern about becoming a haven for crime, while a neighboring church has an interest in keeping prostitutes away. The way Goldstein envisions the question of interests' being explored derives from other legal writings about government interests.80 One begins by asking what the social interests are in the problem (both the interests in curtailing the behavior and the interests in permitting the behavior), and then asking what the government interests are in the problem. Not all social interests should be government interests. Once one identifies the government interests, one can turn to asking what police interests are at stake. Referring to the police's fundamental objectives is a useful way to approach this question (see chapter 2 for further discussion of fundamental police objectives). If the police conclude they have no interest at stake in the problem, there is little justification for their continued involvement regarding it. There are many social problems in which the police are well-advised not to become embroiled. 80See Remington (1952, 1954). In exploring the various nonpolice interests at stake in a problem, it is important to go beyond the most visible and obvious interests. There are often hidden commercial interests involved in many problems, as well as latent social prejudices and biases. These interests should at least be brought out in the open, where they can be considered. The careful probing of these interests is among the most enlightening parts of the problem-solving process. Police officers who engage in this probing of interests begin to appreciate just how many different perspectives there may be regarding the same problem. The multiple and competing interests of the police themselves are often not well- considered. For example, some conventional responses to chronic problems, however ineffective, promote some police interests. In some jurisdictions, police officers rely heavily on either overtime or outside security employment for their incomes. Sometimes an alternative response to a problem has the potential to eliminate the need for the overtime or off-duty assignments, obviously presenting ethical challenges to the police. What Does It Mean To Take Inventory of and Critique the Current Responses to Problems? Another aspect of analysis commonly omitted that Goldstein considers crucial is an inventory and assessment of the current responses to the problem being studied. Many project reports allude only briefly to the inadequacy of current responses, mainly by making the obvious assertion that a new response is needed. Current responses are often described briefly and generally, and casually discredited as being ineffective. One often reads in problem-solving project reports cursory assessments of current practices such as "the traditional response of handling calls, taking reports and making arrests was not working". But brief and general descriptions like these are not illuminating and, often, not entirely accurate. Individual police officers frequently develop their own innovative responses to problems, responses that are not fully and accurately encompassed in their agency's standard operating procedures. Other agencies and groups may be responding to problems in ways that the police are unaware. Some responses, however traditional, may prove more effective upon closer analysis than they might initially appear. For example, many problem-solving initiatives indicate that conventional strategies of arresting offenders have failed. Yet few such initiatives support such assertions by addressing how criminal arrests are actually processed, and what ultimately happens to offenders. These sorts of court-tracking studies that follow criminal arrests through the criminal justice process to assess the outcomes can be illuminating.81 As a routine matter, police agencies typically receive little or no information from the courts about case disposition. Admittedly, court-tracking studies can be somewhat difficult to do (the data are often unreliable and difficult for police to access), and some police agencies don't consider anything that happens after police processing to be of their concern. When the police limit their inquiry to their own arrest actions, they miss a full understanding of the systemic responses to a problem. It takes some effort to discern precisely how problems are being handled and to what extent current practice is effective. 81The early Goldstein and Susmilch (1982a) problem-oriented study of the drinking driver in Madison concluded that, contrary to most police officers' belief, a driver's refusal to take a breath test almost ensured a conviction for the primary charge of drunken driving. This was so because of plea-bargaining practices. Police officers had become conditioned to believe that plea bargaining was contrary to their interests, when under those circumstances, the bargaining practices were entirely consistent with them. The flip side of dismissing the value of current conventional responses, when faced with a problem that is not getting adequate attention, is simply increasing the effort put into conventional responses, without carefully considering their strategic value. Many reports on problem-solving projects leap quickly to judgments that greater police presence, more arrests, more certain prosecution, or stiffer penalties are the best response to a problem. Such judgments are often made without examining the effectiveness of existing levels of these interventions. How Should the Police Develop and Implement New Responses to Problems? Expanding the Range of Response Alternatives Goldstein urges the police to greatly expand their range of alternative responses to problems, responses beyond the conventional increased police presence and criminal arrests. This is perhaps the aspect of problem-oriented policing that thus far the police have most successfully applied.82 A wide range of responses is emerging from reports of problem-oriented policing projects. It may be that the police have long tried many of these responses, but have informally and seldom acknowledged so openly. That doesn't diminish their significance. The concept of problem-oriented policing has successfully brought these new responses into professional and public discourse about policing. 82Deputy Chief Pat McElderry of the Colorado Springs Police Department aptly summarized the police tendency to emphasize responses to problems over the identification, analysis or assessment of problems when he wrote, "[O]ur acculturation in traditional policing still makes it difficult to see 'SARA' as anything other than 'saRa'" (1999). Goldstein intended that new responses to chronic problems be well- considered, following logically from careful problem analysis; that they not be merely a few clever ideas thought up as a hasty reaction. Clever ideas have some value, but without a clear line of reasoning that articulates the basis for the new response, they do not add much to the body of professional knowledge from which other police agencies and communities can draw. Police agencies often copy other agencies' clever or innovative ideas. But, without first assessing how they might work in the local situation, these ideas might well prove ineffective. It is also unfortunate when the police launch problem-solving initiatives with a preferred response in mind. The subsequent problem analysis serves more to justify the preferred response than to inform the decision-maker about the nature of the problem. Most training programs in problem-oriented policing incorporate some structured methods for generating ideas about possible new responses. Brainstorming is the most common. Yet the actual practice of problem-oriented policing does not apply as much structured group decision-making as the training would suggest occurs. Response strategies are more often the product of an ad hoc process involving only a single or a few key decision-makers. When this occurs, there is greater risk that the decision-makers' personal biases will dictate the response strategy. The most common response-related bias is toward using criminal arrest as the primary response. Ideally, the criminal arrest response is considered neutrally, as one possibility among many for addressing a problem. “When a police department takes a problem-oriented policing approach, it turns police work upside down by asking whether the current response is working. It calls for a constant reexamination of what we do, including our relationship with the community.” – Rana Sampson Categorizing Response Alternatives In the literature on problem-oriented policing, there are now several frameworks for considering response alternatives. Goldstein approached the issue by chronicling and then categorizing a wide range of response alternatives that police agencies have actually adopted. He devoted Chapter 8 of Problem-Oriented Policing mainly to describing these categories, their rationale and examples (1990a). His classification scheme is a descriptive one (see Table 3 on the next page). Table 3 Goldstein's Categories of Responses in Problem-Oriented Policing 1. Concentrating Attention on Those Individuals Who Account for a Disproportionate Share of a Problem 2. Connecting With Other Government and Private Services a. Making Referrals to Other Agencies b. Coordinating Police Responses With Other Agencies c. Correcting Inadequacies in Municipal Services, and Pressing for New Services 3. Using Mediation and Negotiation Skills 4. Conveying Information a. To reduce anxiety and fear b. To enable citizens to solve their own problems c. To elicit conformity with laws and regulations that are not known or understood d. To warn potential victims about their vulnerability, and advise them of ways to protect themselves e. To demonstrate to people how they unwittingly contribute to problems f. To develop support for addressing a problem g. To acquaint the community with the limitations on the police, and to define realistically what they can expect of the police 5. Mobilizing the Community 6. Using Existing Forms of Social Control, in Addition to the Community 7. Altering the Physical Environment to Reduce Opportunities for Problems to Recur 8. Increasing Regulation, Through Statutes or Ordinances, of Conditions That Contribute to Problems 9. Developing New Forms of Limited Authority To Intervene and Detain 10. Using the Criminal Justice System More Discriminately a. Straightforward Investigation, Arrest and Prosecution b. Selective Enforcement, With Articulated Criteria c. Enforcement of Criminal Laws That, by Tradition, Another Agency Enforces d. Definition, With Greater Specificity, of That Behavior That Should Be Subject to Criminal Prosecution or to Control Through City Ordinances e. Intervention Without Making an Arrest f. Use of Arrest Without the Intention to Prosecute g. Attachment of New Conditions to Probation or Parole 11. Using Civil Law to Control Public Nuisances, Offensive Behavior and Conditions Contributing to Crime Source: H. Goldstein. 1990. Problem-Oriented Policing. Philadelphia: Temple University Press. Ron Clarke (1997b) has developed a different sort of framework–"the 16 techniques of situational crime prevention.” He derived the categories from the theoretical bases of situational crime prevention and routine activity theory. Clarke also provides examples of each response category, which I have reproduced with minor stylistic changes in Table 4 below. Table 4 16 Opportunity-Reducing Techniques Increasing Perceived Effort 1. Target Hardening Slug Rejector Devices Steering Locks Bandit Screens 2. Access Control Parking Lot Barriers Fenced Yards Entry Phones 3. Offender Deflection Bus Stop Placement Tavern Location Street Closure 4. Facilitator Control Credit Card Photo Gun Control Caller ID Increasing Perceived Risks 5. Entry/Exit Screening Automatic Ticket Gates Baggage Screening Merchandise Tags 6. Formal Surveillance Red-Light Cameras Burglar Alarms Security Guards 7. Employee Surveillance Pay Phone Locations Park Attendants CCTV Systems 8. Natural Surveillance Defensible Space Street Lighting Cab Driver ID Reducing Anticipated Rewards 9. Target Removal Removable Car Radios Women's Shelters Phone Cards 10. Property Identification Property Marking Vehicle Licensing Cattle Branding 11. Temptation Reduction Gender-Neutral Listings Off-Street Parking Rapid Repair 12. Benefit Denial Ink Merchandise Tags Car Radio PINs Graffiti Cleaning Removing Excuses 13. Rule Setting Customs Declaration Harassment Codes Hotel Registration 14. Conscience Stimulation Roadside Speedometers "Shoplifting Is Stealing" "Idiots Drink and Drive" 15. Disinhibitor Control Drinking-Age Laws Ignition Interlocks V-Chips 16. Compliance Facilitation Easy Library Checkout Public Lavatories Trash Bins Source: R.V. Clarke. 1997. Situational Crime Prevention: Successful Case Studies (2nd ed.). Albany, N.Y.: Harrow and Heston. John Eck's problem-solving manual, a work still in progress, offers yet another framework that, like Clarke's, is built around the theories underlying situational crime prevention (see Table 5). The “Problem-Solver” computer program codeveloped by Lorraine Green Mazerolle also includes menus of suggested response alternatives derived from the existing literature on problem-oriented policing. Whether one uses deductive frameworks like Clarke's or Eck's, in which responses are classified according to theoretical premises, or inductive frameworks like Goldstein's, in which responses are classified according to common features of tested responses, these frameworks have the potential to expand the range of responses the police and others use to address community problems. As with problem analysis guides, problem response guides should not be considered comprehensive sets of solutions to problems. What Does It Mean for the Police to Be Proactive? In problem-oriented policing, proactive responses are preferred over merely reactive responses. Reactive responses may be entirely appropriate as a temporary measure to stabilize a problem and to serve other legitimate police objectives. Goldstein stresses proactivity in policing in two senses. First, he asserts that responses to problems should prevent future harm, and not just address past harm. Second, he believes the police should see it as a legitimate part of their role in government and society to speak out about public safety problems that are not being adequately addressed. This relates to the earlier discussion about what sorts of problems the police should address as part of their mandate. Speaking out about problems might simply require calling them to the attention of other officials or community leaders. Beyond that, it might require that the police appeal to others' moral and ethical obligations to take responsibility for problems, or invoke legal authority to compel others to do so. The questions of when and how the police should assign responsibility have not been adequately addressed. Assuming that the police have an obligation to speak out about problems they are aware of, but others may not be aware, in what forum should they speak? Should they speak out when elected officials have consciously chosen not to? What pressure should the police put on uncooperative segments of the community? Goldstein (1996c) has articulated a continuum of pressure the police might apply to get other entities to assume or share ownership for community problems. The degree of pressure the police apply should depend on the strength of the evidence they have regarding the nature of the problem and its causes. Table 6 summarizes this continuum, with each step involving more pressure. Table 5 Eck's Categories of Response Alternatives I. Offender Control Tactics Offenders Offenders' Tools Handlers Handlers' Tools II. Target and Victim Control Tactics Target and Victim Protection Target and Victim Tools Guardianship Guardianship Tools III. Place Control Tactics Places Managers Management Tools Streets and Routes Source: J. Eck. 1998. “A Problem-Solving Manual for the Police and the Public.” Unpublished draft. “Cities need to develop an understanding of, and an explicitness about, the responsibilities of businesses that consume a disproportionate volume of police services–like convenience stores and shopping malls–social responsibilities of companies not to create crime opportunities.” – Rana Sampson Table 6 Goldstein's Continuum of Police Pressure on Others to Accept Ownership of Community Problems (from least degree of pressure to greatest) • develop educational programs regarding responsibility for the problem • make a straightforward informal request of some entity to assume responsibility for the problem • make a targeted confrontational request of some entity to assume responsibility for the problem • engage another existing organization that has the capacity to help address the problem • press for the creation of a new organization to assume ownership of the problem • shame the delinquent entity by calling public attention to its failure to assume responsibility for the problem • withdraw police services relating to certain aspects of the problem • charge fees for police services related to the problem • press for legislation mandating that entities take measures to prevent the problem • bring a civil action to compel entities to accept responsibility for the problem. “Herman Goldstein's original concept was more centralized. I don't think he saw police officers making direct contributions to problem- solving; he saw them more as resources to analysts and researchers. Newport News and Baltimore County showed that police officers can do problem-solving themselves. I know Herman has had mixed feelings about this.” – Darrel Stephens The notion of police proactivity has led to some confusion and abuse. It raises questions about the propriety of police involvement in certain social issues, as discussed earlier. There is also some confusion about what, precisely, constitutes proactive measures vs. reactive measures. The answer depends somewhat on one's frame of reference. For example, with respect to repeat victimization, police measures to assist a repeat victim are reactive with respect to the first victimization, but may be proactive with respect to future victimizations. The notion of proactivity has also been misunderstood and abused when it has been invoked as a code word for aggressive police tactics. Who Should Be Involved in Problem-Oriented Policing, and How? While Goldstein originally encouraged line officers' involvement in problem-oriented policing, he did not anticipate that they would emerge as the leaders in addressing problems. Goldstein originally imagined that command-level police officials and research collaborators would lead most problem-oriented initiatives; that they would be, in essence, research efforts like the early Madison studies of drinking drivers and repeat sex offenders. As originally conceived and tested, problem-oriented policing focused on the police's administrative and investigative operations, more so than on routine patrol operations. Few of the systems, supports and expectations for patrol-level problem-solving that are now widely recognized as critical to problem-oriented policing were part of Goldstein's early vision. The later research done in Baltimore County, Newport News, Va., and other places better defined a role for line officers and made problem- solving part of the daily routine. Problem-oriented policing has seen leadership on projects come from many levels in the police hierarchy. Line police officers have emerged as the leaders of many projects, even when the scope of the project has been quite large. In my analysis of submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing for 1993 to 1999 (see Appendix A), I concluded that about half of the projects were led by one or a couple of line-level police officers.83 In one respect, this provides some evidence of the talent line-level police officers have, talent that police managers do not fully appreciate or exploit. About one-third of the submissions analyzed reported active leadership on the project from a police supervisor.84 Only 15 percent of the submissions cited command-level officers for active project leadership. The results of this analysis are consistent with my own experiences in several police agencies. While this degree of line-level leadership certainly advances Goldstein's intent that line officers be an important part of the decision-making process, it leaves open to question whether line-level officers should be expected to provide their own leadership. One possible explanation for this trend toward line-level leadership is that supervisory and command-level officers are simply not sufficiently engaged in practicing problem-oriented policing. To some extent, the New York City Police Department's experiences during Commissioner William Bratton's tenure showed the weaknesses of exclusively line-level leadership. Bratton believed that the department's community policing and problem-solving efforts in prior administrations depended far too heavily on community police officers to identify and address community problems, without sufficient involvement of precinct commanders (Bratton 1995). The Compstat process was one means by which command-level officers were compelled to become more intimately engaged in resolving community problems. Critics of this approach argue that when commanders are held accountable for problem-solving, problems tend to get defined in their terms, and less so in the community's and the line officers' terms; the community and line officers are likely to be more familiar with the problems than the commanders. Moreover, the real leadership capacity of many line officers may be overlooked, and they may be discouraged from engaging in problem-solving's analytical aspects. 83Determining precisely who led any particular project is sometimes subjective and depends on the project report author's account. Some authors discount potentially significant higher- level leadership to deflect credit to the line officers; others do the opposite. Of the 100 submissions I analyzed, 48 indicated that line- level officers led the project. Of those 48, about one-fourth concerned localized problems, one- half concerned intermediate-level problems, and one-fourth concerned communitywide problems. 84This figure itself is generously high insofar as I attributed supervisory leadership to all projects in which an entire unit of police officers was credited for leading the project. I assumed that if an entire unit was involved, the unit's supervisor likely provided at least some leadership. From experience, I know this assumption is not always accurate. What is the ideal level of police authority for providing leadership in problem-oriented policing projects? The answer, like the answer to so many questions related to problem-oriented policing, is that it depends principally on the scope of the problem being addressed. As a general proposition, supervisors should provide active leadership in localized beat problems; commanders in intermediate-level problems; and top commanders, perhaps including the chief executive, in communitywide problems. In every instance, line officers should be encouraged to be as involved as their time and abilities permit. There is always a need for higher levels of authority to become involved in a particular project, as the situation dictates (e.g., if a lower-level official cannot get cooperation). That higher-ranking police officials seldom actively lead problem- oriented policing initiatives suggests that the problem-solving method of operations has yet to achieve a high level of importance in most police organizations. It tends still to be viewed as something that only beat police officers do. Police chiefs need to pay at least as much personal attention to substantive community problems as they do to administrative and political concerns.85 Some command officers, to the extent they are supportive of problem-oriented policing, see their role as administrative manager, ensuring that systems are in place and resources available for line-level problem-solving. This is fine as far as it goes, but without more personal and direct command-level leadership, few large and complex community problems are likely to be taken on in a sophisticated, problem-oriented way. Line-level officers simply lack the requisite resources in most instances to conduct the sort of analysis and effect the sort of responses necessary to bring about substantial improvements in communitywide problems. From my observations, San Diego's former police chief, Jerry Sanders, widely seen as the champion of problem-oriented policing within his agency, provided an optimal style of leadership with respect to problem-oriented policing projects. Sanders made a habit of attending the department's periodic Problem Analysis Advisory Committee meetings. His attendance demonstrated both his interest in problem- oriented policing generally, and in the line officers' projects particularly, as well as his commitment to become personally involved in a project if circumstances warranted. Through his consistent expressions of interest in community problems, and his consistent availability to intervene in projects, he allowed each problem to receive the appropriate level of leadership. The informal operating principle was that the lowest level of leadership necessary to effectively address a problem was the optimum. Accordingly, one often found at least mid-level police managers personally engaged in large problem-solving projects, and only indirectly involved in smaller, localized problems. 85My own experiences working in the administration of the New York City; Fort Pierce, Fla.; and St. Louis Metropolitan police departments confirm this observation. I typically found that top-level staff meetings were almost entirely consumed by discussions of administrative matters. “At its best, problem-oriented policing engages police officers at the front end, and gets them excited about their work. It gives them a whole new perspective on their job, such that the job can become exciting instead of routine, and that's important.” – Gloria Laycock It makes no sense to dictate, as a matter of policy, that a certain rank police official become personally involved in problem-solving initiatives anymore than it makes sense to dictate this in the handling of incidents or the investigation of crimes. The goal is to get an entire police agency thinking in problem-oriented terms, not merely to have everyone simultaneously working on projects. This question of active leadership should be resolved pragmatically, depending on the particulars of each problem. Given the abundance of community problems in every jurisdiction of sufficiently large scope, supervisors and command-level officers have plenty of opportunities to become personally engaged in problem-oriented policing, and they need to do so for problem-oriented policing to advance and become institutionalized practice. How Should the Effectiveness of Implemented Responses Be Evaluated? The aspect of problem-oriented policing Goldstein has written least about is the measurement of effectiveness. The early experimental projects he conducted in Madison were not fully implemented, so there was little opportunity to evaluate the impact of the proposed response strategies. Goldstein has always asserted that measuring effectiveness is crucial to the process. Without some measurement of impact, the police can learn little about the value of different responses. Several major issues and debates have arisen with respect to this aspect of problem-oriented policing. Process vs. Outcome Measurement Perhaps the single greatest source of confusion relating to the evaluation of problem-oriented policing initiatives surrounds the distinction between the measurement of processes and the measurement of outcomes. The measurement of processes is the documentation of the actions taken in implementing responses, and an assessment of whether the responses were actually implemented as intended. The measurement of outcomes is the assessment of the ultimate impact the responses had on the problem, as defined (i.e., Did the problem improve, worsen or remain the same? Were the outcome objectives achieved?). In many problem-oriented policing projects, these two different types of evaluation are confused. Most commonly, evaluators misconstrue process evaluation for outcome evaluation; that is, they limit their inquiry to determining how well and to what degree the police and others actually implemented their plan of action. While this information is vitally important, it cannot be substituted for some inquiry about what effect the plan of action, however well- implemented, had on the problem. Trained researchers have also fallen into this conceptual trap. The evaluation design of a problem-oriented policing initiative funded by the COPS Office Problem-Solving Partnerships Program provides a common example of this confusion. The problem being addressed was auto theft. A response strategy was developed that focused on the apprehension, prosecution and punishment of juvenile offenders. The written evaluation design, prepared by an outside trained researcher, listed the following outcome measures for the project: • Increase the number of people arrested for auto theft. • Increase the number of juveniles arrested for auto theft. • Focus particular attention on repeat offenders. • Increase the number of cases filed for prosecution. • Increase the number of juvenile cases filed for prosecution. • Increase the conviction rate of auto theft offenders. • Increase the conviction rate of juvenile auto theft offenders. • Increase the punishment for convicted offenders, including sentence lengths. • Increase the actual time of incarceration. None of these outcome measures would reveal anything about the number of auto thefts committed after the police implemented the response strategy, an indicator that logically should be the primary outcome objective. “The analysis of problems by police officers in some projects is impressive, and their responses, creative. Herman Goldstein recognized how creative officers are. Problem- oriented policing allows for that creativity; it is no longer just something that is exercised when the sergeant isn't looking.” – Rana Sampson “Problem-solving seems to happen more naturally at both the bottom and top levels of police organizations. But it's not satisfactory that it happen at only these levels, since most problems we care about are intermediate-size problems, calling for intermediate-level responses, organized and coordinated within the middle layers of police organizations.” – Malcolm Sparrow One possible explanation for the persistent confusion over process and outcome measurement may lie in confusion about the fundamental police objectives (discussed more fully in chapter 2). If one believes the police's primary objective is to enforce the law through apprehension and criminal prosecution, then one can logically understand measures of the sort listed above to be outcome measures. If, on the other hand, one believes that the police's primary objective is to reduce the incidence and seriousness of harm to the community, and that enforcing the law is but a means to that end, then the measures listed above are clearly only process measures, and not outcome measures. Goldstein, of course, holds the latter view, and evaluation designs that are limited to measuring arrests and other process indicators represent a serious distortion in the practice of problem-oriented policing. Ideally, a problem-oriented policing project will include measurement of both processes and outcomes. What Standards of Proof Should Apply in Evaluating Effectiveness? Another major issue relating to evaluation in problem-oriented policing surrounds the evaluation methodology and standards of proof to be applied. This issue presents itself in much the same way the problem analysis issue does. The questions in evaluation are: What are legitimate methods to determine the effectiveness of responses? and, How certain must we be to legitimately claim success? On this matter, Goldstein and other academics disagree. Goldstein advocates a certain degree of flexibility. He acknowledges the many difficulties in establishing precise and certain conclusions in the complex world of human behavior where policing occurs, and accordingly, he is willing to settle for less than the most rigorous tests of effectiveness in most instances.86 In addition, the evaluation of police interventions is not yet sophisticated, so insisting on rigorous standards, however justified theoretically, would likely stifle much experimentation with the problem-oriented concept. As with problem analysis, Goldstein concurs with Ron Clarke's belief in “good enough” measurement. How precise and certain one has to be in problem-oriented policing depends greatly on the consequences of being wrong. The main critic of this brand of eclectic, flexible evaluation has been Lawrence Sherman. Sherman, who endorses the problem-oriented approach to policing generally, has advocated that outcome evaluation entail what he considers the most reliable methodology–controlled experiments–and the most demanding standard of proof–the elimination of all rival hypotheses.87 Sherman is less willing than Goldstein or Clarke to accept the validity of claims based on less than rigorous and controlled evaluations. Few problem-oriented policing projects reported to date have employed anywhere near the level of rigor and control that Sherman considers ideal. Thus, how much the practice of problem-oriented policing has advanced police knowledge about how to reduce crime, disorder and fear depends heavily on one's views about evaluation. Those who share Goldstein's and Clarke's views might conclude that a lot has been learned;88 those who share Sherman's view might conclude that little has been learned (see Sherman et al. 1997).89 Clarke shares some of Sherman's concerns about the inadequacy of evaluation in many problem-oriented policing projects, concluding that the studies conducted under the rubric of situational crime prevention have, on the whole, been more rigorous and reliable than those conducted under the rubric of problem- oriented policing. Most of the work conducted in situational crime prevention, at least the evaluation component, has been led by trained researchers. Most of the work conducted in problem-oriented policing has been led by police practitioners. That there is better evaluation in the situational crime prevention context is therefore not surprising. Yet, in the end, the knowledge gained from the work matters more than the rubric under which the work is done. Goldstein sees the situational crime prevention work as extraordinarily valuable to the police, not so much because he feels they can emulate the research methods, but because they can learn important lessons about the effectiveness of different responses to common problems. 86There may be inherent tension in the concept of problem-oriented policing between the principles related to effective responses and those related to evaluation. Problem-oriented policing encourages the police to develop multifaceted response strategies to maximize the likelihood of success (and, in fact, most problem-oriented policing projects entail the use of multifaceted strategies). Multifaceted response strategies, however, are considerably more difficult to evaluate than single response strategies because it is difficult to isolate each response's effects (Eck 1997). 87Sherman himself was criticized by other scholars for failing to set sufficiently high standards of replicability of results in his domestic violence experimental studies, a charge Sherman defended against, arguing that policymakers must make decisions on the best evidence available, however imperfect (Mastrofski and Uchida 1993). 88Among the other scholars who also endorse less than the strictest evaluation methodology are Bazemore and Cole (1994). They wrote: “Police departments, as they move increasingly toward the community policing model and problem-oriented strategies, will need to assume increased initiative for monitoring and assessing initial implementation of these approaches, as well as evaluation of intermediate impacts… While these evaluations and assessments may not always meet the highest methodological standards for purposes of causal inference, viewed as case studies of strategic interventions and their intermediate impacts on crime and citizen attitudes, local community policing experiments can be expected to add significantly to practical knowledge and theory development” (p. 121). 89See, also, Brame and Piquero (1998), who wrote: “In sum, the evidence on effectiveness of problem-solving strategies seems to have an optimistic tone. Unfortunately, much of it is anecdotal and not scientifically rigorous… Thus, a critical issue in the area of problem-solving is the need for more rigorous research designs and multiple-site studies to evaluate the effectiveness of various measures.” What Are the Specific Objectives of Problem-Solving Efforts? Another issue related to the evaluation of problem-oriented policing concerns the articulation of problem-specific objectives. The Newport News study (Eck and Spelman 1987) first delineated a set of generic legitimate objectives in problem-solving. It grouped those objectives into five categories: 1. totally eliminate a problem; 2. substantially reduce a problem; 3. reduce the harm created by a problem; 4. deal with a problem better (e.g., treat people more humanely, reduce costs or increase effectiveness); and 5. remove the problem from police consideration. The fifth objective, removing problems from police consideration, differs from the first four in that it does not directly address the question of whether the problem, as experienced in the community, will be improved by removing it from police consideration. Taken to the extreme, the police could claim success in problem-oriented policing merely by working to absolve themselves of responsibility for problems. Goldstein did not intend such an outcome; nor did Eck and Spelman. If shifting responsibility for addressing a problem to another entity results in more effective handling of the problem, then the objective is legitimate. If such a shift results merely in some efficiency gains for the police, then it may have some merit, but one cannot consider it an effective resolution. When proper outcome evaluations of problem-oriented policing initiatives are conducted, some prove too limited in their scope; that is, they are limited to measuring only a few indicators of impact, most often the volume of calls for service or the numbers of reported crimes. Often neglected in evaluations are indicators of the prevalence of the problem, the net harm caused by the problem, the possible displacement of the problem, the possible diffusion of response-strategy benefits, and an accounting of the total costs arising out of the problem and responses to it.90 Looking at the prevalence of a problem in addition to the incidence of the problem is interesting because it reveals how widely or narrowly the entire community experiences the harms caused by the problem. An initiative might succeed in reducing the overall incidence of a particular problem within the jurisdiction, but if the problem consequently becomes concentrated in one particular neighborhood, this result may not be desirable. A review of the submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing indicates that police problem-solvers are increasingly recognizing the issue of problem displacement. This is largely due to more specific award submission criteria regarding displacement. Overall, the reports on problem-solving projects far from adequately address displacement, but at least it is becoming recognized as a phenomenon worthy of inquiry..91 Too few problem- oriented policing initiatives entail any real economic assessment of a problem. Economic analyses should not be seen as definitive of success or failure, but they add an important dimension to judging an effort's overall quality. The police, particularly in the United Kingdom, are increasingly being asked to account for the cost-effectiveness of policing strategies (Stockdale, Whitehead and Gresham 1999). 90Tilley (1999) lists the following typical shortcomings in the evaluation of crime prevention measures: “the use of simple before/after comparisons, use of short and arbitrary before-and-after periods, neglect of benchmark statistics, failure to consider possible other (nonintervention) causes of the changes observed, failure to test for displacement or diffusion of benefits, and uncritical promotion of success stories.” 91For a discussion of the displacement of crime, and its opposite phenomenon, the diffusion of benefits, see Clarke (1997b:28-33). Chapter 2 Putting Problem-Oriented Policing and Problem-Solving in the Context of the Whole Police Mission How Does Problem-Solving Fit in With Other Aspects of Police Work? Goldstein's assertion that problem-oriented policing affects virtually everything the police do, and how police agencies are run, can be confusing. Is he saying that the police should discontinue all their conventional methods of operation and engage exclusively in problem-solving processes? Where does this leave the conventional tasks and methods for responding to calls for service or investigating crimes? Police administrators who endorse problem-oriented policing have sought to reconcile the demands on their agencies to continue performing these conventional police tasks with the new demands to engage in substantive problem-solving. Lingering conceptual confusion about how problem-solving is supposed to fit into the context of the entire police mission may account for why the police have not fully integrated the old, unavoidable tasks and methods with the new tasks and methods. One may clear up the conceptual confusion by returning to some first principles of policing. Goldstein's writings on problem-oriented policing are best understood in the context of his earlier writings about the police's role in society (1977). In those writings, Goldstein argued that to understand policing properly, one has to distinguish between the objectives the police are trying to achieve and the methods they use to achieve them. Accordingly, he has argued that investigating crimes and enforcing laws, long thought of as basic policing objectives, are not objectives in and of themselves, but rather methods for achieving other, more broadly stated, objectives. Problem-oriented policing, then, is concerned with expanding on and improving the methods the police use to achieve their more fundamental objectives. “We need problems to become the basic units of work in policing and other city services, and to make that idea real.” – Dennis Nowicki What Are the Fundamental Objectives of Policing? The fundamental objectives of policing (also referred to as the mission of the police or the core functions of policing) are the ultimate purposes for which police agencies have been created. Goldstein was one of a number of scholars who recognized and articulated the breadth and complexity of the police mission. He synthesized his understanding of the multiple objectives of the police in his seminal work, Policing a Free Society, a precursor to his writings on problem-oriented policing. Drawing from even earlier work he had done, Goldstein (1977) characterized the fundamental objectives of the police in free societies as follows: 1. to prevent and control conduct threatening to life and property (including serious crime); 2. to aid crime victims and protect people in danger of physical harm; 3. to protect constitutional guarantees, such as the right to free speech and assembly; 4. to facilitate the movement of people and vehicles; 5. to assist those who cannot care for themselves, including the intoxicated, the addicted, the mentally ill, the physically disabled, the elderly, and the young; 6. to resolve conflict between individuals, between groups, or between citizens and their government; 7. to identify problems that have the potential for becoming more serious for individuals, the police or the government; and 8. to create and maintain a feeling of security in the community. While there are other ways to characterize the police mission, both in greater and lesser detail, Goldstein's formulation remains a comprehensive and useful reference for guiding police actions. Some police agencies have other specialized functions, but most have these basic ones in common. The ultimate aim of problem-oriented policing is to continually make the police better at accomplishing each of the above objectives to better prevent crime, to better assist victims, to make communities feel safer, and so forth. Everything the police do, whether using conventional or innovative methods, should be in pursuit of one or more of these fundamental objectives. “Reactive policing is so much easier. Police officers are trained to prefer order to disorder, and problem-solving seems, to some officers, to be creating disorder, to be upsetting the balance of things.” – Dan Reynolds Properly understood, this broad, though not limitless, set of objectives should be liberating for the police. Theoretically, at least, it frees the police from being bound to certain methods of achieving these objectives, allowing them to develop other methods that might prove more effective. In practice, however, the police remain somewhat bound to conventional methods of operating, for several reasons. One is the sheer force of habit habits not only of the police, but also of the public and of other government institutions. Enormous investments have been made in the form of technology, training and organizational relationships to support conventional methods like criminal investigation, criminal prosecution and rapid response to calls for police service. A second, and yet more profound, reason why the police remain bound to conventional methods is that not all decision- makers accept the notion that law enforcement is but a means to other ends. The idea, that the fundamental purpose of the police is to enforce the law, however idealistic, remains powerfully attractive because it is simple and straightforward, and it seems, on its surface, to be consistent with more deeply held beliefs about the rule of law. The entire edifice of problem-oriented policing is built on the foregoing ideas about the fundamental objectives of the police, the recognition of law enforcement power as a means rather than an end, and all the implications these notions have for the exercise of police discretion and for police authority to operate by administrative rules, and not solely by legislative decree. In other words, problem-oriented policing makes sense to those who share these fundamental beliefs about the police's role and who see policing as a complex and sensitive function, but less so to those who don't. Many of these core beliefs get glossed over in the debates and discussions about problem- oriented policing. The debates and discussions then are about how best to implement problem-oriented policing, rather than whether it is the right approach to policing at all. Problem-oriented policing implicates some of the most important principles governing police power in a society of law. What Are the Various Operational Strategies of Police Work? Assuming, as I do, that Goldstein is correct in his articulation of the fundamental objectives of the police (that there are multiple objectives that overlap and, at times, conflict with, one another, and that law enforcement is but a means to these ends), it is then possible to understand policing in terms of the various methods or strategies used to achieve these objectives. The police employ innumerable specific tactics, but one can better understand these in terms of a few core operational strategies.92 There are five core operational strategies–preventive patrol, routine incident response, emergency response, criminal investigation, and problem-solving–and one ancillary operational strategy–support services. This, of course, is not the only way to conceptualize police work. The first four operational strategies constitute the ways police have conventionally done their work, at least since the 1930s. Problem-solving is a new operational strategy, introduced in Goldstein's problem-oriented policing concept. (See Chapter 1 for a discussion of the distinction between problem- solving and problem-oriented policing.) 92I struggled to find the right term to apply to this concept. After trying “mode”, “method”, “mental construct”, and “core process”, I settled on “operational strategy” thanks to a suggestion by Lt. Ken Bunker of the Reno Police Department. Each operational strategy of police work has unique and distinct features. Each represents a particular process or method for approaching situations the police encounter. Each is taught to police officers (problem-solving, only recently), and officers are taught when each is appropriate. Each has a distinct general procedural framework that guides officers in doing their work within that operational strategy. Each has a distinct general goal or objective. Each entails a unique way of defining a unit of work, and distinct general performance standards and indicators. Each has its own accountability, reporting and record-keeping systems. Preventive Patrol Preventive patrol remains the predominant operational strategy of policing in terms of time spent, all research questioning its effectiveness notwithstanding. It is the operational strategy in which uniformed police officers are expected to operate when they are not otherwise compelled to operate differently. The fundamental logic of preventive patrol is twofold. First, the presence of uniformed police officers is intended to deter citizens from committing offenses, and to enhance their sense of security. Second, the presence of officers is intended to increase the probability that they will interrupt offenses in progress. The objectives of preventive patrol are to prevent and detect offenses, and promote a general feeling of security. Few police departments use formal performance indicators to measure preventive patrol, although many departments still try to quantify the amount of time officers dedicate to preventive patrol by foot, and some capture vehicle mileage.93 Recruit officers are taught methods of preventive patrol, though few experienced officers seriously adhere to these methods. Unlike the other operational strategies of police work, preventive patrol does not lend itself to discrete work units; rather, it is an ongoing activity. Nor are there strong systems of accountability for preventive patrol beyond the occasional chewing out of an officer who fails to detect a commercial burglary on his or her beat. While preventive patrol has been deemphasized by many modern police managers, it remains a strong public expectation of police. Police patrol operations remain principally structured around preventive patrol, emergency response and the handling of routine incidents. 93O.W. Wilson, one of the principal proponents of the value of motorized preventive patrol, reportedly thought of installing electronic sensors on streets in Chicago to monitor the frequency and patterns of motorized police patrols. Such a system would allow for quantifiable performance indicators related to preventive patrol. For further reading on the history and effectiveness of preventive patrol, see Kelling et al. (1974) and Police Foundation (1981). Routine Incident Response Most reactive police business is handled using the routine incident response operational strategy, encompassing the vast majority of what patrol officers and their civilian support staff do (other than preventive patrol). Routine incident response entails the methodical collection of information about a situation, and classification of the situation (crime, information exchange, civil matter, etc.). Most police agencies have over 100 classification categories. The specific police objective will, of course, vary depending on the nature of the situation, but generally, the objective is to restore order, document information or otherwise provide some immediate service to the parties involved. Specific performance indicators are such things as satisfied citizens, no repeat calls for service during that tour of duty, etc. Most routine incidents are packaged as a “call for service,” complete with a permanent record of the incident and the police response. Much of the patrol operation is judged by how officers handle these routine incidents. Emergency Response Police use the emergency response operational strategy far less frequently than the routine incident response operational strategy, yet it is probably the most critical to the police agency's success, because human life is most directly at stake. It encompasses crimes in progress, officers' requests for immediate assistance, traffic accidents with injuries, natural disasters, and so forth. The general objective is to save lives, minimize injury and restore a basic level of order. Until the police achieve these objectives, they can employ no other operational strategy of police work. The unit of work is commonly thought of as a “critical incident” or an “emergency response.” Special reports about major critical incidents and how they were handled are sometimes prepared and reviewed with an eye toward improving future responses to similar incidents. The police are specially trained in emergency response techniques, from vehicle operation to first aid to hostage rescue. Criminal Investigation The criminal investigation operational strategy, while constituting a smaller proportion of police work than most people imagine, dominates the public's perception of police work and the police's perception of themselves; that is, thoughts about investigative work and images of detectives contribute to an idealized understanding of policing. There is a basic framework common to all criminal investigations, from those of the most minor crimes, such as shoplifting, to those of the most complex, such as homicide. Once the police determine that a crime has been committed, the elements of criminal law provide the general framework for investigations, and various techniques have been developed to enable the police to establish the statutory elements of crimes.94 The unit of work in criminal investigations is the “case.” There are special procedures for managing the processing and flow of cases. The standards of proof applied to criminal investigations are legal standards. Police must have “reasonable suspicion” to detain suspects, “probable cause” to arrest them, and enough evidence for prosecutors to establish “proof beyond a reasonable doubt” to secure a conviction. The general objective in this operational strategy is to prepare a prosecutable case. Case clearance and case filing rates provide the specific performance indicators and serve as the foundation for accountability in criminal investigations. There is an abundance of specialized training unique to the investigation of crimes. 94Preliminary crime investigations can lead the police to pursue responses in addition to or other than criminal investigation, but all too often, police investigators limit themselves to a criminal investigation, without broadening the inquiry into the larger or underlying problems. “The sense of emergency in policing has crowded out our capacity to think about problems in the long term.” – Dan Reynolds Problem-Solving The fifth operational strategy of police work is what is now referred to as problem-solving. Historically, it is the least well-developed by the police profession. While the police have always used the mental processes of problem-solving, problem-solving as a formal operational strategy of police work has gained some structure and systematic attention only in the past 20 years. Like the other operational strategies, problem-solving has a distinct framework for guiding action. Problem-solving methodology in policing is known familiarly by such acronyms as SARA or CAPRA. It entails problem identification, analysis, response, and evaluation. The general objective of problem-solving is to reduce harm caused by patterns of chronic offensive behavior. The unit of work in problem-solving is known as a “problem,” a “problem-solving project” or a “POP project.” Performance indicators are significant reductions in harm that are plausibly caused by some specific intended intervention, reductions that hold for some reasonable period of time. Sufficient standards of proof have not been developed, but the current standards are adapted from the social sciences. Problem-solving also involves some specialized training, and systems for reporting and accounting for problem-solving are being developed. Support Services A sixth operational strategy rounds out the picture of the business of policing. This operational strategy, which one might call support services, incorporates the many ancillary services the police provide to the public. The police provide these services routinely, rather than in response to any specific situation. Such services include providing copies of police reports, taking fingerprints for noninvestigative purposes, distributing or teaching generic crime prevention information, operating youth activity programs, and so forth. This operational strategy relates only indirectly to the police's fundamental objectives, although its scope has clearly grown in the era of community policing. It serves primarily to promote and enhance police legitimacy in the eyes of the public by providing nonconfrontational, nonadversarial and noncontroversial services to the public. Table 7 summarizes the operational strategies of police work and their distinct features. Table 7 Operational Strategies of Police Work -Operational strategy -Work Unit -Objectives -Record System -Reporting Requirements -Performance Standards -Specialized Training -Processes -Accountability -Preventive Patrol -None – ongoing -Prevent and detect offenses, promote general feelings of security -Daily activity reports, patrol vehicle mileage -Daily activity reports -Absence of crime, low levels of citizen fear, high rates of police detection of certain types of offenses (e.g., commercial burglary) -Patrol methods (random, directed, conspicuous, inconspicuous) -Limited – some officers use systematic area coverage patterns and plans -Limited – some expectations officers will detect certain offenses on their beats, some command accountability for absence of citizen complaints about police presence -Routine Incident Response -Call -Record incident, resolve dispute, provide or take information -Dispatch records -Report or coded disposition -Complainant satisfaction, no repeat calls that shift, fair treatment of parties, proper completion of report -Special training by type of incident -Procedures according to call type, reporting requirements -Code out call, file report; accountability rests with officer assigned and shift supervisor -Emergency Response -Critical incident -Save life, interrupt crime, protect property, minimize injury -Dispatch records, after- action reports -Critical incident report -No deaths, minimal injuries, order restored -Vehicle operation, first aid, hostage rescue, SWAT, defensive tactics -First aid procedures, critical incident procedures, triage -Primary officer or scene commander, until incident ends (handed off, if necessary) -Criminal Investigation -Case -Establish culpability, make prosecutable case, apprehend offender, clear case -Case files -Case report and file -Case filed by prosecutor, suspect apprehended -Death investigation, crime scene analysis, forensics, interviewing -Criminal investigative procedures -Case file deadlines, case management (handed off, if necessary), rests with detective assigned, unit supervisor -Problem-Solving -Problem or project -Reduce harm, reduce incidence, eliminate problem, improve response -Project files -Sometimes none, project report -Significant reduction in harm, caused by intervention, for reasonable period of time -Problem-solving methods -SARA, CAPRA -Rests with police chief, district commander, supervisor, and officer -Support Services -Program or procedure -Provide service, enhance police legitimacy -Program reports -Program or budget reports -Use/popularity of service -Specific procedures -Written procedure or curriculum -Fiscal At What Levels Is Police Work Done? In addition to understanding police work in terms of the eight fundamental objectives and six operational strategies, one can also understand it in terms of the various levels at which police operate. That is, policing in any given jurisdiction occurs on several scales, ranging from a microlevel (or highly localized) to intermediate levels to a macrolevel (or communitywide). The microlevel refers to how individual, isolated, specific situations are handled. The intermediate level refers to the combination of separate situations into a larger unit of work. The macrolevel refers to the police agency's policies and practices related to an entire class of situations. For simplicity, I use three levels of aggregation to describe the scope or scale of police work. The scale of the police work is roughly proportionate to, and is determined by, the number of people affected by a particular situation–often, the number of victims or complainants.95 95Alpert and Moore (1998) point out that the size (or scope) of problems can be described in various terms: “(1) [the] total resources committed to the problem, (2) [the] amount of time taken to solve, (3) the number of specialized resources required, (4) the extent to which higher-ranking officers must mobilize and coordinate efforts within and outside the department to deal with the problem, and (5) its importance and scale within the community.” There are varying operating levels in each operational strategy of police work. For example, criminal investigation occurs at the microlevel during the investigation of a single crime with a single victim (e.g., a theft or assault). It also occurs at the macrolevel, where the policies and practices for investigating an entire class of crimes, and potentially affecting the entire community, are determined. Criminal investigation also occurs at the intermediate level, where a series of individual crimes are combined for investigative purposes. A rash of burglaries or robberies in a neighborhood might be investigated jointly. Similarly, emergency response occurs at the microlevel (e.g., a single traffic accident, with injuries), the intermediate level (e.g., a natural disaster or large civil disorder), and the macrolevel (e.g., emergency preparedness planning). The same pattern holds for the problem-solving operational strategy, which ranges from highly localized beat-level (microlevel) problem-solving (e.g., one drug house, or even one person) to the intermediate level (e.g., a prostitution strip), to the macrolevel (e.g., juvenile homicides throughout a city). In each operational strategy, the scope of the situation should dictate the level of resources dedicated to addressing it. Almost all police work can be understood within this general conceptual framework of objectives, operational strategies and operating levels. The framework helps explain what the police are trying to achieve, how they are trying to achieve it, and on what scale they are operating. For example, the police might identify a problem related to crowds' congregating on the streets and sidewalks following political rallies. They might then decide that their primary objective is to safeguard the constitutional right to public assembly, with secondary objectives of preventing injury and facilitating the movement of traffic. They might then conclude that, in addition to handling the incident at hand, they need to study this type of problem further to develop a new response, because the current response is inadequate, and similar incidents arise in various contexts. Accordingly, they might then decide that the inquiry needed is sufficiently expansive to warrant making it a high priority for the research and planning unit, and to warrant assigning several police officers and supervisors who regularly handle such incidents to join the planning effort. The inquiry results would then determine the level of resources needed to address future incidents. Table 8 provides additional examples of police work at each level, in each operational strategy. Table 8 Operating Levels and Operational Strategies of Police Work Operating Level Operational Strategy of Police Work -Preventive Patrol -Routine Incident Response -Emergency Response -Criminal Investigation -Problem-Solving Macro -Patrol deployment plans -Policies related to categories of incidents -Policies related to categories of emergencies -Policies and practices related to categories of crimes -Policies and practices related to categories of problems Intermediate -Directed patrols by groups of officers -Traffic control at large public event -Bar fight, multiple-vehicle accident -Rash of burglaries in a neighborhood -Prostitution on a commercial strip Micro -Routine preventive patrol by beat officers -Dispute, minor crime reporting, provision of directions, minor traffic accident investigation -Traffic accident, with injuries; police officer in need of immediate assistance -Shoplifting; assault, with known suspect -Problem individual Note: The flow of the arrows reflects the need for data from the first four operational strategies to be analyzed in the problem- solving operational strategy, which in turn informs and improves the other operational strategies. The ultimate goal of police reform is to enable the police to better achieve the full range of their objectives, effectively, efficiently and in a manner consistent with basic principles of justice. To do so, the police must be able to perform well in each operational strategy of police work, and at each operating level. This requires that the police develop an organizational capacity to employ the appropriate operational strategy of police work with the appropriate level of resources.96 It means having a refined understanding of what particular objectives the police are trying to achieve. It means being able to make smooth transitions between and among the various operational strategies of police work, and up and down the operating levels. 96Sparrow observed that some forms of community or neighborhood policing that deploy the majority of police resources at the neighborhood or beat level inadvertently limit the police agency's capacity to respond to larger crime and disorder problems. He wrote, “[P]olice departments need to build their capacity to perform problem identification and analysis at many different levels of aggregation, and in many different defining dimensions” (1994:48). A good police officer is one who is always clear about his or her objectives, and knows how to transition from an emergency response to a routine incident response, or from a criminal investigation to problem-solving. A good police manager is one who knows how to ensure that each situation is being handled with the right level of resources, and in the appropriate operational strategy. Making the links between and among the cells of this matrix is challenging and demands sophisticated police work and management–knowing, for example, when a pattern of routine incidents indicates a larger underlying problem that might lead to worse disruption of community life if not addressed, and then using the right level of resources and the right processes to address the situation. A good police department is one in which all operational and administrative systems are aligned and prepared to respond to the community's needs. Where policing often goes wrong is in failures to recognize and balance competing objectives, failures to recognize that a different operational strategy is required for a situation, and failures to use the right level of resources for a particular situation.97 Precisely because the dynamics of social conflict change so quickly, police organizations are seriously challenged to become highly sensitized to these changes and to respond appropriately. In its broadest sense, problem-oriented policing is a framework designed to help police meet this challenge. 97I found an outstanding example of a police agency that tries to determine the appropriate operational strategy of response and operating level at the earliest possible time. The Merseyside, England, Police have created what they call incident management units (IMUs). The IMUs, staffed by police constables and analysts, receive notification of most nonemergency citizen complaints to the agency. Once they log a complaint, they begin a preliminary analysis of it to determine if it constitutes part of a larger problem. They then either try to address the problem, or forward the information to the appropriate operational personnel for follow-up (Merseyside Police n.d.). The above conceptualization of police work in terms of the interdependent relationships between and among objectives, operational strategies and operating levels is my own. Herman Goldstein conceptualizes these matters a bit differently than I do. In his view, problem-oriented policing is a mindset that transcends the operational strategies of preventive patrol, routine incident response, emergency response and criminal investigation. It is an analytical way of thinking about and addressing all of the business of policing. In his view, if all of the business of policing, including the handling of incidents, emergencies and criminal investigations, were subjected to a problem-oriented approach, it would ultimately inform the way the police perform those functions. For Goldstein, problem-solving is more like the research behind police operations. I see problem-solving more as part of police operations. Goldstein worries that my characterization of problem-solving as a distinct operational strategy of police work reduces it to a lower level of importance than is warranted, and detracts from the holistic nature of his approach. My intent is quite the opposite: In conceptualizing problem-solving as a distinct operational strategy of police work, I intend to elevate it to a level of importance and attention commensurate with that of preventive patrol, emergency response, routine incident response and criminal investigation. For most of the history of policing, problem- solving has not been recognized as a distinct operational strategy of police work. I contend that, even since the advent of problem- oriented policing, most police agencies still have not elevated problem- solving to the level of the other operational strategies, failing to develop the formal systems needed to sustain it. Goldstein and I agree that the process of problem-solving is at least as important as the conventional processes the police use. How Should the Police Integrate the Need To Address Community Problems With the Desire To Improve Administrative and Procedural Processes? A problem-solving methodology can be applied to almost any endeavor requiring some critical thought before action. In the context of policing, problem-solving methods can be applied to community problems as well as to internal administrative and procedural problems. The mere application of a problem-solving process does not automatically render the undertaking a form of problem-oriented policing in Goldstein's terms. For example, a police department supply clerk could use a problem-solving process to work out difficulties ordering uniforms, but this would not make uniform acquisition part of problem-oriented policing. The “problems” to which Goldstein refers in problem-oriented policing are matters directly relating to the public's safety and security, not to the police agency's inner workings. Table 9 on the next page lists examples of what Goldstein refers to as “substantive community problems,” and examples of administrative and procedural processes. Similarly, the police can apply problem-solving to the process of investigating crimes or responding to emergencies, but if this results only in making these processes more efficient, without creating some overall improvements to the public's safety and security, it does not constitute problem-oriented policing. In Goldstein's terms, problem- oriented policing entails making tangible improvements to the public's safety and security, and increasing police effectiveness, not merely making police processes less burdensome to the police and/or the public. “Compared with other regulatory professions, the police have led the way in the early articulation and implementation of the problem- oriented approach. The police, however, have since run into a specific obstacle, which is their general failure to construct the managerial systems that are required to run problem-solving at higher levels, and as the core of police operations.” – Malcolm Sparrow Table 9 Examples of Substantive Community Problems vs. Administrative and Procedural Problems Substantive Community Problems98 Administrative and Procedural Problems • Auto insurance frauds • Assigning cases for investigation • Auto thefts for export • Automating report writing • Auto thefts in mall parking lots • Constructing new police facilities • Auto thefts in parking garages • Controlling police misconduct • Auto thefts/larcenies in commuter lots • Eliminating discriminatory personnel practices • Bullying in high schools • Establishing satellite police offices • Burglaries at schools and recreation buildings • Evaluating personnel performance • Burglaries at storage facilities • Gaining accreditation • Burglaries at warehouses • Implementing bicycle patrol • Burglaries in the suburbs • Improving media relations • Burglaries/thefts in areas near high schools • Maintaining official records • Carjackings • Negotiating labor contracts • Crack houses and shooting galleries • Preparing a budget • Cruising (automobile) by youths • Preparing patrol deployment plans • Day laborers (problems due to congregation of) • Promoting and rewarding personnel • Disturbances/riots during local festivals • Purchasing equipment and supplies • Drug dealing and pay phones • Recruiting and managing citizen volunteers • Drug dealing in parks • Recruiting police officers • Drug dealing/prostitution in motels • Reorganizing the police department • Drug dealing to schoolchildren • Setting shift rotation schedules • Drug markets on the street • Storing evidence • Drunkenness and fights in entertainment districts • Streamlining booking procedures • False intrusion alarms • Upgrading communications technology • Fights and disturbances at bars/clubs • Fights/weapons in high schools • Gasoline drive-offs • Graffiti in commercial districts • Homeless people loitering in libraries and public buildings • Illicit sexual activity in public places • Motorists running red lights • Muggings/assaults around bus terminals • Panhandling in commercial districts • Pawn shops (trafficking in stolen property) • Private apartment complexes (problems in) • Prostitution strips • Public housing complexes (problems in) • Robberies at convenience stores • Robberies/purse-snatchings of tourists • Shoplifting by juveniles • Shoplifting by professionals • Squeegee men (intimidation, extortion by) • Telephone frauds and shoulder surfing at public transport terminals • Thefts from construction sites 98This list of substantive problems is drawn from one Ronald Clarke, Michael Scott and Rana Sampson compiled for a funding proposal to the COPS Office (September 1999). The distinctions made earlier between problem-solving and problem- oriented policing do not mean that the problem-solving applied to administrative issues or to promote procedural efficiency is not important. Indeed, the police have a continual obligation to use their resources as efficiently as possible, and problem-solving processes can help them do so. Using thoughtful, analytic methods to address administrative matters can sharpen those skills needed to address community problems. However, no amount of efficiency-driven problem-solving can substitute for the more important and more challenging application of problem-solving to community crime, disorder and fear. The application of problem-solving methods to administrative or procedural matters represents one of the most significant sources of confusion about problem-oriented policing.99 A significant proportion of the observable problem-solving undertaken today in the name of problem-oriented policing is not focused directly on community problems, but rather on police agencies' administrative concerns or operational inefficiencies.100 99In the first few chapters of their book, Police As Problem Solvers, Toch and Grant present a generally faithful interpretation of Goldstein's concept of problem-oriented policing, and some useful insights into organizational obstacles to its implementation. However, they then proceed to describe, as an early example of problem- oriented policing, an initiative undertaken by the Oakland, Calif., Police Department in the late 1960s, in which line officers and researchers studied police-citizen conflict in Oakland, and developed programs to reduce it. Aside from the fact that the initiative occurred some 10 years before Goldstein wrote his first article on problem-oriented policing, the problem addressed was not a substantive community problem in the sense that Goldstein defines the term. The problem was certainly important, both to the police and to the citizens, and the officers' work was commendable, but the initiative is not a prime example of problem- oriented policing in practice. Toch and Grant recognized the tension between the Oakland problem and the types of problems Goldstein had in mind, but ultimately concluded the initiative did constitute problem-oriented policing. I respectfully disagree. Although Toch and Grant also described some “group problem- solving” Oakland police officers conducted regarding the police response to family violence, clearly a substantive problem, that effort did not reflect the sort of careful problem analysis Goldstein envisioned. 100Another scholarly article seeks to apply the problem-oriented policing model to sex discrimination in police recruitment (Prenzler 1997). While acknowledging the distinction between external and internal problems, the author argues that problem-oriented policing must address both. However important the sex discrimination problem, and however amenable it is to problem-solving analysis methods, defining this sort of inquiry as an example of problem-oriented policing stretches and distorts one of Goldstein's fundamental principles of problem-oriented policing–that the focus be on the community problems for which the police are responsible. Problem-oriented policing is only indirectly concerned with the administration of police work and with procedural efficiency. It is concerned with these matters only to the extent that they affect the quality of service the police provide to the public, and to the extent that administrative or operational improvements can actually contribute to increased public safety and security. Herein lies a real source of confusion and dilemmas for those trying to implement problem-oriented policing. To bring about a complete reorientation of policing, from an administrative and procedural focus to a substantive focus, many of the existing administrative processes need to change. Goldstein himself describes many of the administrative changes needed to effect this transformation–from hiring processes to training, from records management to information sharing. Making the organizational and administrative changes necessary to support problem-oriented policing, however, is not the same as practicing problem-oriented policing. Only systematic and well-analyzed improvements in policies and practices–those made to increase public safety and security–constitute the essence of problem-oriented policing. All else, however important, is ancillary. It is difficult to overstate the extent to which administrative and political matters can consume the time and attention of the decision- makers most responsible for public safety, including police administrators, other government agency administrators and legislators. Ironically, even when there is a deliberate move to adopt a problem orientation to policing or local government, the business of managing organizational change often crowds out the business of addressing actual community problems, at least among top decision- makers. Personnel matters, budgets and administrative procedures usually dominate staff meeting and legislative agendas, leaving little room for engaged discussion about substantive public safety problems, and how they can be alleviated. Even the research on problem- oriented and community policing is dominated by a focus on the processes of organizational change and the administration of these new styles of policing.101 There have been numerous federally funded studies related to implementing community and problem-oriented policing. They have ranged from surveys of departments claiming to have adopted some new style of policing, to site-specific studies of implementation. A considerable amount of the literature on problem- oriented and community policing has addressed these matters.102 Given the way the concepts of community and problem-oriented policing have been merged from the federal perspective, it is sometimes difficult to determine which studies focus specifically on problem-oriented policing. The literature on substantive community problems addressed using a problem-oriented approach is far less plentiful. Proponents and practitioners of problem-oriented policing have invested a lot of effort preparing police organizations to do problem- oriented policing, by restructuring the organization, rewriting policies, upgrading technology, and developing training programs. The idea has been to realign the organizations to do the new kind of work. Much of the realignment has proven traumatic to the organizations' personnel. It certainly has in the several police organizations for which I have worked. Some of that realignment and resultant trauma may be inevitable. It may turn out, however, that the practice of problem- oriented policing should precede the realignment of the organization. Without a clear understanding of what the final product is–the successful conclusion of problem-oriented policing initiatives that demonstrably improve public safety–the process of realignment is uncertain and threatening. Organizational change in police agencies should flow from the experiences of addressing community problems, in somewhat the same way that assembly-line processes in automobile manufacturing plants should flow from the design of the automobile. In short, form should follow function.103 101At a recent conference on research and evaluation, sponsored by the U.S. Department of Justice, Office of Justice Programs, I noted significantly higher attendance at panels devoted to the study of organizational change in police departments than at those devoted to addressing substantive community problems. 102See Goldstein (1990a: Chap. 9); Peak and Glensor (1999: Chap. 5-7); Geller and Swanger (1995); Wycoff and Skogan (1993); Oettmeier and Wycoff (1998); and Bittner (1990) for discussions of a range of issues related to implementing and managing problem-oriented policing. 103Some private corporations and a few police agencies have explored new methods of effecting organizational change whereby analyses of critical organizational processes dictate changes to those processes and, perhaps, to the organization's structure. This approach is referred to as “business process reengineering” or “core process redesign,” and the specific methodology is known as “process mapping.” For a more in-depth discussion of the application of process mapping to police operations, and of its connection with problem- oriented policing, see Challenge to Change: The 21st Century Policing Project, by Craig Fraser, Michael Scott, John Heisey, and Robert Chapter 3 Relating Problem-Oriented Policing to Other Movements in Police Reform and Crime Prevention The various schools of thought on modern police reform, as well as several parallel or complementary movements and theories, have significance for the problem-oriented policing movement.104 Some of these movements compete with problem-oriented policing for acceptance as a general model for improving policing, while others have nicely complemented problem-oriented policing, drawing from disciplines other than policing. I will briefly discuss the most significant of these movements and theories, and their relationship to problem-oriented policing. Team Policing Team policing, a loose collection of ideas about how the police might more effectively serve the public, is, in hindsight, seen as the precursor to contemporary community policing methods.105 Several key people, like Patrick Murphy, who advocated team policing methods also would later advocate community policing. Many U.S. police agencies tested and implemented team policing in its various forms in the 1970s and 1980s, though its true origins are reportedly traceable to Aberdeen, Scotland, in the 1940s (Sherman, Milton and Kelly 1973). A number of large and medium-sized police agencies can today attribute geographic decentralization of their operations to team policing initiatives. The decentralization of authority, however, which was central to team policing's underlying theories, proved more threatening to many police executives, and did not survive as well as geographic decentralization. Few people today have declared team policing either an unqualified success or an unqualified failure (Walker 1993). There is general consensus today that team policing might have been a bit ahead of its time, but that many of its premises were and remain sound, and that it had sufficient appeal both to the community and to rank-and-file police officers. Indeed, several core features of team policing, such as stability of geographic assignment, unity of command, interaction between police and community, geographic decentralization of police operations, despecialization of police services, greater responsiveness to community concerns, some decentralization of internal decision- making, and at least some shared decision-making with the community, are in place in many of today's police agencies. Even when these features fall short of what some might consider optimal, most police managers generally consider them desirable almost 30 years after the advent of team policing. 104These ideas, including problem-oriented policing, are variously referred to as movements, philosophies, models, paradigms, strategies, theories, programs, schools of thought, etc. Just what they should be called isn't certain, nor is it that important, but Goldstein clearly intended that problem-oriented policing be understood as something more comprehensive than a program, though nothing so grandiose as a philosophy. One observer interpreted Goldstein's writings as calling for an “existential” framework in policing. Whether they do or not, Goldstein never conceived of problem-oriented policing in such philosophical terms. 105One scholar (Brodeur 1998a:vii) suggests that the concept of problem-oriented policing actually originated in the team policing initiatives of the early 1970s. This is not quite true, although there is a connection. As evidence of this connection, Brodeur cites a reference to the term “problem-oriented” in Sherman, Milton and Kelly's 1973 report on team policing (p. 16). Sherman was describing the team policing pilot project in the Dayton, Ohio, Police Department. It turns out, according to Herman Goldstein, that before starting the team policing project, several Dayton police officials attended seminars at the University of Wisconsin Law School. During those seminars, University of Wisconsin police scholars like Goldstein and Frank Remington were introducing the notion of a problem-oriented focus to policing, a notion that grew out of their work on the American Bar Foundation surveys of criminal justice from the 1950s. Goldstein had not yet formulated his full-fledged concept of problem- oriented policing, but it is clear the seeds of the idea have a long history and remain entirely associated with Goldstein's work. Brodeur traces the links between Goldstein's earlier work and his subsequent work in problem- oriented policing in Chapter 2 of the publication (Brodeur 1998b). Community Policing The term community policing began to appear in the professional literature around the mid-1970s. Pioneering police departments, like the Santa Ana Police Department, used the knowledge acquired from team policing experiments to expand some of the elements more broadly into the department's routine operations and into how the police solicited active community participation in preventing crime (Sherwood 1977).106 Much has since been written about the relationship of community policing to problem-oriented policing.107 It is beyond the scope of this writing to explore all the distinctions and similarities, except to summarize some arguments Goldstein made about the distinctions.108 106The Flint, Mich., Police Department's Foot Patrol Experiment, begun in 1979, was instrumental in the subsequent formation of the National Neighborhood Foot Patrol Center (later renamed the National Center for Community Policing), housed at Michigan State University. Founded and led by Professor Robert Trojanowicz, this center became a prominent source of community policing training, publications and advocacy. 107For discussions of the distinctions between community policing and problem-oriented policing, see Brodeur (1998b), Skolnick and Bayley (1988), Toch and Grant (1991, Chap. 11), and Greene and Mastrofski (1988). 108See Goldstein (1985b, 1990a:21-27) for his explanation of the distinctions between problem-oriented and community policing. Most obviously, according to Goldstein, problem-oriented policing primarily emphasizes the substantive societal problems the police are held principally responsible for addressing; community policing primarily emphasizes having the police engage the community in the policing process. Under problem-oriented policing, how the police and the community engage one another will and should depend on the specific problem they are trying to address, rather than being defined in a broad and abstract sense. Community policing implies that responses to problems will involve some sort of collaborative or cooperative working relationship between the police and the community. Problem-oriented policing allows for this possibility, but does not imply that such arrangements are always necessary or appropriate for addressing every problem.109 Carefully analyzing problems before developing new response strategies is given greater weight and importance under problem-oriented policing than under community policing. Problem-oriented policing specifically promotes using alternatives to the formal criminal justice system, redefining the nature of the police's relationship to this and other systems; community policing does not explicitly address this relationship. Community policing strongly emphasizes organizing and mobilizing the community, almost to the point that doing so becomes a central function of the police; problem-oriented policing advocates such efforts only if they are warranted in the specific context of addressing a particular problem. Under community policing, certain features of police organizational structure and policy, like geographic decentralization and continuity in officer assignments to neighborhoods, are deemed essential; under problem-oriented policing, many of these features are seen as helpful, but not essential–problem-oriented policing can be done under a variety of organizational arrangements. Community policing emphasizes that the police share more decision-making authority with the community; problem-oriented policing seeks to preserve more ultimate decision- making authority for the police, even while encouraging the police to solicit input from outside the department. Problem-oriented policing emphasizes officers' intellectual and analytical skills; community policing emphasizes their interpersonal skills. Finally, community policing expands the police's role to advance large and ambitious social objectives, such as promoting peaceful coexistence, enhancing neighborhood quality of life, promoting racial and ethnic harmony, and strengthening democratic community governance; problem- oriented policing is more cautious, emphasizing that the police are more limited in their capacity to achieve these goals than many people imagine, and guards against unrealistic expectations of the police (Goldstein 1992).110 These selected general comparisons are summarized in Table 10. 109The Chicago Police Department has invested heavily in developing and delivering training programs to community groups, instructing them in problem-solving methods. The department has reportedly trained over 12,000 residents in a two-year span (Hartnett and Skogan 1999). Getting community members to understand the principles of problem-solving no doubt has some merit, but it is no substitute for the sort of problem analysis Goldstein advocates that police and trained researchers conduct. Moreover, from my own experiences developing and delivering problem-solving training to St. Louis community groups, doing so is a large undertaking that does not yield significant or immediate improvements in the quantity or quality of problem-oriented initiatives. 110Some scholars and observers who have merged the concepts of community and problem-oriented policing erroneously ascribe the more ambitious goals of community policing to problem-oriented policing, as well (see, for example, Alpert and Moore 1998). Table 10 Selected Comparisons Between Problem-Oriented Policing and Community Policing Principles -Principle -Problem-Oriented Policing -Community Policing -Primary emphasis -Substantive social problems within police mandate -Engaging the community in the policing process -When police and community collaborate -Determined on a problem by problem basis -Always or nearly always -Emphasis on problem analysis -Highest priority given to thorough analysis -Encouraged, but less important than community collaboration -Preference for responses -Strong preference that alternatives to criminal law enforcement be explored -Preference for collaborative responses with community -Role for police in organizing and mobilizing community -Advocated only if warranted within the context of the specific problem being addressed -Emphasizes strong role for police -Importance of geographic decentralization of police and continuity of officer assignment to community -Preferred, but not essential -Essential -Degree to which police share decision-making authority with community -Strongly encourages input from community while preserving ultimate decision-making authority to police -Emphasizes sharing decision-making authority with community -Emphasis on officers' skills -Emphasizes intellectual and analytical skills -Emphasizes interpersonal skills -View of the role or mandate of police -Encourages broad, but not unlimited role for police, stresses limited capacities of police and guards against creating unrealistic expectations of police -Encourages expansive role for police to achieve ambitious social objectives “The practice of problem- solving seems to have stalled, partly because it has not been sufficiently distinguished from its frequent companion (community policing), and has therefore been viewed by many police agencies as a question of professional style for beat-level officers, and not a central challenge for the departmental management structure. Some problems that the police must address don't lend themselves to the sort of community partnership responses envisioned by community policing, and for those kinds of problems, problem-solving has been less well-developed. Those problems nonetheless are amenable to problem- solving interventions.” – Malcolm Sparrow From the perspective of those committed to problem-oriented policing as a framework for police reform, the community policing movement has been a mixed blessing. On the positive side, the general idea of community policing has been enormously popular with the general public and, consequently, with elected officials. More specifically, the promise to the public of more access to the police, more police presence in the community, and greater police responsiveness to community concerns largely accounts for community policing's popular appeal. This popularity has translated into substantial financial and authoritative support for a wide range of programs, policies, training, and research, some of which has also benefited the problem-oriented policing movement. As noted above, to the extent that problem-solving has become at least a central feature of most conceptualizations of community policing, problem- oriented policing has benefited from greater attention to this analytical aspect of police work. Community policing's emphasis on improving the general relationship of the police to the community at large, to minority communities and to organized community groups has undoubtedly helped the police be more effective in their efforts to address particular community problems in a problem-oriented framework. This is no small achievement of the community policing movement. On the negative side, the most politically popular features of community policing have not been the behind-the-scenes analyses of community problems, but the more visible programs that put police officers in all kinds of unconventional settings–on foot and bicycles, in classrooms, in community meetings, at youth recreation functions, etc.–and that have police officers providing unconventional services to the public, such as educating, mentoring and relating to youth. The attraction to these aspects of community policing has drawn some financial and authoritative support away from the analytical aspects of problem-oriented policing. The popularity of community policing has helped problem-oriented policing gain a degree of attention it might otherwise not have so quickly, but has reduced it to the level of a simplified analytical process for guiding police activities. The challenge for problem-oriented policing advocates will be to maintain support for the further development of the concept's less visible, but more critical, elements. Crime Prevention Through Environmental Design Criminologist C. Ray Jeffery first articulated the concept of crime prevention through environmental design (CPTED) in 1971. Along with Jane Jacobs (1961) and Oscar Newman (1972), Jeffery recognized the importance of urban planning, building design and landscape architecture in controlling and preventing crime and disorder (National Crime Prevention Council 1997). CPTED is increasingly becoming essential knowledge and practice. A growing number of police agencies are developing in-house expertise in CPTED, and using that expertise to influence the design and use of land, buildings and other public spaces so that they are less prone to crime. CPTED, while existing as an independent method for analyzing and addressing crime problems closely tied to a geographic setting, has supported the movement toward problem-oriented policing (Saville 1999). Conversely, problem-oriented policing has reinforced the concept of CPTED. It has allowed police officers and others who make design decisions to view crime control from an entirely new perspective other than law enforcement. It has let them see, in tangible ways, a whole range of methods to prevent, or at least reduce, crime. Once exposed to the CPTED principles and methods, many police officers find themselves more open to understanding problem-oriented policing's broader implications. Many police agencies now train officers in CPTED and have them sit on local planning review boards. Situational Crime Prevention Situational crime prevention is perhaps the single most important intellectual development that reinforces and informs the problem- oriented policing movement (Tilley 1999). Its early articulation precedes Goldstein's articulation of problem-oriented policing. The two concepts developed independently, and then began to influence one another. Situational crime prevention is a relatively new branch of criminology, originating in England, that also has built and expanded on the concepts of CPTED and defensible space.111 Ron Clarke (1993)112 succinctly described it as an approach to crime prevention that “is directed at highly specific forms of crime and involves the management, design or manipulation of the immediate environment in as systematic and permanent a way as possible so as to increase the effort and risks of crime, and reduce the rewards as perceived by a wide range of offenders.” More simply put, it refocuses crime prevention away from deterrence and rehabilitation-based efforts to change offenders' underlying attitudes and behaviors, and toward more situation-specific methods of convincing offenders that committing a particular crime in a particular place at a particular time is not worthwhile. 111At least in the United Kingdom, the concepts of “crime prevention” and “situational crime prevention” are seen as related but distinct, with the former seen as the more comprehensive. 112For a brief and highly comprehensible review of the theories underlying situational crime prevention, see Felson and Clarke (1998). See, also, Clarke (1993). The ideas of situational crime prevention theorists like Ronald Clarke and Marcus Felson (see Clarke and Felson 1993, and Felson 1994) have significantly influenced a number of police scholars, who in turn are communicating the concepts to police practitioners. Many of the core elements of situational crime prevention parallel the core 114Some writers seek to align problem-oriented policing with their own favored criminological theories, but usually distort the concept in the process. Fyfe et al. (1997) assert that problem- oriented policing supports a “social conditions theory” of crime in which economic deprivation is seen as a primary cause of crime. They do so by arguing that the police should educate and inspire others to improve social conditions. Whether or not Goldstein would agree with this proposition, his concept of problem-oriented policing is not so explicitly linked to this theory of crime. elements of problem-oriented policing. Herman Goldstein and Ron Clarke have formed a sort of intellectual partnership, advancing the development of problem-oriented policing and situational crime prevention, respectively, while drawing heavily on one another's ideas (see Goldstein 1990a and Clarke 1998). In one respect, problem- oriented policing is the broader concept, not limited to crime problems, but also concerned with the full range of social disorder problems the police must address. In another sense, situational crime prevention is the broader concept, not limited to police actions, but concerned with the actions of any entity capable of preventing crime. Given the high degree of congruence of these two concepts, and the cross-fertilization of ideas, it is reasonable to assume that the two concepts will continue to fuse. Ideally, this fusion will continue to bring the scholars and practitioners of crime prevention closer to the scholars and practitioners of policing. Situational crime prevention has its theoretical roots in criminology. It starts from an intellectual interest in how to get offenders to curtail their crime. It is derived mainly from two theories of crime–routine activity theory and rational choice theory.113 Problem-oriented policing, on the other hand, has its roots in public administration and political science. It starts from an intellectual interest in how to get the police to be more effective in carrying out their functions in democratic societies. Problem-oriented policing as a distinct model of police reform evolved out of Herman Goldstein's early involvement in the American Bar Foundation Survey of Criminal Justice of the 1950s. Thus, in one sense problem-oriented policing is only 20 years old, but its intellectual heritage is several decades older. The findings and conclusions that emerged from the survey provided much of the intellectual foundation for problem-oriented policing (see Goldstein 1993b). 113Routine activity theory holds that predatory crime requires a convergence in time and space of a likely offender, a suitable target, and the absence of a capable guardian against crime. The rational choice perspective holds that offenders make rational choices to commit crimes, even if their information is imperfect or their calculations flawed. A related theory, crime pattern theory, looks at how people interact with the physical environment in terms of nodes, paths and edges (where they go, what routes they take, and the intersections of familiar surroundings of different groups of people) (Felson and Clarke 1998). Problem-oriented policing has at times been criticized for lacking a criminological theory for its foundation. This criticism presumes that a theory for improving police service must first set forth a theory for preventing crime. This, however, is a far more ambitious, and perhaps unrealistic, goal to which problem-oriented policing never aspired. Moreover, any proposal to improve the quality of policing must address the full range of police tasks and responsibilities, and not merely the control of serious crime. Problem-oriented policing is best understood as a framework for organizing the police and their activities so that the police are better positioned to learn how to prevent crime and disorder, and to apply that knowledge. It has no explicit preference for one criminological theory over others.114 It seeks to leave the police open to understanding various criminological theories, and experimenting with practical applications of those theories to determine what works best under what circumstances. This is not to say that problem-oriented policing proponents do not have favored criminological theories. Indeed, among the reasons there has been so much cross-fertilization of ideas between problem-oriented policing and situational crime prevention is that problem-oriented proponents have found merit in the theories underlying situational crime prevention, and police practitioners find the situational crime prevention studies relevant to their own work. But if those theories were ultimately proven wrong, it is unlikely that problem-oriented policing advocates would similarly conclude that the problem-oriented approach was also wrong. It would merely add to the knowledge base from which police practitioners could draw to guide their strategic decisions. Crime Analysis and Compstat Many police agencies have systematically analyzed reported crime data for a number of reasons–to identify potential suspects in specific crimes, to spot geographic and temporal crime trends, and to generally report crime and account for police responses to it. However, crime analysis, as it has conventionally been practiced, is quite different from problem analysis in several respects. Crime analysis focused on Part I Index crimes; problem analysis extends to any and all forms of crime and disorder. Crime analysis was used principally to identify offenders or predict the next crime in a pattern. Problem analysis is used to bring about more permanent reductions in the levels or severity of problems. Crime analysis concentrated on police activities to address crime. Problem analysis explores the whole community's response to the problem. Some agencies now have their crime analysts engaged in broader problem analysis, though mainly by providing, on request, statistical reports and analyses to those line officers leading problem- solving initiatives. Currently, one of the most prominent and popularized crime analysis methods is one patterned after the New York City Police Department's Compstat method (Giuliani and Safir n.d.). Increasingly, as news of the New York method spreads, police agencies are replicating Compstat.115 In essence, Compstat is a crime analysis method by which computerized crime statistics are analyzed and presented to operational commanders, who are then responsible for developing operational tactics to respond to emerging crime patterns. The degree to which this basic method is consistent with problem- oriented policing depends entirely on the details of how it is practiced. When statistics related only to reported Part I crimes are analyzed, the method has little in common with problem-oriented policing. 115The Police Foundation is currently studying the elements of Compstat and how the concept is being implemented in police agencies across the U.S. “Mapping, however, is actually a bit of a red herring. It can even be unhelpful. I worry that people are becoming obsessed with maps and their pretty colors, without thinking much about what information they contain or what can be learned from them. The technology itself becomes what is fascinating, rather than the knowledge to be gained from it. So technology can at times inhibit the development of problem-oriented policing, because it stops people from thinking.” – Gloria Laycock Problem-oriented policing specifically calls for a broad inquiry into many types of community problems demanding police attention. It also calls for analyzing multiple sources of information to develop a fuller understanding of each problem. Where a Compstat-style method results in commanders' selecting from among a limited and conventional set of responses to address problems, such as extra patrol or increased enforcement, it also departs radically from a problem-oriented methodology. Problem-oriented policing calls for a broad and uninhibited search for responses to particular problems, placing special emphasis on responses that minimize the need for the police to use force and large-scale arrest campaigns. A Compstat-style method can foster a hostile atmosphere, more like an inquisition than an inquiry; in this sense, it also differs from problem-oriented policing. Problem-oriented policing, while stressing accountability, also places a high priority on the free exchange of ideas, an exchange that is difficult to achieve in a tension-filled and rigidly hierarchical setting. Finally, a Compstat-style method relies exclusively on police analysis of data and results in decisions made exclusively by the police; in this sense, it also does not resemble problem-oriented policing. Problem- oriented policing puts a high premium on communication, consultation and collaboration with entities outside the police department at all stages of the planning process. Ideally, a Compstat-style method would be entirely consistent with problem-oriented policing. As one way to identify specific problems, a computer-generated pattern of crimes would be only the beginning of a more in-depth and broader analysis of the nature of the problems, their underlying conditions and the limits of current responses. For example, if computerized systems recognized a sudden spate of incidents classified as robberies in a police precinct, this information would not be used merely to mobilize conventional police responses like stakeouts and extra patrol, but instead might launch a closer analysis of the incidents that could reveal several discrete forms of problems, all related to the crime of robbery, each calling for a different set of responses. This should not be understood as an attack on the Compstat method. For many police agencies, this method is a significant advancement in the use of crime data to inform operational decisions. Problem- oriented policing, however, is a considerably more sophisticated and involved approach to handling police business than a Compstat method simplistically practiced. Hot-Spot Policing and Crime Mapping Over the last decade, many police scholars and practitioners have developed theories and applications for understanding crime and disorder in terms of geographic patterns. This has been in part fostered by research that shows that reported crime and disorder tend not to be evenly distributed across the landscape, but concentrated significantly in certain areas. Researchers have since developed many tools to allow the police to map these concentrations, to better understand crime and disorder and direct their resources in response. The basic idea is hardly new in policing, though the technology has made such mapping infinitely more possible and potentially useful. Its practical utility depends as much on how the data are organized, what questions are asked about the data, and what conclusions are drawn from the data as it does the volume of data and computing power and sophistication. Because of the new computer mapping technology, crime mapping has reached new levels of prominence in many police operations. It is becoming a specialized field in policing.116 116The National Institute of Justice established the Crime Mapping Research Center in 1997, and the Police Foundation recently established the Computer Mapping Laboratory. For further reading on crime mapping and its implications for crime prevention, see Weisburd and McEwen (1998), and for its implications for policing, see Harries (1990) and Greene (1998b). See, also, La Vigne and Wartell (1998), Reuland (1997), and Block, Dabdoub and Fregly (1995). Crime mapping and its links to crime prevention can strongly support problem-oriented policing (La Vigne 1999; La Vigne and Wartell 1999; Taxman and McEwen 1998). Crime mapping is enabling police practitioners and researchers to think about crime and disorder and their relationship to other geographic phenomena in ways that were previously unimagined or impractical. Problem-oriented policing specifically calls for, among other things, an analysis of police incidents in terms of location as a potentially useful way to aggregate incidents into clusters. A spatial incident pattern can help stimulate a better understanding of the underlying causes of certain community problems. Crime mapping also fits well with situational crime prevention, in which understanding crime in the specific context of its location is critical. For example, crime maps might reveal a pattern of storage-facility burglaries, and that revelation might then prompt a closer analysis of those facilities' physical layout and management. Seldom will crime mapping alone suffice as problem analysis, but it is a potentially useful analytical tool. The spatial and temporal concentration of crime and disorder has led some scholars to propose what they call “hot-spot policing.” Hot-spot policing, in essence, requires that the police concentrate their attention and resources on places where and times when there is a significantly high volume of demand for police services. At this basic level of understanding, the idea also is compatible with problem-oriented policing. Crime mapping and hot-spot policing, however, are not comprehensive approaches to policing, as is problem-oriented policing. Using mapping as an exclusive means to identify and analyze community problems would leave many problems hidden, and artificially limit the analysis of even those problems with some spatial patterns. Many problems the police must contend with do not lend themselves to spatial concentrations, and thus will not show up on any hot-spot maps. Crimes such as credit card fraud, domestic violence or child abuse are prevalent throughout jurisdictions. Overreliance on mapping limits police inquiries to data that can be mapped, and much of the information the police need to get a complete and accurate picture of community problems is not readily captured in data that are mapped. To the extent that those who use computerized maps to analyze problems become fascinated by the technology itself, there is a risk that the reliability of the data underlying the maps will be taken for granted. In fact, a lot of police data relating to the location of crimes and incidents are ripe for misinterpretation.117 Broken Windows and Zero Tolerance The basic notion underlying what some have called the “broken windows” theory of crime and disorder is that, by having the police and community address the many minor community incivilities and signs of neglect, more serious crimes and disorder will be prevented (Wilson and Kelling 1982, 1989).118 This idea has spawned as a consequence, intended or not, an idea popularly referred to as “zero tolerance.”119 Zero tolerance indicates that the police will restrict or eliminate the use of discretion in enforcement, that they will enforce laws as strictly as possible within their means. The idea is also popularly linked with the perceived practices of the New York City Police Department during much of the 1990s. Whatever merits the broken windows theory and zero tolerance strategy may have, how these ideas have developed in practice has little in common with problem-oriented policing.120 In so many respects, the idea of zero tolerance is antithetical to problem-oriented policing. If Herman Goldstein has stood for nothing else in his academic career, it is that the police, of necessity and largely for good cause, exercise enormous discretion in choosing which laws to enforce, when, where, and how (Goldstein 1963, 1977, 1990a, 1993b). Problem-oriented policing builds on that premise, drawing into enforcement decisions even greater input from the community, prosecutors and other government officials. Optimally, the refined use of the police's arrest powers and the exploration of the many alternatives to arrest will result in less reliance on criminal sanctions to address crime and disorder. Problem-oriented policing does allow that brief periods of concentrated law enforcement might be entirely appropriate to intervene in and disrupt a pattern of crime or disorder, but rejects the wholesale adoption of anything like “zero tolerance law enforcement” as a standing remedy for most community problems. 117The address from where an incident is reported is easily confused with the address where the incident occurred. Consequently, the locations of public pay phones often appear to be “hot spots” of activity merely because many people use such phones to report incidents to the police. Many other incidents are attributed to nonspecific addresses, such as those occurring in large open spaces like parks and wooded areas. Computer-aided dispatch data require careful interpretation in order to reach valid conclusions. 118Recently published research indicated there is no strong evidence that either social or physical incivilities in a neighborhood significantly affect residents' fear of crime, neighborhood decline or incidence of crime. The researchers concluded that “study results warn against problem-oriented policing or community- oriented policing efforts that concentrate too heavily on fixing physical problems as a way to revitalize a neighborhood or reduce residents' fear. Neighborhood status and low crime are more important than 'broken windows' in a neighborhood for long-term stability and low fear” (Taylor 1999). 119A number of writers and observers have asserted a connection between the broken windows and zero tolerance concepts, but George Kelling, one of the originators of the broken windows theory, does not endorse it (Rosen 1999, Goldstein 1999). Nonetheless, politicians and lay observers often view broken windows and zero tolerance as the same concept, and furthermore, both are often held out as an alternative to community policing (see Massing 1998). 120The broken windows thesis actually evolved out of some highly specific problem-solving efforts in the New York City subway system, in which George Kelling participated, but over time the concept lost its problem-specific focus. For a critique and refutation of the broken windows thesis, see Sampson (1999) and Harcourt (1998). For a more detailed critique of the principles underlying zero tolerance and “ordermaintenance policing,” see Cole (1999). For a critique of the New York City Police Department's zero tolerance strategy, and a comparison with the San Diego Police Department's neighborhood and problem- oriented policing strategies, see Greene (1999). For further discussion of the distinction between zero tolerance and problem-oriented policing, see Cordner (1998). For a critique of the zero tolerance concept and an explicit distinction of it from problem-oriented policing, see Goldstein (1999). Summary All these movements in the realm where policing, crime prevention and research intersect, from community policing to zero tolerance, have influenced, and been influenced by, problem-oriented policing. Some of these movements can be said to be variations on themes found in problem-oriented policing, emphasizing one or another element. Much of what is referred to as community policing or community-oriented policing121 is but a variation on problem-oriented policing themes. Other movements are more properly understood not as rival comprehensive theories of policing, but as specialized trends that, properly used, support a problem-oriented approach. Crime mapping is such an example. Yet other movements, like zero tolerance, while purporting to be a variation on problem-oriented policing, in practice are countermovements that reject problem-oriented policing's most basic premises. 121Most uses of the terms community-oriented policing and neighborhood-oriented policing appeared after Goldstein coined the term problem-oriented policing. Goldstein chose this term carefully because he fully intended that the police organize and align their actions (i.e., orient their actions) around the notion of problems. It is less clear whether those who use the terms community-oriented policing and neighborhood-oriented policing similarly intend that the police should organize and align their actions around communities or neighborhoods, and if so, what that means, exactly. Efforts to understand the literal meaning of these terms help expose these concepts' strengths and deficiencies. Chapter 4 Major Challenges to Advancing Problem-Oriented Policing In the 20 years since Herman Goldstein first proposed his problem- oriented approach to policing, we have learned a lot about how it can and might be implemented and improved. We have learned more, perhaps, about the limits of reform than about the successes of reform, but in the whole history of policing, that seems always the case. The development of problem-oriented policing in the past 20 years is encouraging even though quite limited. Perhaps this is to be expected given that the police profession, certainly as compared to most other professions, is relatively young and still in an early developmental stage. It is still developing systems, standards and methods for accumulating and applying research knowledge to practice. Police, government and community leaders must come to appreciate the value that research can add to their decision-making about how to address complex problems of crime, disorder and fear. They must reflect thoughtfully on complex problems, even in the face of demands for immediate action, and resist adopting simplistic responses. Problem-oriented policing's full potential will not be achieved in a climate of haste and impatience. Problem-oriented policing, as an idea rather than a program, has no particular central institution controlling or guiding its development, if indeed that is desirable or feasible. No single institution controls the operations and administration of the thousands of U.S. police agencies. Various efforts at implementation have focused on different aspects of problem-oriented policing. Certain aspects are especially appealing or relevant to different police agencies and research organizations at various times. Considerable progress has been made with respect to some aspects of the concept, and less with respect to others. “We have many more police agencies and officers involved in problem-solving, and there are many exceptional efforts at addressing problems. But for the most part, no one has taken what was done by Herman Goldstein or by the Newport News Police Department and expanded upon it in any substantial manner.” – John Eck Setting an Agenda for Avancing Problem-Oriented Policing There are only a few institutions capable of setting a national agenda for the advancement of problem-oriented policing, but to date, no institutions have done so. Several have incorporated parts of Goldstein's vision into their overall agendas. The National Institute of Justice funded the Madison pilot project in problem-oriented policing in the early 1980s, and has since funded a number of other projects and initiatives related to problem-oriented policing. The COPS Office made “problem-solving” a component of its funding programs through a wide range of initiatives falling under the general definition of community policing. So while some NIJ, COPS Office and other Department of Justice funding has clearly helped advance problem-oriented policing, it cannot be said that problem-oriented policing is the central component of any of these agencies' agendas (National Institute of Justice 1999; Office of Justice Programs 1998; Office of Community Oriented Policing Services 1999).122 122NIJ's former director, Jeremy Travis, wrote about the need for the federal research community to engage in more problem-oriented research of the type conducted by Harvard University and the Boston Police Department in Operation Cease-Fire. He wrote, “The research profession needs to catch up with policing and to define a role in the problem-solving process” (1999). Initially, PERF was the primary institution advancing problem- oriented policing, much of it through Department of Justice funding. In PERF's early years, when Gary Hayes, a former student of Goldstein's, was the executive director, problem-oriented policing was a priority on PERF's research agenda. During this time, the Madison, London, Baltimore County, and Newport News experiments in problem-oriented policing were launched. After Hayes' untimely death, PERF continued to advance problem-oriented policing when Newport News' police chief, Darrel Stephens, took over as executive director. Projects replicating the Newport News study were launched, out of which grew the annual Problem-Oriented Policing Conference and strong San Diego Police Department leadership. The conference gave problem-oriented policing some base of support that was only partially tied to PERF's research agenda. Although conference attendance continued to grow, further research and experimentation in problem-oriented policing have been less prominent parts of PERF's research agenda after Stephens' tenure. Problem-oriented policing principles and methods are still incorporated into many of PERF's current projects, programs and publications, but few are as directly related to advancing the practice of problem-oriented policing as were the early projects. Herman Goldstein has set forth his own priorities for seeing the concept further developed, but as yet those priorities have not been translated into a coherent policy agenda for the profession. Goldstein has proposed a national agenda to support problem-oriented policing (Goldstein 1993a, 1994a, 1994b; see, also, Rosen 1999), though several elements remain largely unaddressed. Goldstein has proposed the following major elements to a research and technical support agenda. • Fund and promote applied research on specific community problems, including experimentation with different response strategies. • Publish case studies of effective practices related to specific community problems. • Compile, synthesize and disseminate research and practice related to specific community problems. • Develop high-quality training programs and materials for various audiences. • Develop materials to better communicate the concept to various audiences. • Help police agencies develop capabilities to do problem-oriented research. • Develop guidance in the analysis of problems and evaluation of results. • Identify and develop information technology that supports problem-oriented policing. • Train researchers to conduct problem-oriented research. • Explore internal and external organizational issues related to implementing problem-oriented policing. Among the many questions and concerns about the future of policing and the future of problem-oriented policing, I raise and address those I think are the most critical for advancing the concept. Many of the items in Goldstein's agenda are reflected in this discussion as well. The first set of questions relate to how the problem-oriented policing concept and substantive knowledge about community problems will be advanced and shared. The second set of questions relate to the role those other than the police must play if problem-oriented policing is to be practiced effectively and fairly. “One of the strengths of the concept is its simplicity, so it's hard to understand why it's so difficult for some people and agencies to do it… Those who struggle with the concept tend not to appreciate the value of data, the time it takes to make use of it, and the patience required to reflect on the real nature of problems.” – Gloria Laycock Advancing Problem-Oriented Policing Through Training, Research and Practice How Will the Principles and Methods of Problem-Oriented Policing Be Taught? Training in the principles and methods of problem-oriented policing for the many different audiences who might benefit from it remains sporadic and of varying quality. PERF continues to offer training in problem-oriented policing, some of which is now offered under the auspices of the Community Policing Consortium. The Community Policing Consortium has produced a standard training curriculum in community policing, one module of which is an eight-hour session on “community problem-solving” that was developed principally by PERF. The written curriculum adheres to the basic problem-oriented policing model. A short training course, however good, cannot possibly convey a complete understanding of, and proficiency with, the principles and methods of problem-oriented policing anymore than a short course could suffice to make police officers proficient in any other police operational strategy. The Community Policing Consortium also produced a six-part video series on community policing and problem-solving. Like the written curriculum, the series adheres to the basic problem-oriented policing model and is useful for an introductory-level audience. PERF and the Community Policing Consortium now offer some of this training on-line. In the absence of other high-quality, professionally developed training materials, many problem-oriented policing trainers continue to rely on an ad hoc collection of training videos and handouts. Some of these training videos are now almost 15 years old, still in use for want of anything better or more recent. At least 13 of the approximately 30 regional community policing institutes that were provided start-up funding by the COPS Office provide training in problem-solving, but the institutes had considerable latitude to design their own curricula and courses. Some institutes reportedly provide good-quality training in problem-oriented policing, and regional agencies heavily depend on them. While there were good and valid reasons to encourage innovation and local control of the institutes' curricula, it is unfortunate that the institutes' training in problem-oriented policing was not mandated and standardized.123 Mandating training in problem-oriented policing would have ensured a wider exposure of the concept to the field and standardizing the problem-oriented policing curriculum would have ensured greater control over the quality of the instruction. Much of the balance of national training programs in problem- oriented policing is provided by small training and consulting firms, and individuals. A few colleges and universities also offer courses related to problem-oriented policing.124 So while a number of organizations offer courses in problem-oriented policing, the number of training experts remains remarkably small. Even in police agencies that offer training courses in problem-solving as part of their preservice or in-service curriculum, the trainers are often the same subject-matter experts from the earliest days of the agencies' experimentation in problem-oriented policing; that is, the same few individuals are relied on at the local, regional and national level to provide training in problem-oriented policing. A challenge in the advancement of problem-oriented policing is to get those with an introductory-level understanding of the concept to progress to intermediate levels and those at intermediate levels to progress to an advanced level, thereby increasing the pool of people participating in and promoting the problem-oriented policing movement. 123I served as a technical consultant to several of the regional community policing institutes from 1998 to 1999, observed several training courses and met with other institute consultants. 124Among the colleges and universities that have recently listed courses in problem-oriented policing are Florida State University, the University of Maine at Presque Isle, Dalhousie University (Nova Scotia), Northwestern University (Traffic Institute), and Charles Sturt University (New South Wales, Australia). It is also still common for individuals and units other than the department's training unit to develop and conduct in-house training programs related to problem-oriented policing.125 This suggests that police agencies and professional training organizations have not yet fully adopted problem-oriented policing into their organizational missions. Most in-house training in problem-oriented policing, including that offered as part of preservice academies, is limited to one or two days of instruction. Such limited instruction, offered in discrete blocks of time, can familiarize participants with only the basic concepts; it can hardly be expected to make them proficient in practicing problem-oriented policing. A few police agencies go further by having recruit trainees actually do some problem-solving as part of their field training experience. Among the more encouraging recent developments in problem-oriented policing training is a project to develop new police field training guidelines and a model field training program. Funded by the COPS Office, the Reno Police Department and PERF seek to update what is known as the San Jose field training model to better reflect the new expectations in police work brought about by community and problem-oriented policing. The San Jose model, a considerable advancement for its time, is now almost 30 years old, and may not be suited to instruct new officers in the principles and methods of problem-oriented policing. 125The Seattle Police Department, with some funding from the U.S. Department of Justice, has developed a set of basic and advanced training curricula in problem-oriented policing. These materials were developed principally by the department's community policing division, and not by the training academy. The amount of training in problem-oriented policing today is far greater than it was even 10 years ago even if the total need is still not being met. Most organizations that provide police training offer at least a few courses in community or problem-oriented policing. Most training programs provide some instruction in the rationale for problem-oriented policing, the basic elements of the concept, and the process of problem-solving, as well as case studies or practical exercises in problem-solving. Some additional courses cover various related administrative matters like supervision or implementation. It is far rarer to find training courses in specific substantive community problems. Ideally, training in problem-oriented policing will move beyond simply covering the mechanics of problem-solving to a more advanced treatment of the state of knowledge about common community problems the police confront. One can imagine someday finding a range of courses covering police responses to such problems as commercial robbery, street-level narcotics trafficking, shoplifting, domestic violence, and so forth. Such training would not be limited to teaching enforcement procedures, investigative methods, or laws and policies, but would cover the nature and known causes of the problem, and proven methods of effective prevention, intervention and reduction. When such training courses become commonplace, problem-oriented policing will have moved out of its experimental stages and into a more sophisticated and detailed stage. There is a need for national training programs to provide police officials, including chief executives, middle managers and analysts, with intensive guidance in applying problem-oriented policing methods to difficult substantive community problems. Training programs like the U.S. Department of Justice's National Executive Institute, Law Enforcement Executive Development Seminars and National Academy concentrate more on the mechanics of police administration and leadership than on the substantive concerns of problem-oriented policing. One can imagine new programs, as well- funded and supported as these others, that provide police officials with more direct guidance on practicing problem-oriented policing. Some work has been done developing written instruction in problem- oriented policing methods for researchers and practitioners. The Bureau of Justice Assistance (1993) published a guide for police, written by John Eck and Nancy La Vigne, on methods for surveying citizens and the physical environment. The COPS Office has funded Michigan State University Professor Tim Bynum to produce a guide to problem analysis (forthcoming). NIJ has funded one of the principal researchers in the Boston Youth Violence project, Anthony Braga, to prepare a primer in environmental criminology (a branch of criminology that views the physical setting of crime as a causative factor) and problem-oriented research methods. As part of the COPS Office-funded project to produce a series of problem-specific guides for police, a companion guide to assessing the impact of problem- solving initiatives will also be produced. While there has been some progress in developing training courses and instructional materials for police practitioners and researchers, little has been done to develop materials to convey the concept to other audiences. Among the target audiences whose particular interests have not been adequately addressed are judges, prosecutors, elected officials, other government agency leaders, and community organization leaders. There are a few promising efforts. With funding from the Soros Foundation, Herman Goldstein is preparing a publication intended to reach some of these neglected audiences. Malcolm Sparrow's book Imposing Duties (1994) examines the underlying principles of problem-oriented policing in the contexts of the police, tax administration and environmental protection. It addresses some of the neglected audiences, too. How Will the Police Accumulate and Transfer Knowledge About Substantive Community Problems? The concept of problem-oriented policing is grounded in research methods. Thus, police agencies should place a high premium on written information–reviewing studies of similar problems and reporting in writing the results of the problem-solving process. The spread of knowledge about problem-oriented policing occurs at least in part through publications. How Substantive Knowledge is Shared in the Police Profession The police culture leans more toward an oral tradition than a written one.126 In other words, police officers gather information about practicing their work principally by listening and talking to other practitioners. This holds true from the field training of new recruits to the exchange of ideas about problem-solving. One reason the annual Problem-Oriented Policing Conference is so popular among police practitioners is that it affords them an opportunity to meet and talk with fellow police officers about practices that do and do not work. Officers can gauge the veracity and reliability of reports about problem-solving practices by getting a feel for the character of the people reporting them, and by asking detailed questions.127 By contrast, the Problem-Oriented Policing Conference is not as well-attended by police researchers. There are not formal presentations of written papers at the conference, yielding less grist for the mill of academic research. However much this oral tradition strengthens the police's social bonds, it inhibits the transfer of reliable, accurate knowledge. As many myths are perpetuated as truths. 126Mark Moore of Harvard University recounted how doctors in the late 1800s began to distinguish their profession from others, largely by developing a body of written information about effective medical practices (Law Enforcement News, June 15, 1999). 127In 1999, the Lancashire, England, Constabulary held an internal problem-oriented policing conference at which police constables from throughout the large police agency presented problem-solving initiatives. From what I observed, the presentations were well- received by fellow police officials, and the conference proved an effective means of exchanging ideas, promoting good practices and encouraging others to become engaged in problem-oriented policing. To a large extent, the concept of problem-oriented policing is designed to bridge the worlds of police practice and research by establishing common interests and a common lexicon. This bridge is far from complete. The transfer of collected knowledge and wisdom occurs differently among academicians than it does among the police. Researchers are expected to be familiar with the relevant literature on a particular subject. There is no similar expectation in policing. The police expect to learn about new practices by attending training courses and, occasionally, conferences, not so much by keeping abreast of the latest professional literature. Paid sabbaticals, by which a professional temporarily relocates to another professional organization in order to exchange knowledge, are common among academics, but not among police. Extended time spent outside one's own organization in an environment conducive to learning is every bit as necessary for spreading ideas and good practices in policing as it is in academia. A number of the leading police agencies in problem-oriented policing can trace the introduction of the concept to a key individual's sabbatical.128 Many other agencies that have implemented problem-oriented policing have been headed by police chiefs with experiences in other police departments, or have had key individuals who traveled extensively, providing training in problem-oriented policing. There remain far too few such opportunities for police officials, however. 128Learning through sabbaticals strongly influenced the implementation of problem- oriented policing in Edmonton, Sacramento, Reno, and Merseyside. Writing Down Problem-Oriented Practice The practice of problem-oriented policing has suffered from a lack of quality writing about project work. Some police departments, mindful of other duties that compete for officers' time, have minimized the reporting requirements for problem-solving, emphasizing the work rather than the writing. However much this may encourage officers to engage in problem-solving, it has left little lasting evidence that the work occurred. Without written evidence, the transfer of knowledge about problem-oriented police work is limited to the storytelling of the particular officers involved. Once they lose interest in telling their stories, the knowledge dissipates. The efforts to chronicle good problem-oriented practice at the national level have been beneficial, but modest. NIJ funded a project to collect case studies in effective police problem-solving, the results of which have been published by the COPS Office (Sampson and Scott 2000).129 In 1999, NIJ and the COPS Office funded the publication of the best submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. Several Community Policing Consortium publications include narratives about problem- solving submitted by police agencies. Since 1988, PERF has published Problem Solving Quarterly, a newsletter that chronicles police problem- solving projects. Publication of this newsletter has been sporadic in recent years, however, largely due to the lack of sufficient quality submissions. The electronic database POPNet, maintained by PERF, is another means of disseminating information about problem-solving initiatives, but as noted earlier, it has only a limited number of entries, only summary information, and little or no quality control. These few efforts represent a much smaller investment than Goldstein had in mind, and few of the case studies entail rigorous research methods. 129The categories of problems addressed in this volume are alcohol-related crime, apartment complex and other rental-property crime, burglary, college-related crime, cruising, domestic violence, drugs, false alarms, gangs, graffiti, group homes, homeless-related crime, mental illness, neighborhood disorder, parks, prostitution, robbery, and theft from vehicles. Practitioners must be encouraged to continue using problem-oriented approaches to community problems, and to maintain records of their actions. Researchers, whether in-house or external, must be encouraged to do the more formal writing about problem-oriented projects, writing that serves two audiences: researchers and practitioners. To do so effectively, they should spend enough time with police officers to genuinely understand the issues from the officers' perspectives. My own experiences chronicling police officers' problem- solving efforts confirm this. The most effective method for gathering information about problem-solving projects has been to interview the officers involved, independently review data about the problem, and write a narrative about the project. Self-reporting alone yields little, and without some independent verification, lacks reliability. If the profession desires and values good written reports of problem- oriented policing, then it must use people with substantial research and writing skills to produce them. Collecting, Synthesizing and Disseminating Research and Practice on Specific Community Problems A number of publications sponsored and/or published by the federal government have attempted to capture the state of research and practice with regard to specific types of community problems the police confront (see Appendix B). These various publications have not been organized into a centralized reference system, nor have they followed a standardized format. While many of the publications are of good quality and offer useful information to police practitioners, there still is no coherent research agenda that would lead to a comprehensive and current body of knowledge about specific types of community problems and/or common types of responses to them.130 The COPS Office recently awarded funding to this author and several colleagues to produce a series of problem-specific guides for the police, that will partially address this element of a national research agenda.131 The Office of Justice Programs and the COPS Office have sponsored conferences in recent years on research and evaluation. The conferences are intended in part to bridge the gap between researchers and police practitioners, and to focus on research lessons that would be of interest to practitioners. However valuable the research may be, a police official would not have found much in those conferences that directly related to the police response to community crime and disorder problems: much of the substance of the conferences related to organizational and management issues.132 How Can Problem Analysis Be Improved, and a Systematic Body of Research on Substantive Community Problems Be Developed? The police field continues to lack an organized and substantial body of knowledge about effective methods for addressing common community problems.133 There are few sources readily found that provide information about the causes of, and effective responses to, most such problems. This is largely because there simply hasn't been much relevant research conducted. A standard literature search on any particular problem would lead the researcher to a host of different professional journals, books and technical reports, many of which would provide only a theoretical perspective, rather than a practical perspective from which one might adopt proven interventions or fashion new ones. The amount of potentially useful information is no doubt much greater than most police officials realize, but because it has not been systematically compiled and annotated for use by practitioners, it remains largely unavailable to the police. The police profession would do well to begin such a systematic compilation and digesting of relevant knowledge if problem-oriented policing is to become an even more viable approach. A systematic program to conduct applied research on substantive community problems, and to compile and disseminate the results of the research findings to the police, would begin to build a body of knowledge both about specific types of problems and about types of responses to address them. 130Peak and Glensor (1999) dedicate a chapter to describing in summary fashion police agencies' innovative responses to problems related to drugs, gangs, graffiti, special populations (mentally ill, homeless, alcoholic offenders), domestic violence, housing problems and neighborhood disorder, prostitution, cruising, teen hangouts, and false alarms. 131These guides will provide practitioners with a summary of the state of knowledge about effective responses to specific community problems the police commonly confront. The guides will draw from the fields of both problem-oriented policing and situational crime prevention. These problem-specific guides, along with a listing of community problems, a classification scheme for community problems, and a companion guide on assessing problem- oriented policing projects, are expected to be available in late 2001 or early 2002. The initial guides in this series will cover the police response to assaults in and around bars, drug dealing in privately-owned apartment complexes, street prostitution, graffiti, thefts of and from cars in parking facilities, false burglar alarms, school bullying, shoplifting, residential burglary, commercial burglary, disorderly youth in public places, speeding in residential areas, loud car stereos, panhandling, rave parties, clandestine methamphetamine labs, robbery at automated teller machines, 911 hangups and acquaintance rape on college campuses. 132Of the 45 workshops and panels at the 1998 National Conference on Community Policing: What Works: Research and Practice, by my estimation, only seven addressed substantive community problems faced by the police. Most of the research panels addressed matters related to community policing implementation, police sociology, community perceptions of the police, police misconduct, information technology, and police management and administration. 133Toch and Grant (1991) articulated the need to develop a problem-oriented network of knowledge, both about effective responses to common community problems and about the implementation of problem-oriented policing. They argue that, to be useful, such a network must be more than a simple distillation of published research; it should incorporate the practical experiences of police officers engaged in problem-solving. In contrast to the earliest experimental initiatives in problem-oriented policing conducted by Goldstein in Madison and by the London Metropolitan Police, the vast majority of problem-oriented policing initiatives since that time have originated and been conducted at the operating level of police organizations. Police officers and their supervisors have led most of the problem-solving projects and have, for the most part, focused on problems concentrated in specific locations, neighborhoods or districts. There are few projects conducted at a policy level that involve reasonably sophisticated analysis and that focus on a large problem affecting an entire community.135 Goldstein's earliest works on the problems of drunken drivers and repeat sex offenders in Madison are prototypes of policy-level problem analysis.136 New York City provides two examples of commendable efforts at policy-level problem analysis. One was a study of the problem of runaway children in the city (Ryan and Doyle 1986). It captured the scope and nature of the problem, though stopped short of recommending particular changes to policy or practice. The second was a study of safety in city schools (Travis, Lynch and Schall 1993). It concluded with recommendations for structural and policy changes in the school system to enhance safety. Both studies were conducted and/or led by the New York City Police Department's administrative and research branches. From among the best submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing, approximately one-third of the projects addressed problems that affected an entire community, yet only about one-sixth of those had demonstrable leadership or involvement of policy-level administrators and/or professional research staff (see Appendix A). The best example of high-level problem analysis from the Goldstein awards was the study of the problem of youth gang violence in Boston (Boston Police Department 1998). This was a joint effort of university researchers, police officials and community leaders. From among those sophisticated crime reduction efforts reported in Situational Crime Prevention (Clarke 1997b), only a few involved substantial police involvement in problem analysis and subsequent response.137 134As noted by Harvard University's Mark Moore, the most substantial body of research on the effectiveness of police interventions is in the area of domestic violence, and this consists of only six significant experimental studies, from which it remains difficult to draw any firm conclusions (Law Enforcement News, June 15, 1999). The research on the multitude of other problems faced by the police is far more miniscule. Wrote another Harvard researcher, David Kennedy, “Over and over again, when consulting the literature to address particular problems, one realizes that the most basic questions about these problems have not been answered in a way that is very useful for informing policy and shaping practice” (1999). 135The geographic scope of a problem does not necessarily correspond to the degree of sophistication of the problem analysis. Some problems, though concentrated geographically, are quite large and warrant sophisticated analysis. The study of telephone fraud at New York's Port Authority Bus Terminal is such an example (Bichler and Clarke 1997). By contrast, some problems affecting an entire community are addressed by line-level personnel using less sophisticated research methods. The Fremont, Calif., Police Department's study of domestic violence repeat victimization is such an example (1997). 136For discussions of the many issues involved in conducting policy-level problem-oriented research, see Goldstein and Susmilch (1982c), and Kennedy (1999). 137The chapters that report some police involvement are those by Matthews (prostitution); Knutsson and Kuhlhorn (check forgery); Poyner (crime in parking lots); Brown (crime and disorder in public spaces in town centers); Hunter and Jeffrey (convenience store robberies); Anderson and Pease (repeat burglaries and car crime); Laycock (burglary); Homel et al. (drunkenness and violence near nightclubs); and La Vigne (crime in subways). The police involvement in these crime prevention projects varied. In some, police officials were instrumental in designing the intervention; in others, they were instrumental in implementing the intervention. From a crime reduction perspective, it matters less whether the police or some other entity is responsible for effecting the changes that bring down crime rates, but from a police perspective, the police miss many opportunities to effectively address crime and disorder problems because they are not engaging in much policy-level problem analysis.138 Large police agencies with research and planning units should consider shifting more of those units' focus to analyzing large-scale community problems. This also requires that police research units reorient their approach, expanding beyond conventional methods such as identifying spatial patterns of crime through mapping.139 Police agencies without such in-house expertise or resources should consider collaborating with outside researchers.140 Police researchers should have the skills necessary to conduct advanced problem analysis or, at a minimum, be able to make intelligent use of what literature exists on substantive problems. 138Bichler and Clarke (1997) reported a significant missed opportunity for the police to address the problem of telephone fraud in a bus terminal. 139The Westminster, Calif., Police Department recently reported a commendable effort to enhance its internal capabilities to conduct sound evaluation studies of departmental programs and initiatives (Community Policing Consortium 1999). The examples of evaluation studies cited, however, are either process evaluations or administrative-initiative evaluations; none are outcome evaluations of efforts to address substantive community problems, the sort of evaluations essential to the practice of problem-oriented policing. 140Goldstein and Susmilch (1982c) discuss the challenges faced by external researchers collaborating with police agencies in problem- oriented research. The sort of police practitioner-researcher collaboration envisioned for problem-oriented policing has not occurred more often as a result of difficulties on both sides. For their part, some police officials are impatient with extensive research, preferring to work on smaller-scale problems with rudimentary research than to wait for more sophisticated research to shed new light on larger problems. Researchers, for their part, sometimes find it difficult to make the transition from pure social science research methods to the action research called for in problem-oriented policing. Pure research frequently produces either interesting theory with little practical value, or exquisitely precise findings about factors that do not lend themselves to effective intervention. Consequently, they have limited value for problem-oriented policing. As a general proposition, the entire criminology field and related criminal justice sciences have been slow or reluctant to substantively engage in problem-oriented policing (see Clarke 1997a).141 The number of environmental criminologists is growing. Environmental criminology, from which the concept of crime prevention through environmental design flows, has a theoretical kinship with problem- oriented policing, although many environmental criminologists have yet to connect directly with police concerns. Environmental criminology to date remains on the periphery of mainstream criminology. There are few academic researchers with much practical experience in problem-oriented policing, so some police agencies would be hard-pressed to find the right kind of research assistance, even if they sought it.142 For their part, the police have viewed criminology as abstract and, accordingly, have not sought to incorporate the lessons of criminology into their practices. The field of environmental criminology is giving the police good reason to do so, however, and they would be well-advised to become more familiar with its lessons. 141I reviewed a list of the panel discussions offered at the 1997 Annual Meeting of the American Society of Criminology. Of the 442 discussions listed, only 42 (about 10%) of the titles related to the substantive community problems the police face. 142One of the objectives of NIJ's Strategic Approaches to Community Safety Initiative is to train researchers in problem-oriented policing research methods. In recent years, some research funding has promoted action research in policing. As mentioned earlier, the COPS Office's Problem-Solving Partnership and School-Based Partnership grants and NIJ's Strategic Approaches to Community Safety Initiative (SACSI) and Locally Initiated Research Partnerships in Policing programs (described in chapter 1, action research model section) encourage collaborations between the police and researchers. NIJ has also announced a funding program called Computer Mapping, Planning and Analysis of Safety Strategies (COMPASS). This program is intended to enhance the internal data analysis capabilities of police agencies in ways that will be compatible with community and problem-oriented policing. The COMPASS model expands on the Compstat model, combining police data with other community-related data (e.g., data on fear levels, vacant housing, street lighting, sanitation, housing, school safety, hospital records, courts and corrections, victimization, and drug use). In England and Wales, the national government is offering substantial funding (£250 million, or the equivalent of almost $400 million) to implement and evaluate crime reduction practices. The government will provide police and other agencies with research advice and expertise as part of the program. The program specifically promotes “placing greater emphasis on problem-oriented policing” (Home Office 1998). However much reliable research all these programs produce, they fall short of constituting a coherent and comprehensive research agenda regarding the specific types of crime and disorder problems the police confront. “Since the early 1990s, it has been frustrating that, after 10 years, most problem-solving efforts remained small local efforts by line officers, not intermediate or communitywide sorts of problems.” – David Kennedy Research units can't and shouldn't do all the work in problem-oriented policing, but they should do problem-solving of a different type than beat- level problem-solving, and should engage in some experimentation. – Rana Sampson Summary What these gaps in knowledge and learning methods mean for the further development of problem-oriented policing is not yet clear. Perhaps researchers will do more research that is directly relevant to police practitioners. Perhaps researchers will learn to write more intelligibly for a practitioner audience, and attend more practitioner conferences to convey their knowledge in person. Perhaps the police will come to read more and rely more on published information. Perhaps the expanded use of Internet technology will make written information more available, obviating the need to develop and staff extensive police libraries. The experiences of the past two decades suggest that the best avenue for systematically advancing knowledge is one that requires contributions from both practitioners and researchers. I don't know whether improvements in the research community will generate greater interest among the police in using research to address community problems, or whether a greater police demand for such research will spur researchers to action. One thing is clear: The quality and quantity of the underlying research and the writing about problem-oriented projects need substantial improvement, even while the current, more modest efforts should be recognized and encouraged. Defining Roles for Others in Practicing Problem-Oriented Policing Are New Alliances Between the Police and the Community Healthy? Problem-oriented policing, like community policing, stresses police collaboration with the community to address problems. This general proposition is hardly controversial, though the particulars of actual collaboration can spark debate and controversy. The general proposition that the police collaborate more extensively with the community has, on balance, been a positive development, for many reasons. However, under certain conditions, these new collaborations between police and community present significant challenges in a constitutional democracy. At times, the “majority rules” philosophy of the community and the conservative traits of the police combine to support police practices that the courts find threatening to the constitutional order. A couple of examples typify this concern. In June 1999, the U.S. Supreme Court struck down a Chicago ordinance that empowered the police to order people seen loitering on the streets with reputed gang members to disperse or be arrested.143 The Chicago police had ultimately endorsed this ordinance after initially voicing opposition. It also had support from segments of the predominantly minority communities in Chicago most likely to be affected by it. National police and municipal executives' organizations supported legal arguments in favor of the ordinance (though national police organizations representing black police officers opposed it). Civil libertarians and some neighborhood association groups also opposed it. The Supreme Court found that the ordinance granted excessive discretion to the police, even while acknowledging the legitimate interest the police had in reducing intimidation of the public by gangs.144 In the early 1990s, the city of Santa Ana experienced problems with transients' congregating around civic buildings, intimidating others. The Santa Ana police, in trying to respond to widespread citizen complaints, developed a strategy that entailed strict enforcement of all laws and ordinances against those deemed to be transients. Both the strategy and the insensitive tactics the police used–targeting their enforcement only on suspected transients, rounding them up and marking their bodies with booking numbers–led to a harsh legal judgment against the police for violating the offenders' rights.145 The Chicago police did not specifically undertake the initiative in the name of problem-oriented policing, although the department generally espoused the principles of community policing and problem-solving. 143City of Chicago v. Morales et al. No. 97-1121. Decided June 10, 1999. 144For further explorations of the issues addressed in the Morales case, see Roberts (1999). 145See “Homeless Rousted in Santa Ana,” Orange County Register, Aug. 17, 1990; and “Cases Dismissed in Santa Ana's Homeless Sweep,” Los Angeles Times, Feb. 7, 1991, for journalistic accounts of the situation. The Santa Ana police undertook their initiative as a problem-oriented project. Regardless of how the police label such initiatives, Goldstein imagined that the processes used in problem-oriented policing, in which the police carefully develop responses based on thorough research, and subject those responses to review and input from many perspectives, would reduce the possibility that the courts would challenge and strike down police actions. Had Chicago studied the problem of gang intimidation using a problem-oriented approach, it might have yielded a range of new responses, and any new legislation might have been better drafted to respond directly to the problem and thereby survive judicial scrutiny. Whether the police in the instances cited above were wrong in their actions is largely a matter of opinion (even the U.S. Supreme Court was not unanimous in the Chicago ruling). Some might see these examples as wrong-headed judicial opinions rather than wrong-headed police decisions. What bears remembering is that the mere application of a problem-solving process to community problems does not guarantee that all the interests of a constitutional democracy will be protected. Goldstein specifies that, when choosing from among various alternatives for responding to a problem, the police must consider the constitutionality of the response, the effectiveness of the response, and the potential for negative consequences. However, he does not, nor can he, specify what conclusions decision-makers might draw when considering these factors. Are New Alliances Between the Police and Other Government Agencies Healthy? The many new alliances between the police and other government agencies hold potential for overreaching. As a general proposition, Goldstein's model of problem-oriented policing endorses closer and more collaborative working relationships between the police and their counterparts in other municipal, state and federal agencies. By combining their respective expertise, resources and authority in creative ways, the police and other agencies can often accomplish more working together than they can working independently. Some jurisdictions have extended the principles of problem-oriented policing beyond the police to encompass the entire local government,146 though I am not aware of any local government that has fully adopted a problem-oriented approach. Nonetheless, there will be instances in which the respective agencies' independence is necessary to protect against overzealousness and abuses of authority. Partnerships should not be abandoned because of the possibility of overreaching, or even because of occasional incidents of overreaching, but administrators and oversight bodies should remain aware of the risks. 146The city of Westminster, Colo., described its version of these principles as “community oriented governance” in a brochure developed in the late 1990s. The city of Sacramento restructured the way it delivered many municipal services, to better connect with police efforts to practice problem-oriented policing. During the early 1990s, when problem-oriented policing was being strongly promoted in St. Louis, city building inspectors were encouraged to work closely with the police. The inspectors and the police frequently found themselves teaming up to close down suspected drug houses. In one such incident, to enforce a condemnation order, building inspectors (wearing body armor) broke into a building with a sledgehammer and ordered the occupants out. The occupants complained, and the inspectors were disciplined and nearly charged with a crime. The police, who had been working with the inspectors to enforce the condemnation order, defended themselves by claiming they didn't actually swing the sledgehammer. Fortunately, the incident did not end the working relationship between the agencies, though it did place some restrictions on it. In this case, the building inspectors began to overidentify with the police role and lost sight of the limits of their authority. There has also been some criticism of the extensive collaboration between police and federal prosecutors in their efforts to reduce shootings and homicides by aggressively prosecuting offenders under federal rather than state law.147 Some critics, including members of the federal judiciary, are concerned that such routine and expansive collaborations between local and federal authorities undermine important principles of federalism and result in some degree of unfairness to defendants. In Liverpool, England, the local (Merseyside) police have begun to work more collaboratively with housing authority agents. The agents have wider authority to conduct surveillance on tenants than do the police, and the police liaison between the agencies acknowledges the potential for abuse and confusion about their proper respective roles.148 147This practice has received considerable attention in Richmond, Va., but occurs routinely elsewhere, as well. 148Personal conversation with Chief Inspector Simon Byrne, local authority liaison officer, Merseyside Police, July 5, 1999. Similar concerns can and do arise when the police work collaboratively with such other government personnel as liquor law agents and probation and parole officials. Again, whether these new practices are unfair is a matter of opinion (though, sometimes, a legally binding one), but they serve as reminders that some novel forms of collaboration, however effective they may prove, raise important issues about procedural fairness and the checks and balances of government power. What Should Be the Role of Prosecutors? Historically, prosecutors have related to the police almost exclusively in terms of the criminal investigation function. From the policing standpoint, prosecutors' interests are just a few of many the police must take into account. Much of police work does not result in even the consideration of criminal prosecution. However, the police are central to almost everything prosecutors do. Prosecutors prosecute cases, and few prosecutions are initiated by anything other than police work. (This is especially true in local prosecutors' offices, where the vast majority of criminal cases are prosecuted.) So while prosecutors continue to see their role as narrowly limited to prosecuting criminal cases, they also see the police role in similarly narrow terms. The police generate prosecutors' workload and make prosecutors' work easier by conducting thorough investigations. Understandably, from this perspective, prosecutors jealously guard against any diversion of police resources away from criminal investigation. Prosecutors exert a powerful influence on police practices, despite the reality that only a small percentage of police work culminates in criminal prosecution. Most prosecutors view themselves as the chief law enforcement officer in the jurisdiction, implying some supervisory role over the police, and many police officials do not challenge this, having themselves adopted a belief that the police function primarily to collect evidence for criminal prosecution. At a minimum, prosecutors represent a powerful constituency for part of police work. They represent a demand for criminal cases, and they set standards for the quality of criminal investigations against which the police are measured. Supported by popular images of police work as principally being an enterprise of catching and convicting criminals, prosecutors hold the police both accountable for portions of their work and, to some degree, captive to it.149 149One of the longest-running and most critically acclaimed U.S. television shows is “Law and Order.” Every episode begins with the police investigation of a serious crime, and culminates with the prosecution of the resultant case. As each show opens, the narrator says, “In the criminal justice system, the people are represented by two equally important groups: the police who investigate crimes, and the district attorneys who prosecute the offenders.” There have been some efforts to reconsider prosecutors' role in the larger enterprise of promoting public safety (Coles and Kelling 1999; Glazer 1999).150 A few local prosecutors' offices around the United States have experimented in what has come to be known as community prosecution, borrowing from the notions of community policing.151 Typically, in community prosecution, prosecutors are assigned to geographic areas and are responsible for prosecuting all or most of the crimes that arise out of them. The prosecutors are expected to try to learn more about their area's public safety concerns. If community prosecution, however, is limited to prosecuting criminal cases along geographic lines, it is not a significant departure from conventional practice, and does not necessarily reinforce problem-oriented policing. If prosecutors actually reconsider their function as one of solving community crime, disorder and fear problems, rather than just prosecuting individual cases, they reinforce problem-oriented policing. One of the earliest and best articulations of this shift in the prosecutor's role was written by a former organized-crime prosecutor, Ronald Goldstock (1991). Goldstock argued that prosecutors should see their roles as more than just prosecutors of individual cases, but rather as leaders of efforts to reduce crime. He went on to argue that prosecutors, even more so than police or elected officials, are in the best position to exercise leadership in problem-oriented responses to crime. While that particular argument was less than fully persuasive, his advocacy for stronger prosecutor involvement in crime reduction was welcome. 150From 1986 to 1990, Harvard University's John F. Kennedy School of Government sponsored the Conferences on New Directions in State and Local Prosecution, a series similar to its Executive Sessions on Community Policing. Community prosecution and problem-solving were among the topics, but the conferences have not had the impact on the field of prosecution that their counterpart sessions had on the field of policing. 151Among the better-known initiatives in community prosecution are those in Montgomery County, Md., and Multnomah County, Ore. (See Washington Post, “Community Prosecution: Montgomery relocates law enforcement to the neighborhoods,” July 11, 1999. Op-ed piece by Montgomery County State Attorney Douglas Gansler.) Coles and Kelling (1999) refer to community prosecution efforts in Seattle; Portland, Ore.; Oklahoma City; Miami; Brooklyn, N.Y.; Kansas City, Mo.; Boston; Austin, Texas; and Indianapolis. The U.S. States Attorney's Office for the District of Columbia has also undertaken a pilot effort in community prosecution. Goldstock's vision of the prosecutor as problem-solver has gone largely unfulfilled. There have been a number of experiments with community prosecution and diversions from prosecution, including such alternatives as restorative justice programs,152 all of which move generally in the direction of problem-oriented policing. There are also instances in which prosecutors have collaborated in police problem- solving initiatives, but there are few reports of systematic, policy-level problem analysis by prosecutors. Some descriptions of community prosecution sound more like commendable efforts by prosecutors to aggregate individual cases into larger cases, and to improve the coordination of law enforcement responses among various agencies; they don't describe the sort of comprehensive problem analysis Herman Goldstein proposed. 152The COPS Office recently published two documents describing the links between restorative justice and community policing (Nicholl 1999a, 1999b). Prosecutors' failure to fully engage in a problem-oriented approach to reducing crime, disorder and fear is unfortunate for two related reasons. First, without prosecutors, a valuable perspective on crime problems is missing from many police-led initiatives. Prosecutors are better-aware of how cases are processed through the court system and, accordingly, are more aware of the effectiveness of existing means for disposing of cases. Prosecutors also are more aware of the range of legal responses that might be used to address a particular problem, as well as some of the risks of alternative approaches. Prosecutors have access to data and to judges, and research skills the police often lack. When prosecutors are open-minded and take a broad perspective on their role, they can greatly facilitate problem- oriented policing.153 The second reason it is unfortunate that prosecutors are not more engaged in problem-oriented policing is that their absence from the process conveys a powerful signal to the police that problem-solving is not valued as highly as criminal investigation. This can discourage the police from investing more fully in problem- solving. Detectives are especially sensitive to prosecutors' signals. Prosecutors' general disengagement with problem-oriented policing partially explains why so few police detectives have engaged with the concept, as well.154 The emerging movement toward community prosecution is a positive development toward advancing problem-oriented policing, but it is far from complete. This new orientation toward prosecution remains rare among prosecutors' offices, and it will require every bit as much effort to reorient prosecutors to their work as it is taking to reorient police officers to theirs.155 It will require some changes in how law schools train students, especially those aspiring to become government lawyers. Currently, conventional legal training offers little that would prepare a prosecutor for problem-oriented prosecution. 153The prosecutor's office in Mecklenburg County, N.C., assigned one of its best prosecutors to the Charlotte-Mecklenburg Police Department, where he became engaged with problem-oriented policing initiatives. 154In October 1999, the COPS Office sponsored a conference (“When the Heat's On: Leadership Sessions To Support Problem-Oriented Policing”) dedicated largely to the matter of involving investigators in problem-solving. In preparing for this conference, as well as in analyzing the best submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing (see Appendix A), I found it difficult to locate more than a handful of problem-solving initiatives led by detectives. For further discussion of the general lack of detective involvement in problem-oriented policing, see Eck (1999). See Cosgrove and Wycoff (1999) for a discussion of the trends among U.S. police agencies with respect to the role of investigations units in community policing and problem-solving. 155Several people with whom I spoke, including Dr. Ellen Scrivner, a deputy director of the COPS Office and a trained psychologist, and Dr. Mike Chatterton, the director of the Henry Fielding Centre at the University of Manchester (England), hypothesized that the transition to problem-oriented policing and prosecution may be proving so difficult because it is essentially a shift from the predominantly inductive reasoning methods of conventional policing and the law to the more deductive reasoning methods of science. What Should Be the Role of Local Government Leaders? If prosecutors have had limited involvement in problem-oriented policing, local government leaders have probably had even less so. To be sure, many, if not most, local government officials have found reasons to support community policing. At a minimum, it receives much federal support in the United States. Unlike those forms of community policing that readily translate into highly marketable programs whereby extra police officers provide new services to the public, problem-oriented policing does not lend itself to sound bites and simple political imagery. Consequently, local government leaders must first invest their time and energy in understanding problem- oriented policing's full implications, if they are to support it. They must invest in such areas as research and analysis, and information technology–investments that are not guaranteed to pay off at any particular time, but are highly likely to pay off in the long term. San Diego provides one of the strongest examples of local government leadership in problem-oriented policing. From the police department's earliest initiatives in problem-oriented policing, then- Assistant City Manager Jack McGrory came to understand its value and fully supported its implementation. When he subsequently was appointed city manager, he provided the strong leadership necessary to incorporate problem-oriented principles throughout the structure of the city government. Police departments are by no means the only government agencies susceptible to what Goldstein described as the “means-over-ends syndrome.” Fire departments, building inspection departments, public works agencies, social service agencies, and all others can become just as complacent as the police about their work. They, too, can easily come to understand their work in narrow terms, seeking to make themselves ever more efficient, and not necessarily more effective. They do not naturally see themselves in broader terms, as problem- solvers, anymore than do the police. Goldstein has long argued that one of the central objectives of the police is “to identify problems that have the potential for becoming more serious problems for individuals, the police or the government” (1977). He argues that the police are in a position to observe the indicators of a wide range of problems affecting the community, not just those directly related to crime. This implies that the police might identify problems that do not fit squarely or exclusively within their domain. Moreover, the processes of problem-oriented policing promote the development of police partnerships with other government agencies, partnerships that enable all involved to more comprehensively understand community problems. Whether the police just identify problems for others to address, or address those problems themselves, effective collaboration requires strong leadership from local government leaders. These leaders must work to overcome the traditional divisions of labor and responsibility among the various government departments. One of the hazards of the more popularized versions of community policing is that other government agencies may believe the police can provide more services and assume responsibility for more social problems than they did before. Without leadership to create new expectations that departments collaborate on public safety problems, such collaboration is not likely to happen. Problem-oriented policing is not featured prominently in the most recent conferences and publications of the major U.S. professional associations of local government leaders (the U.S. Conference of Mayors, the National League of Cities and the International City/County Management Association). Local government leaders have other concerns in addition to public safety, and many do not directly control police operations. However, given the importance of public safety, one would expect these leaders would think more about coordinated and systematic ways of improving it. For the most part, local government leaders still attribute primary responsibility for public safety to the police, fire and ambulance services, despite growing evidence that crime, disorder and fear are greatly influenced by land-use planning, economic development, business regulation, code enforcement, architecture, public housing management, and traffic engineering. Police officers have often assumed the roles of land-use planners, architects and economic developers, as the literature on problem-solving initiatives clearly indicates. It isn't that the police should be local government leaders' primary focus, but that the responsibility for public safety should be more evenly distributed among local government agencies. Were this the case, local government leaders would play a primary role in coordinating and guiding problem-oriented initiatives to reduce crime, disorder and fear. “If the public demands better police service, the police will likely provide it. But demand is a very localized thing, and demand isn't always expressed in a very clear or sophisticated fashion. Where there are good police leaders, they can help translate that demand into better policing.” – John Eck Community leaders still cling to the idea that government and the police can and will solve all their problems, and they tend not to want to take responsibility for solving their own problems. – Darrel Stephens Should the Police Be Held More Accountable for Reducing Crime, Disorder and Fear? Should the experience of the past two decades in problem-oriented policing lead us to conclude that the police, in fact, can effect reductions in crime, disorder and fear? Goldstein has long argued that problem-oriented policing is an approach that recognizes the limits of police authority and the limits of police practices alone to bring about significant changes in public safety. Goldstein argues that when the police and the community accept that the police are not omnipotent, the police can solicit and receive the active support of the community and other government agencies to more effectively address the problems of crime, disorder and fear. This makes sense, properly understood, but in light of how theory gets translated into public policy, there is a lot of room for misunderstanding. Whatever else the experience of the New York City Police Department in the 1990s has meant, for a good many political leaders and some police executives, it has represented a bold claim to police efficacy not heard in many years. The distilled lesson from New York, as taught by those who claim credit for its methods, is that the police, given sufficient resources, support and latitude, can bring about significant reductions in crime and disorder. Many elected officials and police executives around the world scrambled to emulate New York's methods, whatever they perceived them to be. It is unlikely that anyone will ever adequately describe precisely what methods the police department used during that time.156 The city and the police department are simply too enormous for simple explanations. How much the policies developed at police headquarters or the mayor's office actually reflected the practices of police officers in the city's 76 precincts and innumerable specialized bureaus, divisions and units might never be known. The broad characterizations of official policy emphasize high-volume arrests for even low-level offenses, and constant pressure on commanders to respond to emerging crime patterns.157 Official policy does not appear to have emphasized collaborative problem-solving. 156For further reading on the strategies and tactics of the New York City Police Department during the 1990s, see Silverman (1999), Maple with Mitchell (1999) and Bratton with Knobler (1998). 157The aggressive stop-and-search and arrest strategies used by various task forces and street-crime units have come under considerable criticism due to abuses of authority, though former Commissioner Bratton has publicly denounced those practices as being antithetical to his intended policies. By contrast, the San Diego Police Department is held up as the model for a different set of official police policies that yielded statistical reductions in crime and disorder that were at least as impressive as, if not more so than, New York's during the same time. The department stands as the model agency for problem-oriented policing. However, broad claims about accomplishments in San Diego do not necessarily reflect the daily practices of police officers there either.158 To what extent San Diego police officers engaged in New York-style aggressive arrest tactics, or New York officers engaged in San Diego-style problem-solving tactics, nor what impact these various tactics had on crime rates, is not fully known.159 158Professor Gary Cordner of Eastern Kentucky University is currently researching the nature, quantity and quality of problem-solving by line officers in the San Diego Police Department. He hopes to develop an understanding of why some officers do effective problem-solving and others do not. The research project, “Enhancing the Analysis and Response Components of Problem- Solving in Community Policing,” is funded by the National Institute of Justice. 159See Greene (1999) for comparisons between the New York and San Diego policing models. So, after two decades of experimentation with problem-oriented policing, we are not really much closer to answering the question of whether the police should be held more accountable for reducing crime, disorder and fear, and if so, what approach would best achieve this. Of course, the answers to these questions do not turn exclusively on the efficacy of problem-oriented policing. Elected officials and media representatives remain bent on having these questions answered in simple terms. Their search for simple answers remains in vain. After 20 years, problem-oriented policing has demonstrated an internal logic, has been successfully applied at the project level, and remains a promising approach for the foreseeable future. In light of the growing contributions to professional knowledge emerging under the framework of situational crime prevention and crime prevention through environmental design, there is growing reason to believe that collaborations of police, governments, businesses, and communities, committed to carefully analyzing community problems and developing tailored responses, can bring about significant changes to public safety levels. Beyond that, claims about the police's capacity to singlehandedly reduce crime, disorder and fear at the community or higher level are simply not warranted. The greatest promise of problem- oriented policing may be that it is the approach most likely to maintain the delicate balance between freedom and order, and minimize the likelihood that police actions will undermine their legitimacy in society. This is so largely because the problem-oriented approach rejects the very excessive reliance on the enforcement of criminal law, and the use of force that accompanies it, that so often leads to abuse and consequent erosion of public trust in the police. Achieving that much, while incrementally and systematically improving our understanding about how police and communities can effectively reduce crime, disorder and fear, is a considerable improvement from past approaches to policing. “Public policymakers and the uninvolved public develop their view of policing from watching TV cop shows. We haven't communicated the core elements sufficiently and consistently enough for them. – Dennis Nowicki Conclusion How Will We Know If Problem-Oriented Policing Works? The ultimate test of problem-oriented policing is whether it proves successful in enhancing police service. One can evaluate the progress of problem-oriented policing in several different ways. At a minimum, asking whether problem-oriented policing works, and asking whether the problem-oriented policing movement has been successful, are separate matters. The former relates to the ultimate outcomes of policing. It is a search for proof that the problem- solving methodology reduces crime and disorder, makes communities safer, and does so better than any other approach to policing.160 The latter relates to the process of implementing problem-oriented policing. It is a search for proof that problem-oriented policing has become the standard approach to policing. I will address each in turn. Asking whether problem-oriented policing works is tantamount to asking whether the police are effective in achieving their socially mandated objectives. This depends, of course, on what one believes to be the objectives of the police. If one uses Goldstein's eight objectives, discussed earlier, as a guide, the matter is indeed quite complex. Successful policing, in the broadest sense, is policing that achieves each objective. If there are competing and, at times, conflicting objectives, as Goldstein argues there are, then there can be no such thing as maximally effective policing.161 In addressing a particular community problem or handling some incident, the police often must compromise some objectives to fully achieve others (e.g., the police must block traffic to allow for a political demonstration, or the police must release a suspect because they cannot obtain evidence without violating the suspect's constitutional rights). Thus, there can be only optimally effective policing, meaning that the police have balanced their objectives. At the microlevel, one can determine problem-oriented policing's success only in a problem-specific way; that is, the best answer to the question of how one measures success in problem-oriented policing is “one problem at a time” (to play on Morgan Stanley Dean Witter's marketing slogan: “We measure success one investor at a time.”). One should assess police effectiveness with respect to each discrete social problem the police are at least partially responsible for addressing.162 160Sherman et al.'s exhaustive review of crime prevention research (1997) concluded that problem-oriented policing has proven promising as an approach to preventing crime and disorder, more so than community policing. Sherman correctly points out that problem- oriented policing is “essentially about insight, imagination and creativity,” and the scientific method itself. Thus, it is a fundamentally sound approach that does not depend on any one crime prevention theory for its viability. 161Mastrofski (1999) also argues that talk about the “bottom line” in policing is difficult, because there are competing and conflicting objectives. 162Some evaluators of problem-oriented policing measure impacts at a level of aggregation that does not correspond to the level of the problem-solving interventions (for example, the police respond to a highly specific problem, using a highly specific intervention, and the evaluators measure the intervention's impact based on the area's aggregate crime rate). This was the case in an evaluation of problem-oriented policing in Lawrence, Mass. (Bazemore and Cole 1994). Rosenbaum and Lurigio (1998) attempted to review the evaluations of problem-oriented policing and draw conclusions from them, but, in my opinion, also confused the appropriate levels of aggregation and focus of evaluation. Jesilow et al. (1998) acknowledged that one should measure effectiveness at a problem-specific level, but themselves adopted more-abstract levels of measurement in their study of problem- oriented policing in Santa Ana, Calif. Stockdale, Whitehead and Gresham (1999) concluded that applying economic analyses to policing activities made the most sense at the project or initiative level, and less so at more abstract levels. Because problems of crime, disorder and fear arise and abate through a complex interaction of social norms, laws and technology, there really can be no end point to policing.163 As one class of problems abates, new classes of problems arise. An obvious example is the entirely new class of problems the police face with the rise of the Internet as a means of conducting business and communicating. Indeed, police work is always described in the present participle–policing–and never in the past tense. A community is never considered to have been policed. Thus, while it is appropriate to judge problem-oriented policing by the degree to which it is effective in addressing society's current problems, one should also judge it by the degree to which it prepares the police to identify and respond to future problems. 163It may turn out that the search for definitive measures of effectiveness and causation in the realm of crime control and policing is in vain if conventional social science methods are employed in that search. Complex systems, which antisocial behavior and the efforts to control it surely are, may demand an entirely different scientific method than static systems that merely react to forces rather than adapt to them (Waldrop 1992, Lewin 1992). A full exploration of this possibility is well beyond the scope of this work, but it is a highly important possibility. One can claim the problem-oriented policing movement has succeeded once police agencies have integrated the problem-solving operational strategy of police work into their operations at least as completely as they have the other operational strategies of preventive patrol, routine incident response, emergency response, and criminal investigation. Once integrated, each operational strategy will always have room for improvement. One can make this assessment of the success of the problem-oriented policing movement with respect to particular police agencies as well as to the profession as a whole.164 164Capowich and Roehl (1994) used problem- specific measures of effectiveness in their study of problem-oriented policing in San Diego, as did Hope (1994) in his study of problem-oriented policing in St. Louis. Perhaps too many police agencies remain focused on measuring the quality of their efforts to implement the administrative systems that support problem- oriented policing–the training, information systems, supervision styles, performance measurement tools, personnel selection processes, records systems, etc.– and insufficiently focus on the quality of their efforts to do problem-oriented policing–the quality of their problem identification, analysis, response, and assessment. Wrote Mark Moore, “Thus the mark of an effective police department will not be how successful it is in implementing the most recent national model of a successful program, but instead, in how thoughtfully it crafts a local solution to a local problem, taking into account the local character of the problem and the local means of dealing with it” (1998). In addition, one can claim the problem-oriented policing movement has succeeded once the imbalance between policing's “means” and “ends” has been altered to better reflect a direct concern on the part of police administrators and researchers with the substantive aspects of police business. In his early writings on the concept, Goldstein (1981) identified several areas of police administration and research where this imbalance needed to be corrected: a. police administration texts, b. police conferences, c. police administration and criminal justice university curricula, d. police training programs, e. police chief selection criteria, f. police chief calendars, g. police journal content, h. state planning agency agendas, i. substantive policymaking participation, and j. police research-unit agendas. While there has been some move toward a greater substantive focus, it is my distinct impression that all the areas listed above still primarily have an administrative and organizational focus, to the exclusion of a focus on the substantive problems the police confront. As is probably true in all fields, the development of an important idea, or of several important ideas simultaneously, is not neat and clean. There is no central policymaking entity, at least not in American policing. Scholars and practitioners alike shift through time in their understanding and support of the various ideas. The ideas themselves are shaped by factors other than pure theory or tested practice: by political and popular interest, available funding and the desire to achieve distinction. While the uneven and sometimes contradictory way these various movements push and pull the police profession frustrates those who are committed to one idea or another, in the long run, this is for the best. It is best for society as a whole, and best for the problem-oriented policing movement. The diversity of ideas and the highly decentralized way they are implemented have ultimately led to refinement of the best of them. Were it even possible for the development of problem-oriented policing to be centralized and made more consistent, it would likely weaken the idea. A single wrong turn in centralized policymaking results in many wrong turns in police practice. There are risks to promoting homogeneity in the implementation of problem-oriented policing, whether through the requirements of federal funding programs or through other means. An idea such as problem-oriented policing, which has yet to be fully developed, needs diversity to grow. And so it is that problem-oriented policing competes in the messy marketplace of ideas about how to improve policing. Problem-oriented policing must pass the rigorous tests of academic scrutiny and criticism to prevail as a path for improving policing. To be tested properly, it must be implemented with at least basic fidelity to the fundamental principles laid out by Herman Goldstein. Goldstein never intended that problem-oriented policing, at least as he articulated it, be understood as a finished or definitive product. Indeed, according to the scholar Jean-Paul Brodeur: “[I]t would seem as difficult as it is futile to measure with precision the extent to which the new strategy has been implemented. Such a measurement implies freezing a paradigm that is characterized by its open-endedness” (1998b). Goldstein intended that problem-oriented policing be understood as a basic framework to be tested, refined and improved on. Problem-oriented policing has come a long way in 20 years, from the chalkboards and classrooms of the University of Wisconsin, to the squad rooms, community meeting halls and conference rooms where modern policing is played out. It has achieved a degree of professional interest, and some measure of public and political interest, that must be heartening to Herman Goldstein and those who believe in his idea.165 The development of problem-oriented policing, however, is far from complete. Ironically, the popularity of the idea puts it at risk of burning out, and that would be unfortunate.166 It is precisely because problem-oriented policing is so deeply rooted in what Goldstein calls the basic arrangements for policing in a free and open society–the most fundamental challenges for establishing domestic tranquility and order–that police, community and government officials can ill afford to rest comfortably on the progress made to date. 165As Toch and Grant so aptly concluded: Problem-oriented policing cannot afford to be insensitive to public sentiment, but it must have faith in the process whereby its solutions are derived. The point of such faith is not to ignore popular opinion, but to subordinate getting along to doing right, where facts and fashions differ. The dangers of not doing so are illustrated by past experiences, such as the saga of team policing, which was often aborted (prematurely) because it had been instituted as a gambit and not as an intervention responsive to an analysis of needs. Sentiments (in the case of team policing, the fear of riots) are often evanescent, while needs (such as the slum conditions that sparked riots) stay around and remain unmet. The fact that problem-oriented policing is now “in” should please us, but we must not confuse this fact with the reasons why the reform makes sense, which existed before the strategy was “in”…and should remain long after drug-related pressures subside (1991:285-86). 166Gilling offered a number of important reminders and cautions about using scientific or quasi-scientific research methods to make public policy as problem-oriented policing prescribes. He correctly noted that every step of the problem-solving process involves what are essentially political judgments about what problems are important, what facts are to be gathered, and what conclusions are correct. He concluded: Followed correctly, [the problem-oriented approach] provides the best opportunity of making a significant and lasting impact upon the growing levels of crime that have been a characteristic feature of most of the postwar developed world, and thus it offers liberation from the “nothing works” pessimism that still lies beneath the surface of crime control discourse. However, given its tenuous position as a relatively new paradigm, the problem-oriented approach cannot afford to underestimate the strength of the opposition manifested in traditional perspectives, alternative agendas, and the limitation of existing data sources and interpretative frameworks. There is a considerable amount of pressure being exerted on the problem- oriented approach to be stretched in a particular political direction (1996:21). Appendices Appendix A: An Analysis of the Best Submissions for the Herman Goldstein Award for Excellence in Problem- Oriented Policing, 1993-1999 In 1993, the Police Executive Research Forum (PERF) created an award program in conjunction with its annual Problem-Oriented Policing Conference (POP Conference). The award recognizes exemplary police projects that address community problems using a problem-oriented approach. The award was named in honor of Herman Goldstein. It is officially known as the Herman Goldstein Award for Excellence in Problem-Oriented Policing (hereinafter the Goldstein award). A panel of police supervisors judged the submissions for the award in the first few years of the program. Since 1996, a panel of police practitioners and researchers has judged the submissions. I have been one of the judges since 1996.167 I read and analyzed the best submissions available to me. These comprised the project reports that were awarded top honors from 1993 to 1995, and all submissions that survived an initial screening by the award committee from 1996 to 1999. I analyzed a total of 100 projects. (I didn't set out to analyze 100 projects–it just happened to total to a nice round number, thereby greatly simplifying my calculations of percentages.) The total number of submissions for the award far exceeds the number I reviewed for this analysis. In recent years, the program has received approximately 90 submissions per year. This sample of projects is not representative of all submissions to the award program, but rather is representative of what the judges have deemed to be the best submissions. Accordingly, my conclusions do not necessarily reflect an assessment of the state of all problem- solving as it is currently being practiced, but rather reflect an assessment of the state of what is being submitted for recognition as high-quality work. Undoubtedly, there is other high-quality problem- oriented work undertaken by police agencies that, for a variety of reasons, is never submitted to any award program for recognition. The projects I analyzed, arranged by police agency, are listed below. 167The judges from 1996 to 1999 have been Ron Clarke (criminal justice professor–Rutgers University, and chair of the committee), Gary Cordner (criminal justice professor–Eastern Kentucky University, and former police chief), Ron Glensor (deputy chief, Reno, Nev., Police Department; and adjunct professor–University of Nevada-Reno), Rana Sampson (police consultant; former public safety director–University of San Diego; former sergeant, New York City Police Department), Greg Saville (criminal justice research associate –University of New Haven, and former police constable at the Peel Regional Police in Ontario), and the author, Mike Scott (police consultant and former police chief). Since 1997, Nancy La Vigne (National Institute of Justice) has also been a judge. In 1996, Karen Lea (sergeant, St. Petersburg, Fla., Police Department) served as a judge. Project Title 1. Chronic Truancy Abatement Program 2. Operation Cease-Fire 3. San Juan Del Centro Housing Complex 4. Electric Avenue 5. Apartment Watch 6. Carolwood Park Apartments 7. ABC Enforcement Efforts 8. Gill Park Project 9. Raby Rebels Youth Project 10. Hartlepool School Watch Scheme 11. Apartment Managers' Hotline Program 12. Mario's Market 13. The Elite Arcade 14. San Jacinto Park Renovation Action Plan 15. Quality Neighborhoods Program 16. Transient Enrichment Network 17. Domestic Violence Revictimization Prevention 18. El Dorado Park “Stone Soup” Partnership 19. Group Homes 20. Local Ordinances and Conditional-Use Permits: The Empowerment of Law Enforcement 21. Child Custody Disputes and Court Order Violations 22. Blue Hole Park Project 23. Theft Reduction Auto Program 24. Day Laborer Project 25. Street Sweeping, Broadway Style 26. District 4 Thefts From Rental Vehicles 27. Methacathinone Laboratories 28. Center Court Apartments 29. Conflict Resolution in Farragut School 30. Black Tiger Karate Studio 31. Creston Apartments 32. Vehicle Accident Reduction Plan 33. The M.A.N.E.R.S. Project 34. Mission: Mission Lake Plaza 35. Northfields Project 36. Mental Evaluation Team 37. Virgil/Burns Area 38. Hollywood-Area Domestic Violence 39. Speeding in West Division 40. Harbor Area's Gateway Neighborhood Recovery Project 41. MacArthur Park Revitalization Project Police Agency Year 1. Baltimore Police Department 1999 2. Boston Police Department 1998 3. Boulder , Colo., Police Department 1996 4. Calgary, Alberta, Police Service 1994 5. Calgary Police Service 1996 6. Carol Stream, Ill., Police Department 1997 7. Charlotte-Mecklenburg, N.C., Police Dept. 1999 8. Chicago Police Department 1997 9. Cleveland, England, Police 1998 10. Cleveland, England, Police 1999 11. Colorado Springs, Colo., Police Department 1998 12. Delray Beach, Fla., Police Department 1996 13. Delta, British Columbia, Police Department 1997 14. El Paso, Texas, Police Department 1996 15. Fairfield, Calif., Police Department 1997 16. Fontana, Calif., Police Department 1998 17. Fremont, Calif., Police Department 1997 18. Fresno, Calif., Police Department 1996 19. Fresno Police Department 1996 20. Fresno Police Department 1997 21. Fresno Police Department 1999 22. Georgetown, Texas, Police Services Division 1995 23. Glendale, Ariz., Police Department 1997 24. Glendale, Calif., Police Department 1997 25. Green Bay, Wisc., Police Department 1999 26. Honolulu Police Department 1998 27. Indiana State Police 1997 28. Joliet, Ill., Police Department 1996 29. Joliet Police Department 1996 30. Joliet Police Department 1997 31. Kansas City, Mo., Police Department 1994 32. Kansas City, Mo., Police Department 1997 33. Lancashire, England, Constabulary 1999 34. Lauderhill, Fla., Police Department 1996 35. Leicestershire, England, Constabulary 1999 36. Long Beach, Calif., Police Department 1999 37. Los Angeles Police Department 1993 38. Los Angeles Police Department 1996 39. Los Angeles Police Department 1996 40. Los Angeles Police Department 1997 41. Los Angeles Police Department 1998 Project Title Police Agency Year 42. 1100 Block, 59th Place Los Angeles Sheriff's Department 1997 43. Tourist-Oriented Police Program Metro-Dade, Fla., Police Department 1996 44. Escort Services Metropolitan Bureau of Investigation, Fla. 1997 45. Fifth District, Levis Street Metropolitan D.C. Police Department 1997 46. Hawthorne Huddle Minneapolis Police Department 1999 47. Damascus Gardens Montgomery County, Md., Police Department 1996 48. Dilapidated House in Oceanside, Long Island Nassau County, N.Y., Police Department 1996 49. Crimes Against the Elderly Nassau County Police Department 1998 50. Roosevelt Avenue Project National City, Calif., Police Department 1997 (Anti-Prostitution Effort) 51. The R.A.I.D. Squad Initiative New Zealand Police 1998 52. PRIDE Program Newport News, Va., Police Department 1998 53. The Barrow Temperance Project of Public Safety North Slope Borough, Alaska, Department 1995 54. Tiffany Gardens and Western Hills Apartment Complexes Overland Park, Kan., Police Department 1998 55. The Last-Drink Program Peel, Ontario, Regional Police Service 1996 56. Turner-Fenton Peel Regional Police Service 1996 57. Nightclub Problems Phoenix Police Department 1997 58. Angela/Chanslor Area Pomona, Calif., Police Department 1999 59. Whitfield Towne Apartments Prince George's County, Md., Police Dept. 1998 60. Stop Break Queensland, Australia, Police 1999 61. The Power of Partnerships Racine, Wisc., Police Department 1999 62. North Side Redondo Gang Redondo Beach, Calif., Police Department 1996 63. Graffiti Task Force Richmond, Va., Police Department 1999 64. Vanier Project Royal Canadian Mounted Police, Quebec 1996 65. Auto Theft Royal Canadian Mounted Police, British Columbia 1997 66. New Helvetia and River Oaks Sacramento, Calif., Police Department 1996 67. Prostitution Restraining Order Program San Bernardino, Calif., Police Department 1999 68. Drag Racing San Diego Police Department 1996 69. Pallet Project San Diego Police Department 1996 70. The 501 Blues: The La Fripe International Project San Diego Police Department 1996 71. Auto Theft San Diego Police Department 1997 72. Macho's Nightclub Project San Diego Police Department 1997 73. Start Smart San Diego Police Department 1997 74. Mission Valley River Preserve San Diego Police Department 1998 75. Operation Hot Pipe, Smokey Haze and Rehab San Diego Police Department 1998 76. San Diego Traffic Offender Program (S.T.O.P.) San Diego Police Department 1998 77. San Ysidro Boulevard San Diego Police Department 1998 78. Truancy Control Project San Diego Police Department 1999 Project Title Police Agency Year 79. Lewd Conduct at San Elijo Lagoon and I-5 Viewpoint San Diego Sheriff's Department 1996 80. Harbor Plaza Santa Ana, Calif., Police Department 1993 81. Street Cruising Santa Ana Police Department 1997 82. Dennis Palmer Elk's Lodge Santa Barbara, Calif., Police Department 1996 83. Options, Choices and Consequences Seattle Police Department 1996 84. West First Revitalization Project Spokane, Wash., Police Department 1997 85. 911 Abuse St. Petersburg, Fla., Police Department 1996 86. Project Respect St. Petersburg Police Department 1996 87. Prostitution St. Petersburg Police Department 1997 88. Repeat Alcoholic Offenders in Downtown St. Petersburg St. Petersburg Police Department 1997 89. Unsolved Homicides St. Petersburg Police Department 1997 90. Eighth Street Temple, Texas, Police Department 1996 91. South Florida Seaports Internal Conspiracy Project U.S. Customs Service, Fla. 1997 92. Stop Stick Project U.S. Customs Service, Texas 1997 93. Perception of High Crime on Campus University of Alabama-Birmingham Police 1996 94. UW Police Response to Alcoholic Vagrants University of Wisconsin-Madison Police 1997 95. Intersecting Solutions Vancouver, British Columbia, Police Dept. 1999 96. South Central Prostitution Project Wichita, Kan., Police Department 1996 97. 21st Street Community Renovation Strategy Wichita Police Department 1997 98. Spectator's Club Project Wichita Police Department 1998 99. Hilltop Community Project Wichita Police Department 1999 100.Tropicana Motel Yuma, Ariz., Police Department 1998 Table 1 lists the number of projects I analyzed for each year. Table 1 Year No. of Projects 1993 2 1994 2 1995 2 1996 30 1997 30 1998 17 1999 17 Total 100 Note: Beginning in 1998, the committee decided to reduce the number of submissions to be reviewed by the entire committee (to approximately 15). They felt this better reflected the differing levels of quality of the submissions and permitted the judges to review the best submissions more carefully. I entered the following information from each project report into a simple database program: a. year of submission, b. title of project, c. name of police agency, d. state or province of police agency, e. country and region of police agency, f. type of police agency, g. nature of problem in terms of behavior, h. nature of problem in terms of place, i. nature of problem in terms of people involved, j. nature of problem in terms of time or event, k. scope of problem, l. type of responses used to address problem-oriented policing, m. references to “zero tolerance,” n. references to “restorative justice,” o. references to “crime prevention through environmental design,” p. position of project leader, q. position type of project leader, and r. level of recognition in award program. These data allowed me to analyze the following questions about observable trends among the best submissions to the award program: a. Where are the best submissions coming from? Which agencies? Which states, provinces or countries? What types of agencies? b. What types of problems are the police addressing? In what terms do police define problems? What is the scope of the problems the police are addressing? Who is leading the projects? c. What types of responses are police using to address problems? How often are certain popular response types used? These are questions of interest to me in assessing observable trends over time in the Goldstein award program. I did not capture detailed information about the methods used to identify, analyze and assess problems, though that information is surely of great interest to others. This survey of the 100 best submissions to the Goldstein award program from 1993 to 1999 is, of course, limited in scope and methodology. Even the best submissions to the program do not necessarily reflect the best of problem-oriented policing. In my own travels, I have visited a number of police agencies that have claimed to be engaged in problem-oriented policing routinely and for several years, yet that have seldom, if ever, submitted project reports to the program. It takes time and effort to prepare the submissions, and not all police officials are prepared to make that investment. There may also be good crime prevention initiatives occurring in which the police are only tangentially involved. Reports of those initiatives also might not find their way into the program. And, as noted earlier, problem- solving efforts from non-English-speaking countries usually are not translated for the program. These qualifications aside, I would nonetheless argue that the 100 projects I reviewed represent an important portion of the total amount of problem-oriented police work occurring. Allowing for some errors in judgment by the program judges, these 100 projects generally represent the best of the work submitted to the program. As to the limits of my methodology, I concentrated more on the nature of the problems addressed and the responses to those problems than I did on such other important steps in problem-solving as problem analysis and assessment. I will reserve that for future work. I reviewed only the project reports submitted to the program, reports that are limited in length by the program rules. I did not interview the problem-solvers or review supporting materials not submitted to the program, though doing so would surely improve an understanding of the work actually done. I classified elements of the projects in terms that make sense to me, though not according to any widely accepted analysis framework. Finally, I did not solicit any independent review of the projects to control for my own biases and errors. So what conclusions, however tentative and qualified, do I draw from a review of these best efforts? I offer the following summary conclusions, followed by a more detailed description of my findings related to each of my research questions. Summary of Conclusions a. The range of response alternatives used is the best aspect of problem-oriented policing being demonstrated in the Goldstein award program. The police continue to frequently use the criminal justice system, but usually more selectively and in conjunction with alternative responses. The police are willing to use informal and noncoercive response alternatives in addition to formal and coercive measures. This positive development is tempered in more recent years by increased resort to stock responses such as “zero tolerance” and “crime prevention through environmental design,” responses that may be sensible, but that too often are crudely applied. b. Problem-solving initiatives usually use a combination of response types to address problems. Multiple interventions, while complicating efforts to determine causes and effects, usually address problems more effectively than single strategy responses. c. Problem analysis remains generally weak, with most analysis serving merely to substantiate the existence of the suspected problem rather than to develop a more insightful understanding of why it is occurring. d. Assessment of response effectiveness is typically cursory, lacking in precision and certainty, although this aspect of problem-solving is improving, with greater attention being paid to such matters as control groups and displacement effects. e. Inadequate research resources are being dedicated to problem- oriented policing. Research expertise, technology and funding remain in scarce evidence in the Goldstein award projects. f. Police executives and mid-level managers are conspicuously absent from many good problem-solving initiatives. The good work of line-level police officers would likely (though not certainly) be improved by stronger involvement of higher-ranking officials in the process. g. Overall, most of the best submissions to the Goldstein award program come from the southwestern part of the United States, especially from southern and central California. There are high- quality efforts from other regions, but no other region produces the same amount of high-quality projects. Some states with large populations, as well as some police agencies reputed to be engaged in problem-oriented policing or community policing, are conspicuously absent from representation in the program. h. Problems ranging from serious crimes to nuisances, disorder and accidents are addressed using problem-oriented policing methods. This contradicts some claims that problem-oriented policing is applied only to lower-level disorder problems, and not to serious crime problems. i. Problem-solving initiatives addressed problems ranging from highly localized ones to those affecting entire communities. In fact, there were more of the latter than the former. This contradicts some claims that problem-oriented policing addresses problems that are only limited in scope. The level of analysis and response to communitywide problems, however, seldom matched the scope of the problem. Detailed Findings Where Are the Best Submissions Coming From? The Goldstein award program is open to police agencies throughout the world. Because the program is run by PERF, a predominantly, though not exclusively, American organization, most submissions come from U.S. police agencies.168 Due to language barriers, all submissions have come from English-speaking countries (including the bilingual province of Quebec), even though problem-oriented policing is practiced in non-English-speaking countries. Table 2 shows from which countries the best submissions have come. Table 2 Goldstein Award Submissions, by Country Country No./Pct. United States 86 Canada 8 United Kingdom 4 New Zealand 4 Australia 4 Of those projects submitted by U.S. police agencies, the majority (58%) were submitted by agencies in what I defined as the southwestern region of the country,169 with 41 percent coming from California alone (see Tables 3 and 4 on the next page). There are several possible explanations for this trend. The award program is closely linked with PERF's POP Conference. The award solicitations are sent out in conjunction with announcements about the conference, and the awards are presented at the conference. Since its inception in 1990, the POP Conference has been held in San Diego and cohosted by the San Diego Police Department. If for no other reason than logistics, police officials from southern California and surrounding regions have attended the conference in greater numbers than have those from other parts of the country. The San Diego Police Department is recognized as a leader in the practice of problem- oriented policing, and has exercised this influence worldwide, but especially in the Southwest region of the United States. It is also possible that the preponderance of the best submissions' coming from the Southwest reflects, in a general way, the high caliber of police personnel and management typically found in this part of the country. I don't know which of these factors best explains this trend, but it is perhaps the most obvious observable trend about the award. Table 3 lists the breakdown of U.S. submissions by region, as well as the total percentage of submissions each region supplied. Table 4 does the same, by state. 168The United Kingdom has recently started a new award program to recognize problem- oriented policing there. It is known as the Tilley award, in honor of Professor Nick Tilley of Nottingham Trent University, widely recognized as one of the premier experts in problem- oriented policing in the United Kingdom. In coming years, the Tilley award program may draw away from the Goldstein award program submissions from the United Kingdom, although that isn't certain. 169I categorized the states into the following six regions: Northeast: Connecticut, Maine, Massachusetts, New Hampshire, New York, Rhode Island, Vermont Mid-Atlantic: Delaware, District of Columbia, Maryland, New Jersey, Pennsylvania, Virginia, West Virginia Southeast: Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee Midwest: Illinois, Indiana, Iowa, Kentucky, Michigan, Minnesota, Missouri, Ohio, Wisconsin Southwest: Arizona, California, Colorado, Hawaii, Kansas, Nevada, New Mexico, Oklahoma, Texas, Utah Northwest: Alaska, Idaho, Montana, Nebraska, North Dakota, Oregon, South Dakota, Washington, Wyoming Appendix A Table 3 Goldstein Award Submissions From the United States, by Region of United States Region No. Pct. Southwest (Arizona, California, Colorado, Hawaii, Kansas, Nevada, New Mexico, Oklahoma, Texas, Utah) 50 58 Southeast (Alabama, Arkansas, Florida, Georgia, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee) 12 14 Midwest (Illinois, Indiana, Iowa, Kentucky, Michigan, Minnesota, Missouri, Ohio, Wisconsin) 12 14 Mid-Atlantic (Delaware, District of Columbia, Maryland, New Jersey, Pennsylvania, Virginia, West Virginia) 6 7 Northeast (Connecticut, Maine, Massachusetts, New Hampshire, New York, Rhode Island, Vermont) 3 3 Northwest (Alaska, Idaho, Montana, Nebraska, North Dakota, Oregon, South Dakota, Washington, Wyoming) 3 3 Table 4 Goldstein Award Submissions From the United States, by State State No. Pct. California 35 41 Florida 10 12 Illinois 5 6 Kansas 5 6 Texas 4 5 Arizona 3 3 Maryland 3 3 Wisconsin 3 3 Colorado 2 2 Missouri 2 2 New York 2 2 Virginia 2 2 Washington 2 2 Alabama 1 1 Alaska 1 1 District of Columbia 1 1 Hawaii 1 1 Indiana 1 1 Massachusetts 1 1 Minnesota 1 1 North Carolina 1 1 Total 86 97 Note: Percentages do not total to 100 due to rounding. That so many submissions come from California may be explained by the factors mentioned above. Other states are clearly underrepresented (or not represented at all) in the program. The San Diego Police Department is the source of more of the best submissions than any other single police agency (see Table 5). Table 5 Goldstein Award Submissions, by Agency Agency No./Pct. San Diego Police Department 11 Los Angeles Police Department 5 St. Petersburg Police Department 5 Fresno Police Department 4 Wichita Police Department 4 Joliet Police Department 3 Calgary Police Service 2 Cleveland, England, Police 2 Kansas City, Mo., Police Department 2 Nassau County Police Department 2 Peel Regional Police 2 Royal Canadian Mounted Police 2 Santa Ana Police Department 2 U.S. Customs Service 2 Baltimore Police Department 1 Boston Police Department 1 Boulder Police Department 1 Carol Stream Police Department 1 Charlotte-Mecklenburg Police Department 1 Chicago Police Department 1 Colorado Springs Police Department 1 Delray Beach Police Department 1 Delta Police Department 1 El Paso Police Department 1 Fairfield Police Department 1 Fontana Police Department 1 Fremont Police Department 1 Georgetown Police Services Division 1 Glendale, Ariz., Police Department 1 Glendale, Calif., Police Department 1 Green Bay Police Department 1 Honolulu Police Department 1 Indiana State Police 1 Lancashire Police 1 Lauderhill Police Department 1 Leicestershire Constabulary 1 Long Beach Police Department 1 Los Angeles Sheriff's Department 1 Metro-Dade Police Department 1 Metropolitan Bureau of Investigation 1 Metropolitan Police Department 1 Minneapolis Police Department 1 Montgomery County Police Department 1 National City Police Department 1 New Zealand Police 1 Newport News Police Department 1 North Slope Borough Department of Public Safety 1 Overland Park Police Department 1 Phoenix Police Department 1 Pomona Police Department 1 Prince George's County Police Department 1 Queensland Police 1 Racine Police Department 1 Redondo Beach Police Department 1 Richmond Police Department 1 Sacramento Police Department 1 San Bernardino Police Department 1 San Diego Sheriff's Department 1 Santa Barbara Police Department 1 Seattle Police Department 1 Spokane Police Department 1 Temple Police Department 1 University of Alabama-Birmingham Police Department 1 University of Wisconsin-Madison Police Department 1 Vancouver Police Department 1 Yuma Police Department 1 Total 100 Table 6 Goldstein Award Submissions, by Type of Police Agency Type of Agency No./Pct. Municipal 75 County 9 Regional 8 Federal 4 Campus 2 State 2 Total 100 The best submissions are predominantly from municipal police agencies (see Table 6). What Types of Problems Are the Police Addressing? One can identify problems in various ways, and Herman Goldstein recommends that the police do so in whatever way best characterizes the specific situation being addressed. As a general proposition, one can define problems in terms of the offensive behavior, the location, the people involved, or the time or event during which the situation occurs. Accordingly, I classified the 100 projects according to this general framework. I tried to capture the way the project reporters described each problem. In all cases, the project reporters could readily classify the problem being addressed in terms of the offensive behavior. In fewer instances was the location, people involved or time/event central to how the project reporters defined the problem. This analysis reflects as much how problem-solvers define problems as what behaviors, locations, people, and times/events are involved in problem-solving initiatives. Problems, by Behavior Table 7 on page 103 shows the types of problems addressed in the projects in terms of behavior, and is further organized by my own categorization of generic problem types (traffic-related, drug-related, alcohol-related, intimidation/fear, fire-related, sex offenses, assault, deadly assault, stealing, disorder/disturbances, other deadly behavior, environmental crimes/disorder, youth-related, and miscellaneous). My generic categories are not mutually exclusive. For example, the problem of sexual assault, which I placed along with other sex offenses, can also be defined as a nondeadly assault problem. In most instances, I classified the problem in precisely the terms reported in the submissions; in other instances, I modified the description slightly to fit into an existing category. Most submissions described the problem in reasonably specific terms, although some described it in more general terms, which I captured by such labels as “general crime” or “neighborhood deterioration.” Describing a problem in such generic terms is not all that helpful from a problem-oriented perspective, but I did so when there was no more-specific characterization of the problem behavior. Most submissions described several discrete problem behaviors within the context of a problem- solving project. Some submissions described the project in terms of a single problem behavior, while others described as many as 14 discrete problem behaviors. This analysis provides a rough indication of the types of problems being addressed in the best projects. The most common problem was “drug dealing,” followed by “assault,” “prostitution,” “vandalism,” “theft,” “disorderly conduct,” and “loitering.” The generic categories of “stealing” and “drug-related” were the largest. Table 7 Problem Types, by Behavior Problem Type, by Behavior No. Stealing Thefts Robberies Burglaries Auto thefts Thefts from autos Fraud Carjacking Fencing of stolen property Telephone fraud Theft of metal 57 13 12 11 9 5 3 1 1 1 1 Drug-Related Drug dealing Public drug use Drug smuggling Drug manufacturing 40 34 3 2 1 Intimidation/Fear Loitering Intimidation Panhandling Extortion Fear of crime Shots fired Bias crime Harassment Window washing (squeegee) 38 13 8 6 3 3 2 1 1 1 Miscellaneous Vandalism General crimes Trespassing Mental illness-related problem Public health 35 14 12 7 1 1 Assault (nonsexual, nondeadly) 33 Assaults 21 Fights 7 Domestic violence 5 Disorder/Disturbance 27 Disorderly conduct 13 Noise 7 Disturbances 3 Crowd disorder 1 Domestic disturbances 1 Juvenile disorder 1 Vagrancy 1 Deadly Assault 26 Shootings 12 Homicides 10 Drive-by shootings 2 Domestic homicide 1 Stabbing 1 Traffic-Related 25 Motor vehicle accidents 6 Parking 5 Drunken driving 4 Speeding 3 Traffic complaints 2 Cruising 1 Drag racing 1 Illegal public transportation (wildcatting) 1 Traffic congestion 1 Unlicensed driving 1 Environmental Crime/Disorder 24 Graffiti 11 Litter 9 Illegal dumping 2 Environmental waste dumping 1 Vehicle dumping 1 Sex Offenses 23 Prostitution 17 Sexual assaults 3 Escort service (prostitution) 1 Indecent exposure 1 Public sexual activity 1 Alcohol-Related Public intoxication Underage drinking Public drinking Alcohol-related offenses Alcohol incapacitation Sale of alcohol to minors and intoxicated people 15 4 4 3 2 1 1 Youth-Related Truancy Child custody dispute Missing children Runaway 7 4 1 1 1 Fire-Related Arson Explosion Fire hazard 3 1 1 1 Other Deadly Behavior Suicides Drowning 3 2 1 Problems, by Place While project reporters could characterize all the problems in terms of behavior, they did not define all the problems in terms of place. Ten of the projects did not lend themselves to description by place, largely because the problem occurred in various places throughout the affected community. Table 8 lists the types of places for the 90 projects in which the problem locations were specific. For 17 projects, the project reporters defined the problem, at least in part, as one of neighborhood decline or decay; that is, the general deterioration of a neighborhood was at least part of the problem to be addressed. Table 8 Problems, by Place Place Type No. Apartment complex 12 Commercial strip/district 10 Residential neighborhood 10 Licensed liquor establishment 6 Park 6 Roadway 6 School 6 Mixed-use neighborhood 4 Nightclub 4 Urban neighborhood 3 Border crossing 2 Riverbed 2 Shopping mall 2 Airport 1 Apartment building 1 Car dealership 1 Clandestine drug laboratory 1 College campus 1 Dilapidated house 1 Entertainment district (bars, clubs, taverns) 1 Indian reservation 1 Karate studio 1 Motel 1 Port 1 Recreation area 1 Group home 1 Retail clothing store 1 Video arcade 1 Problems, by People Involved As with place, the project reporters did not define all problems in terms of the people involved. Obviously, there are people involved in or affected by every problem, but project reporters defined the problem in terms of people in only 59 of the submissions. For example, although 17 submissions cited prostitution as a problem, only three defined prostitutes themselves as the focus of the problem. Table 9 shows the categories of people who were the focus of the problem-solving efforts. The predominant category was “gangs.” Table 9 Problems, by People Involved People Involved No. Gangs 26 Transients 7 Juvenile offenders 5 Students 4 Chronic alcoholics 3 Prostitutes 3 Car enthusiasts (hot-rodders, cruisers) 2 Merchants 2 Tourists 2 Day laborers 1 Drug couriers 1 Elderly victims 1 Indian youths 1 Homosexual men 1 Mentally ill people 1 Port employees 1 Domestic violence 1 victims Problems, by Time/Event Project reporters defined few problems by time or event. Only five of the 100 projects lent themselves to a temporal definition. Of those five, three related to the times when schools were in or out of session, one to bar closing hours and one to postgame victory celebrations. In addition to place and people, most problems also have some sort of temporal dimension, but the police officials leading the initiatives rarely used the time/event element as a way to organize their thinking about the problem. Table 10 Projects, by Scope Scope of Problem No./Pct. Localized 21 Intermediate 47 Communitywide 32 Total 100 Table 11 What Is the Scope of the Problems the Police Are Addressing? I classified each project in terms of the scope of the problem-solving initiative. By scope, I refer to the extent to which the problem affected the entire community. I classified each problem as either “localized,” “intermediate” or “communitywide” in scope. “Localized” problems typically affected a single residence, building, intersection, etc. “Intermediate” problems typically affected an entire apartment complex or neighborhood. “Communitywide” problems affected the police agency's entire jurisdiction (or, in some instances, just the entire jurisdiction of an agency subunit). These are rough categories and, accordingly, provide only a rough estimate of the scope of problems high-quality projects are addressing. Table 10 lists the number of projects in each category. Somewhat surprisingly, almost one-third of the projects addressed problems affecting the entire community. I found it interesting that, despite the fact that problem-solvers quite often identified problems of considerable scope, the level of resources the agencies dedicated to researching and responding to those problems seldom matched the scope; that is, quite often, problem-solving officers and supervisors found themselves trying to address large problems without the benefit of a lot of research assistance or substantial resources. Table 11 reports the level of police leadership for each scope of problem, and partially confirms this conclusion. Only 17 percent of the submissions reported significant command-level leadership (typically, lieutenant and above). Line officers led almost one-half of the projects (47%). In 38 percent of the communitywide projects, line officers alone provided the leadership. Command-level officers more typically provided direct leadership on intermediate-scope projects (in nine of the 47 reported). Line officers' being listed as project leaders does not mean that supervisors and commanders were disinterested in the projects or did not provide indirect leadership and support, but only that they did not provide direct oversight and were not closely engaged in the project. Scope of Problems, by Level of Police Leadership Level of Police Leadership Number (Percentage of Row) Scope of Problem Line Supervisor Command Support Row Total Localized 13 (62) 6 (29) 2 (10) 0 (0) 21 Intermediate 22 (47) 15 (32) 9 (19) 1 (2) 47 Communitywide 12 (38) 14 (44) 6 (19) 0 (0) 32 Column Total 47 35 17 1 100 Looking more specifically at the types of assignments of the officers leading the projects, I found that about one-half of them were assigned to some sort of specialized unit, or had assignments of which community policing or problem-solving was the defining feature. [Project reporters variously referred to these officers as community policing officers (or teams or units), neighborhood police officers, problem-oriented policing units (or POP teams), task forces, or some variation thereof.] About one-fourth of the projects had leadership from officers (including supervisors and commanders) in general patrol assignments. Detective and administrative officers exercised far less leadership. Table 12 lists the project leadership, by type of assignment. Table 12 Project Leadership, by Type of Assignment Type of Assignment No. Community policing, problem-oriented policing, neighborhood, task force, etc. 46 General patrol 26 Detective or special investigative unit 6 Drug or vice unit or officer 5 Traffic or DUI unit 5 Administrative command 3 Crime prevention 3 Police chief 3 School officer 2 Unknown 2 Note: The numbers do not add up to 100, as some projects had multiple leaders working from multiple assignments. What Types of Responses Are Police Using To Address Problems? The submissions reported a wide array of specific responses to problems. This was expected, given the range of types of problems addressed, and entirely consistent with Goldstein's insistence that police carefully tailor responses to the specific problem. To make some sense of the range of responses, I identified each specific response reported in the submissions, and classified each response according to the descriptive response categories Goldstein developed in Problem-Oriented Policing (1990: 104-141). Goldstein described 11 general categories of responses (three of which have multiple subcategories). These 11 categories, along with some specific examples of each, are listed below, along with the number of projects in which the police used the particular response. (Goldstein's categories are listed below followed by specific examples drawn from the 100 submissions. The number in categories with multiple subcategories can total to more than 100; otherwise, the number also reflects the percentage of projects for which the particular response was reported.) Alternative Response Categories in Problem-Oriented Policing 1. Concentrating Attention on Those Who Account for a Disproportionate Share of a Problem (25) • Repeat-offender/career-criminal initiatives • Repeat-victimization initiatives • Repeat-location initiatives 2. Connecting With Other Government and Private Services (65) a. Making Referrals to Other Agencies (10) • Counseling and shelter for domestic violence victims • Support services for rape victims • Dispute resolution for landlords and tenants • Counseling and shelter for runaways • Demolition of buildings • Investigations of child abuse/neglect b. Coordinating Police Responses With Other Agencies (33) • Joint monitoring of probationers and parolees • Joint truancy enforcement with schools and juvenile authorities • Joint alcohol licensing inspection and enforcement • Joint driver's license inspection and enforcement c. Correcting Inadequacies in Municipal Services, and Pressing for New Services (22) • Increased garbage collection and forestry services • Increased recreational activities and facilities • Increased enforcement of building code and zoning violations • Improved street lighting • Improved public health services • Improved transportation systems • Creation of refugee assistance facilities • Creation of graffiti removal programs • Creation of detoxification facilities • Creation of job training programs • Creation of system for checking on the welfare of the elderly 3. Using Mediation and Negotiation Skills (5) • Landlord-tenant disputes • Merchant-customer disputes • Neighbor disputes • Labor-management disputes • Gang disputes • Political protests • Negotiations about rules of the street among users in conflict • Domestic disturbances 4. Conveying Information (111) a. To Reduce Anxiety and Fear (8) • Providing accurate and reliable information • Dispelling rumors • Calming victims b. To Enable Citizens To Solve Their Own Problems (13) • Providing instructions on accessing government services c. To Elicit Conformity With Laws and Regulations That Are Not Known or Understood (23) • Explaining liability to liquor vendors • Explaining parking regulations • Explaining American laws to recent immigrants • Explaining liquor laws to juveniles d. To Warn Potential Victims About Their Vulnerability, and Advise Them of Ways To Protect Themselves (20) • Children, about strangers, drugs, sexual assault • Shoppers, about thefts from cars • Elderly, about con artists • Shoppers, about bogus merchandise • Car owners, about auto theft • Homeowners, about burglary • Hotel patrons, about storing valuables and area safety e. To Demonstrate to People How They Unwittingly Contribute to Problems (14) • Contributing to panhandlers f. To Develop Support for Addressing a Problem (30) • Identifying problems about which the public is unaware • Explaining harms of seemingly innocuous offenses g. To Acquaint the Community With the Limitations on the Police, and To Define What the Community Can Realistically Expect of the Police (3) • Limitations on ability to remove undesirable people from public places 5. Mobilizing the Community (30) • Establishment of neighborhood watches • Identification of abandoned vehicles • Promotion of community interaction to reduce fear, mistrust or tension • Installation of telephone notification systems to alert potential victims • Formation of citizen patrols • Solicitation of information on criminal activity 6. Using Existing Forms of Social Control, in Addition to the Community (38) • Parents over children • Teachers over students • Landlords over tenants (residential and business) • Employers over employees • Contractors over subcontractors • Universities over fraternities • Friends over one another • Neighbors over one another • Youth over one another, as members of a club • Banks over account holders • Bar owners over patrons • Motel/hotel owners over guests • Businesses over private security companies • Military commanders over soldiers 7. Altering the Physical Environment To Reduce Opportunities for Problems to Recur (57) • Redesigning buildings • Closing streets or rerouting traffic • Improving lighting • Cleaning up neighborhoods • Changing merchandising layouts • Erecting barriers • Greasing poles and fences • Relocating bus stops • Demolishing buildings • Cleaning graffiti • Towing abandoned vehicles • Removing or altering pay telephones • Installing metal detectors • Demolishing buildings • Using plastic rather than glass receptacles 8. Increasing Regulation, Through Statutes or Ordinances, of Conditions That Contribute to Problems (21) • Establishing minimum standards for locks and lighting, to reduce burglary • Establishing regulatory and fining schemes for false alarms • Establishing specific crime prevention requirements (e.g., two clerks in convenience stores) • Requiring soundproofing in apartment complexes • Restricting merchandising practices that make theft easy • Restricting sale of spray paint to minors 9. Developing New Forms of Limited Authority To Intervene and Detain (15) • Giving police power to detain without charging • Giving police power to make involuntary mental commitments • Giving police power to make involuntary detoxification commitments • Giving police power to conduct involuntary transports to shelters for homeless in danger • Establishing cite-and-release procedures • Securing agency authority from private property owners to enforce trespassing laws 10. Using the Criminal Justice System More Discriminately (124) a. Straightforward Investigation, Arrest and Prosecution (45) • Reactive investigations and arrests • Proactive investigations and arrests b. Selective Enforcement, With Articulated Criteria (22) • Crackdowns (aka zero tolerance) • High-volume traffic enforcement at select locations c. Enforcement of Criminal Laws That, by Tradition, Are Enforced by Another Agency (12) • Merchant fraud • Environmental laws • Building code violations • Consumer protection laws • Immigration laws d. Greater Specification of Behavior That Should Be Subject to Criminal Prosecution or to Control Through City Ordinances (4) • Aggressive panhandling laws • Loitering-for-the-purpose-of-(e.g., prostitution, drug dealing) laws (efforts to refine the law to focus on specific harm, without being overbroad) e. Intervention Without Making an Arrest (26) • Stopping, warning, educating offenders • Giving conspicuous warnings to offenders • Confiscating contraband without charges • Setting up DUI roadblocks f. Use of Arrest Without the Intention To Prosecute (2) • As a means to get drug users into treatment • As a means to get batterers into counseling g. Attachment of New Conditions to Probation or Parole (13) • Mapping offenders out of an area • Prohibiting contact with specific individuals 11. Using Civil Law To Control Public Nuisances, Offensive Behavior and Conditions Contributing to Crime (44) • Liquor licensing • Zoning • Conditional-use permits • Business licenses • Asset forfeiture • Padlock laws • Nuisance abatement • Restraining orders and injunctions • Vehicle impoundment • Health inspection • Fire code inspection • Building code inspection Tables 13 and 14 list the frequency with which each of these general response categories and subcategories was reported in the projects. Table 13 General Response Categories, by Frequency Category No. General Response Category No. of Times Response Was Used 10 Using the Criminal Justice System More Discriminately 124 4 Conveying Information 111 2 Connecting With Other Government and Private Services 65 7 Altering the Physical Environment To Reduce Opportunities for Problems to Recur 57 11 Using Civil Law To Control Public Nuisances, Offensive Behavior and Conditions Contributing to Crime 44 6 Using Existing Forms of Social Control, in Addition to the Community 38 5 Mobilizing the Community 30 1 Concentrating Attention on Those Who Account for a Disproportionate Share of a Problem 25 8 Increasing Regulation, Through Statutes or Ordinances, of Conditions That Contribute to Problems 21 9 Developing New Forms of Limited Authority To Intervene and Detain 15 3 Using Mediation And Negotiation Skills 5 Note: The totals can exceed 100 because the police could use multiple subcategories in any given project. Table 14 Response Subcategories, by Frequency Response Subcategory No. Response Subcategory Description Percentage of Projects in Which Response Was Used 7 Altering the Physical Environment To Reduce Opportunities for Problems to Recur 57 10a Straightforward Investigation, Arrest and Prosecution 45 11 Using Civil Law To Control Public Nuisances, Offensive Behavior and Conditions Contributing to Crime 44 6 Using Existing Forms of Social Control, in Addition to the Community 38 2b Coordinating Police Responses With Other Agencies 33 4f Conveying Information To Develop Support for Addressing a Problem 30 5 Mobilizing the Community 30 10e Intervention Without Making an Arrest 26 1 Concentrating Attention on Those Who Account for a Disproportionate Share of a Problem 25 4c Conveying Information To Elicit Conformity With Laws and Regulations That Are Not Known or Understood 23 2c Correcting Inadequacies in Municipal Services, and Pressing for New Services 22 10b Selective Enforcement, With Articulated Criteria 22 8 Increasing Regulation, Through Statutes or Ordinances, of Conditions That Contribute to Problems 21 4d Conveying Information To Warn Potential Victims About Their Vulnerability, and Advise Them of Ways To Protect Themselves 20 9 Developing New Forms of Limited Authority To Intervene and Detain 15 4e Conveying Information To Demonstrate to People How They Unwittingly Contribute to Problems 14 4b Conveying Information To Enable Citizens To Solve Their Own Problems 13 10g Attachment of New Conditions to Probation or Parole 13 10c Enforcement of Criminal Laws That, by Tradition, Are Enforced by Another Agency 12 2a Making Referrals to Other Agencies 10 4a Conveying Information To Reduce Anxiety and Fear 8 3 Using Mediation And Negotiation Skills 5 10d Greater Specification of Behavior That Should Be Subject to Criminal Prosecution or to Control Through City Ordinances 4 4g Conveying Information To Acquaint the Community With the Limitations on the Police, and To Define What the Community Can Realistically Expect of the Police 3 10f Use of Arrest Without the Intention To Prosecute 2 Note: Frequency is expressed as a percentage because each subcategory was recorded only once per project, even if the police used several different responses of the particular type. As the tables above indicate, the most commonly reported type of response was some use of the criminal justice system. This was not surprising, given the police's longstanding function as enforcers of criminal law. Within that general category, straightforward investigation, arrest and prosecution was the predominant specific response category. In reading the submissions, I had the impression that, in most instances in which the police used straightforward investigation, arrest and prosecution, they used it more as the backdrop to other, more carefully developed interventions than as the primary intervention itself. It served as a reminder, both to the people whose offensive behavior was being addressed, and to the police themselves, that the most restrictive response alternative–arrest and prosecution–remained available if less-restrictive measures failed to correct the behavior. The next most common specific response category within this general category was selective enforcement, with articulated criteria. I was generous in classifying these sorts of responses in that the articulated criteria were not often explicit. This response category contained all references to “zero tolerance” criminal law enforcement; 15 submissions reported “zero tolerance” as a response strategy. Within this general criminal justice system category were two of the least frequently reported specific response types–defining with greater specificity that behavior that should be subject to criminal prosecution or to control through city ordinances, and using arrest without the intention to prosecute. The police rarely reported drafting new legislation to target specific forms of behavior. Rather, they creatively used existing laws to fashion a response strategy. They also rarely reported making arrests without intending to prosecute, most likely because this practice appears, on the surface, to be ethically and legally problematic. In fact, carefully considered, with appropriate safeguards, this response can be effective. Drug courts and domestic violence courts frequently use this strategy to compel offenders to seek professional treatment. The unwillingness either to use this response or to admit to using it may also be attributable to the low level of prosecutor involvement in these problem-solving initiatives. For police to use this response appropriately, prosecutors should be involved in the process. The second most frequently reported general response category was conveying information, followed by connecting with other government and private services. Like the category related to using the criminal justice system, these categories are quite broad, and so naturally encompass many of the police's specific responses. Their breadth is reflected by the subcategories Goldstein developed to better convey what these sorts of responses entail. I found it more instructive to look at the frequency of the subcategories described in Table 14 than at the frequency of the general categories in Table 13. The frequency with which the specific response categories were reported in the submissions better reflects the degree to which the police used responses other than criminal law enforcement. The more specific breakdown reveals that the single most frequently used type of response was altering the physical environment to reduce opportunities for problems to recur. Situational-crime-prevention and crime-prevention-throughenvironmental- design advocates will be heartened by this finding. The police in these problem-solving initiatives demonstrated a willingness and capacity to modify the environment in which problems occurred as an effective means of modifying the behavior of offenders and potential victims. Fifteen of the submissions specifically referred to crime prevention through environmental design. The police also relied heavily on informal social relationships as leverage to modify behavior (reflected in category 6), and use of civil law to regulate conduct (reflected in category 11). Somewhat surprisingly, there were few reports of the use of mediation, either by the police themselves or through trained professional mediators (category 3). Only one submission reported use of a strategy sometimes referred to as “restorative justice.” Restorative justice, like mediation, is a form of alternative dispute resolution that is growing in popularity. I expected it to be more widely used in the projects. While I did not record response types other than those in Goldstein's categories, I did note that the police frequently used increased surveillance as a response to many problems. Increased surveillance includes extra or conspicuous police patrols, video surveillance, police satellite offices in the problem areas, or covert police surveillance. I estimate the police used some form of increased surveillance in at least one-third of the projects, with some form of electronic surveillance (typically, through video cameras) being the most common. The submissions reported as few as one category of response to as many as 15 categories of responses per project. Table 13 lists the distribution of the number of response categories per project. The average number of response categories used per project was five.170 This finding partially confirms the idea that the most effective problem-solving initiatives are those that combine several types of responses. The use of multiple responses does complicate the assessment of effectiveness, as it becomes increasingly difficult to isolate the effective interventions as the number of interventions increases. This tension requires greater consideration by those interested in determining which are the best practices for various types of problems. 170The mean was 5.68, the median was 5 and the mode was 5. Appendix A Table 15 Number of Response Categories, by Number of Projects No. of Response Categories Reported No. of Projects 1 5 2 10 3 16 4 8 5 17 6 10 7 13 8 4 9 4 10 3 11 0 12 3 13 3 14 2 15 2 Total 100 Conclusion I am not the first to analyze the Goldstein awards. In 1997, Ron Clarke, the chair of the award committee, reported on an analysis of the 88 submissions to the 1995 program.171 Clarke and his research assistant explored the type and quality of problem analysis and assessment in greater detail than I did. In those aspects, they found the projects largely to be lacking, a finding I cannot dispute. Clarke classified projects differently than I did in several respects. First, he used somewhat different response-type categories than I did, although they, too, incorporated most of Goldstein's categories. Second, he classified the scope of problems only as either “beat level” or “jurisdiction-wide,” whereas I added an intermediate category. Third, he classified projects as either “problem types” or “place types.” I made no firm distinction, classifying each project in multiple ways. Finally, he classified responses as either “enforcement” or “situational,” a classification I did not make. 171Clarke reported his findings in an appendix to his report to the National Institute of Justice, “Problem-Oriented Policing and the Potential Contribution of Criminology,” Feb. 26, 1997, and expanded on them in a chapter in Problem- Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work, published by PERF in 1998. Several other studies have analyzed various collections of problem-solving projects, including a study of POP projects in the San Diego Police Department (Capowich and Roehl 1994), and a study of POP projects in the Leicestershire and Cleveland, England, police forces (Leigh, Read and Tilley 1998). While the precise methods and categorization schemes of the studies vary, many of the general findings are consistent. Together, the studies provide insights into the actual practice of problem- oriented policing. Many of Clarke's findings are consistent with mine, although he expressed more disappointment in the quality of the projects than do I.172 (This may be partly because he analyzed only one year's worth of submissions, including many that did not survive an initial quality screening. I analyzed only the best submissions from all years.) Clarke found the same average number of responses used per project as I did–five.173 He found similar patterns in the frequency of response types–frequent use of the criminal justice system, coordination with other government agencies and private services, and provision of information. Interestingly, he found exactly the same percentage of instances in which the police altered the physical environment to reduce opportunities for problems to recur (57%), and similar levels of the use of surveillance. One must put both my analysis and Clarke's in their appropriate context. It is easy to find deficiencies among the projects when they are being compared to an ideal model of problem-oriented policing. What is more remarkable, in my opinion, is the high level of dedication, innovation and apparent effectiveness demonstrated by the police officers who undertook these projects. They confirm for me the real potential that lies in the problem-oriented approach to policing, an approach that, after all, is a mere 20 years old. Each year, one or two projects are designated as winners, and several others as “finalists” or “honorable mentions.”174 There are many interesting and high-quality projects, among both those formally recognized and those not so recognized. Each judge has his or her favorites, and although the program coordinators report that the judges' scores are becoming increasingly consistent, there will always be a degree of personal preference in the judging.175 Every one of the 100 projects I reviewed has something interesting and valuable to offer readers, and collectively, as well as individually, these projects make an important contribution to the developing body of knowledge about effective police practice. 172Clarke and I integrated some of our findings in a chapter in a volume on problem-oriented policing (Scott and Clarke 2000). 173Capowich and Roehl (1994) found that San Diego police officers used an average of seven responses per problem-solving project. Their counting rules were not necessarily the same as either mine or Clarke's, but their result is generally consistent with ours. They, too, found heavy use of environmental redesign and informal social control as response strategies. 174In the first few years of the program, projects were divided into “individual” and “team” projects, a distinction that has since been abolished. The “honorable mention” designation of early years has been replaced with a “finalist” designation. 175Among my personal favorites over the years are the Blue Hole Park project (Georgetown, Texas, 1995); the Barrow temperance project (North Slope Borough, Alaska, Department of Public Safety, 1995); the New Helvetia and River Oaks project (Sacramento, Calif., Police Department, 1996); the Elite Arcade project (Delta, British Columbia, Police Department, 1997); the domestic violence revictimization prevention project (Fremont, Calif., Police Department, 1997); the day laborer project (Glendale, Calif., Police Department, 1997); the street cruising project (Santa Ana, Calif., Police Department, 1997); Operation Cease-Fire (Boston Police Department, 1998); the Transient Enrichment Network (Fontana, Calif., Police Department, 1998); “Operation Hot Pipe, Smokey Haze and Rehab” (San Diego Police Department, 1998); and “Street Sweeping, Broadway Style” (Green Bay, Wisc., Police Department, 1999). Appendix B: A Partial List of Problem-Focused Literature The following is a partial list of problem-focused publications that were published either by the U.S. Department of Justice, Office of Justice Programs (OJP), or by police research organizations, with funding from OJP agencies. By the term, “problem-focused,” I mean only that the subject matter of the publication is a substantive community problem, not that it was necessarily a product of Goldstein's model of problem-oriented research. Some of these publications describe actual problem-oriented policing efforts; some merely provide information about the nature and scope of a problem, without assessing any intervention efforts. It is not a comprehensive list, but rather reflects a review of recent publication lists put out by these agencies. The list is meant to illustrate, in a general way, the sort of research publications that reflect the kind of substantive focus that Herman Goldstein advocates in problem-oriented policing. U.S. Department of Justice-Sponsored Bureau of Justice Assistance Addressing Community Gang Problems: A Model For Problem-Solving Addressing Community Gang Problems: A Practical Guide The BJA Firearms Trafficking Program: Demonstrating Effective Strategies To Control Violent Crime Developing a Strategy for a Multiagency Response to Clandestine Drug Laboratories Strategies for Reducing Homicide: The Comprehensive Homicide Initiative in Richmond, California National Institute of Justice Arrestees and Guns: Monitoring the Illegal Firearms Market Batterer Programs: What Criminal Justice Agencies Need To Know Confronting Domestic Violence: A Guide for Criminal Justice Agencies Controlling Drug and Disorder Problems: Oakland's Beat Health Program Crime, Grime, Fear, and Decline: A Longitudinal Look Crime in the Schools: Reducing Conflict With Student Problem-Solving The Crime of Stalking: How Big Is the Problem? Crime Prevention Through Environmental Design in Parking Facilities The D.A.R.E. Program: A Review of Prevalence, User Satisfaction and Effectiveness “Designing Out” Gang Homicides and Street Assaults Evaluation of Violence Prevention Programs in Middle Schools The Expanding Role of Crime Prevention Through Environmental Design in Premises Liability Fraud Control in the Health Care Industry: Assessing the State of the Art Juvenile Gun Violence and Gun Markets in Boston The Kansas City Gun Experiment Modern Policing and the Control of Illegal Drugs: Testing New Strategies in Two American Cities Police Problem-Solving Strategies for Dealing With Youth and Gang-Related Firearms (ongoing study) The Police Response to Gangs: A Multisite Study (ongoing study) Police Response to Special Populations (Handling the Mentally Ill, the Public Inebriate and the Homeless) Policing Drug Hot Spots Preventing Gang- and Drug-Related Witness Intimidation Proceedings of the Homicide Research Working Group Meetings, 1997 and 1998 Reducing Crime and Drug Dealing by Improving Place Management: A Randomized Experiment Reducing Violent Crimes and Intentional Injuries Revictimization: Reducing the Heat on Hot Victims Solving Crime Problems in Residential Neighborhoods: Comprehensive Changes in Design, Management and Use Stalking in America: Findings From the National Violence Against Women Survey Threat Assessment: An Approach To Prevent Targeted Violence Trends, Risks and Interventions in Lethal Violence: Proceedings of the Third Annual Spring Symposium of the Homicide Research Working Group Understanding and Preventing Violence Violence Among Middle School and High School Students: Analysis and Implications for Prevention Office of Juvenile Justice and Delinquency Prevention Environmental Approaches to Reducing Underage Drinking Strategies to Reduce Underage Alcohol Use: Typology and Brief Overview Promising Strategies To Reduce Gun Violence Safe Start–Child Development–Community-Oriented Policing Police Executive Research Forum A Time for Dignity: Police and Domestic Abuse of the Elderly Dispute Resolution and Policing: A Collaborative Approach Toward Effective Problem-Solving Fighting Fear: The Baltimore County COPE Project Finding and Addressing Repeat Burglaries Illegal Money Laundering: A Strategy and Resource Guide for Law Enforcement Agencies Improving the Police Response to Domestic Elder Abuse Mental Illness: Police Response Police and Drug Control: A Home Field Advantage Police Antidrug Tactics: New Approaches and Applications The Police Response to Gangs: Case Studies of Five Cities The Police Response to People With Mental Illnesses: Trainer's Guide The Police Response to People With Speech and Hearing Disabilities: Trainer's Guide The Police Response to the Homeless: A Status Report Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work Problem-Solving: Problem-Oriented Policing in Newport News Special Care: Improving the Police Response to the Mentally Disabled Strategies for Success: Combating Juvenile DUI Tackling Drug Problems in Public Housing Take Another Look: Police Response to People With Seizures and Epilepsy Toy Guns: Involvement in Crime and Encounters With Police Under Fire: Gun Buy-Backs, Exchanges and Amnesty Programs Police Foundation Arresting Shoplifters: An Experiment in Lesser Crimes and Sanctions Creating the Multidisciplinary Response to Child Sex Abuse: Implementation Guide Domestic Violence and the Police: Studies in Detroit and Kansas City Drug Enforcement in Public Housing: Signs of Success in Denver Kansas City Preventive Patrol Experiment: A Summary Report Minneapolis Domestic Violence Experiment Newark Foot Patrol Experiment The Police and Interpersonal Conflict: Third-Party Intervention Approaches Reducing Fear of Crime in Houston and Newark: A Summary Report Spouse Abuse Research Raises New Questions About Police Response to Domestic Violence Other Publications Listed by the National Criminal Justice Reference Service Case Studies of Community Antidrug Efforts Mental Illness and Violent Crime National Evaluation of G.R.E.A.T. A Policymaker's Guide to Hate Crimes Stopping Hate Crime: A Case History From the Sacramento Police Department British Home Office-Sponsored Alcohol and Crime: Taking Stock Armed Robbery: Two Police Responses Arresting Evidence: Domestic Violence and Repeat Victimisation Burglary Prevention: Early Lessons From the Crime Reduction Programme Clubs, Drugs and Doormen Hot Products: Understanding, Anticipating and Reducing Demand for Stolen Goods Keeping Track? Observations on Sex Offender Registers in the U.S. Missing Presumed…? The Police Response to Missing Persons The Nature and Extent of Construction Plant Theft The Nature and Extent of Light Commercial Vehicle Theft New Heroin Outbreaks Amongst Young People in England and Wales Opportunity Makes the Thief: Practical Theory for Crime Prevention Police Antidrugs Strategies: Tackling Drugs Together Three Years On Policing Drug Hot Spots Policing Problem Housing Estates Preventing Repeat Victimisation: The Police Officer's Guide Preventing Residential Burglary in Cambridge: From Crime Audits to Targeted Strategies Repeat Victimisation: Taking Stock Tackling Local Drug Markets Tackling Street Robbery: A Comparative Evaluation of Operation Eagle Eye Theft, Stolen Goods and the Market-Reduction Approach: Operation Radium and Operation Heat Vehicle Crime Reduction: Turning the Corner Appendix C: A Summary of Interviews With Selected Problem-Oriented Policing Practitioners and Researchers To inform my writing of this report, I interviewed selected experts with extensive knowledge about, and experience with, problem- oriented policing. I initially identified 44 people who, in my estimation, have a thorough grasp of the concept of problem-oriented policing (hardly a complete list). Over the course of a year, I spoke with 30 of them at varying lengths about problem-oriented policing, and conducted a structured interview with 12 of them. The following are excerpted responses to my questions regarding problem-oriented policing, quoted from the following: John Eck: Currently an associate criminal justice professor at the University of Cincinnati. Formerly the evaluation coordinator for the Washington, D.C./Baltimore High-Intensity Drug Trafficking Area, vice president of the Crime Control Research Corp. and associate director for research at the Police Executive Research Forum (PERF). Has published extensively on problem-oriented policing. Bob Heimberger: Currently a sergeant and special assistant to the chief of police in the St. Louis Metropolitan Police Department. Has provided extensive training and technical assistance in problem- oriented policing, and has directed problem-oriented policing initiatives in St. Louis. David Kennedy: Currently a senior researcher at the Program in Criminal Justice Policy and Management at the John F. Kennedy School of Government, Harvard University. Oversees the National Institute of Justice's (NIJ's) Strategic Approaches to Community Safety Initiative. Has published on problem-oriented policing, and directed the research component of Operation Cease-Fire, a collaborative effort with the Boston Police Department to reduce youth gang violence. Gloria Laycock: Currently the director of the Jill Dando Institute of Crime Science, School of Public Policy, University College London and recently a visiting fellow at NIJ. Formerly head of the Home Office Policing and Reducing Crime Unit in London. Has conducted extensive research on police practices, particularly crime prevention strategies and tactics, and has edited the Home Office's police research papers for a number of years. Nancy McPherson: Currently an administrator at the Portland, Ore. Police Bureau and formerly the director of the Community and Information Services Bureau for the Seattle Police Department, the manager of neighborhood policing for the city of San Diego, and a field advisor to the San Diego Police Department for PERF. Has provided extensive training and technical assistance in problem- oriented policing. Received the Gary P. Hayes Award for innovation in policing in 1999. Dennis Nowicki: Currently the director of the Center for Public Service and Leadership at Pfeiffer University in Charlotte, N.C. Formerly the chief of police at the Charlotte-Mecklenburg Police Department and the Joliet, Ill., Police Department. Has published on problem-oriented policing, and has implemented problem-oriented policing practices in police agencies. Dan Reynolds: Currently a deputy chief at the Savannah, Ga., Police Department. Has provided extensive training and technical assistance in problem-oriented policing, and has implemented problem-oriented policing practices in a police agency. Rana Sampson: Currently a police consultant, operating as Community Policing Associates in San Diego. Formerly the director of public safety at the University of San Diego, a senior researcher at PERF, and a sergeant with the New York City Police Department. Has published on problem-oriented policing, and has implemented problem-oriented policing practices in a public safety agency. Malcolm Sparrow: Currently a professor of practice at Harvard University's John F. Kennedy School of Government. Formerly a detective chief inspector in the British Police Service. Coauthor of Beyond 911: A New Era for Policing. Author of Imposing Duties: Government's Changing Approach to Compliance, and The Regulatory Craft: Controlling Risks, Solving Problems and Managing Compliance (forthcoming). Darrel Stephens: Currently the chief of police at the Charlotte- Mecklenburg Police Department. Formerly the city administrator of St. Petersburg, Fla.; the chief of police at the St. Petersburg Police Department; the executive director of PERF; and the chief of police at the Newport News, Va., Police Department and the Largo, Fla., Police Department. Has published on problem-oriented policing, and has implemented problem-oriented policing practices in police agencies. 1. Do you still consider problem-oriented policing a viable approach to improving police service? John Eck: Yes, there really is no alternative. The problem-oriented policing label will come and go, but if the police don't embrace problem-oriented policing, they risk becoming a marginalized public service agency, as has happened to other agencies. The police need to get on board with problem-oriented policing, or some of their functions may be taken over by other sectors, leaving only a narrow role of report taking and patrolling. David Kennedy: Yes. Seemingly intractable crime and disorder problems remain that way because we continue to do the same things that don't work. Sustained thinking works. The basic framework of the President's Crime Commission of 1968 remains in place today–the use of the criminal justice system as the main crime control mechanism. Most major reform efforts are merely efforts to improve the functioning of that system. The criminal justice system may be suited for individual justice, but it is not well-suited for reducing crime. Gloria Laycock: Yes. I've seen it work, so, de facto, it's viable. We now know it's deliverable at some level in policing. Whether that means it's deliverable across a whole agency, as I originally thought it would be, I'm not sure, at least not in a short time frame. Dennis Nowicki: Yes, definitely. In some ways, all the attention given to the “problem-oriented policing movement” is much to do about nothing. Despite the fanfare, problem-oriented policing is really just a common-sense approach to policing. I don't see any other way of doing our business. It results in a much clearer purpose for the activities police officers should be and are engaged in. Most social services should be done this way. Dan Reynolds: The problem-solving process is universal, and has been throughout time. It's logical and rational. It should always have been used in policing. The education system perhaps has not done a good job inculcating problem-solving as a mental process. The sense of emergency in policing has crowded out our capacity to think about problems in the long term. I am seeing changes in the style of discussion in policing, even in our staff meetings. We talk in terms of problem-solving all kinds of issues. Rana Sampson: Yes. Problem-oriented policing is not a panacea to all public safety problems, but it is more effective than what we've been doing. Research has demonstrated the success of using analysis to help the police develop and use less-blunt instruments for addressing problems–if not for the total elimination of problems, at least to reduce them. Problem-oriented policing is a larger concept than mere problem-solving. It has tremendous ramifications for the structure of police organizations. Malcolm Sparrow: Yes, absolutely. I see how difficult it is to implement, but I see it as critical to improving police operations, and to controlling a whole range of large public safety problems. The problem-oriented approach has rich potential, but still, little of that potential has been discovered. Darrel Stephens: Absolutely. The rationale underlying it is just as sound as it ever was. Most of police work is repetitive, and the experiences applying a problem-oriented approach to repetitive problems have so far been positive. Problem-oriented policing is more effective than mere reliance on the criminal justice system. Adaptations to the concept have been made. Herman Goldstein's original concept was more centralized. I don't think he saw police officers making direct contributions to problem-solving; he saw them more as resources to analysts and researchers. Newport News and Baltimore County showed that police officers can do problem-solving themselves. I know Herman has had mixed feelings about this. Problem-oriented policing hasn't progressed as much as I would have hoped, though. 2. What aspect of the problem-oriented policing movement over the past 20 years has most impressed you? John Eck: It's sort of like the talking-horse phenomenon: It's not how well it talks, but that it talks at all, that's impressive. The fact that we still talk about problem-oriented policing at all is a testament to its staying power. Many people are still interested in and still struggling with the concept. It has had a lot of subtle influences on police thinking. The police are more likely to look outside their departments for help today, and more interested in using data to make decisions. Bob Heimberger: The national attention the process has received. Problem-oriented policing has given communities hope that their police can be effective. But all the federal money and the popularity of community policing have also hurt the movement; they have led to lots of small, poorly implemented programs that may be determined to have failed. David Kennedy: The string of wins since the early to mid-1980s. There have been good, concrete examples of effective problem- solving that came out of such places as Newport News, Houston and Tampa. Tough problems were addressed using this model. It has always been clear there was juice behind the concept. Gloria Laycock: Its ability to deliver outcomes, bottom-line outcomes such as the reduction of crime. Problem-oriented policing is an idea that was ahead of its time. Twenty years isn't the relevant period of time in the United Kingdom, because Herman Goldstein didn't become relevant until more recently. Engagement with problem oriented policing in the U.K. really came about after the Harvard Executive Sessions on Community Policing, and then, only in the Metropolitan Police Department in London. It was not widespread. Independently, we in the Home Office were developing something called the preventive process, which paralleled the SARA model and had been developed from the early 1980s, although we applied it only to crime prevention. After I read Herman's book, I saw we were doing some of the same things. This reinforced many of our ideas related to crime prevention. The SARA model has been very useful, even though I recognize its drawbacks and limitations. The police like formulas and acronyms that guide them toward action. The SARA model and the crime triangle are now used widely in the U.K. At its best, problem-oriented policing engages police officers at the front end, and gets them excited about their work. It gives them a whole new perspective on their job, such that the job can become exciting instead of routine, and that's important. Some of the junior officers are way ahead of their superiors in their comprehension of the concept. It isn't necessary to have the entire agency embracing problem-oriented policing at this point. One can find good applications of it at the individual level in just about any agency. It's really not at the agency level, but at the individual level where one is impressed by the development of the idea. Nancy McPherson: The concept provides a clear focus on crime and disorder, as opposed to administrative matters or politics. But many police agencies haven't really focused on this. Few police organizations have really made the organizational changes necessary to support problem-solving. Two of the most positive impacts of problem- oriented policing have been the inspiration it has provided to police officers to do good work–they see the positive results from their problem-solving–and community members see it as more than fluff, as an approach that actually makes a difference on problems they care about. Dennis Nowicki: The way the concept has been embraced by the community. Problem-oriented policing is not just a police initiative. In some cities, there is a critical mass of support for this way of policing that didn't exist 20 years ago. Although the concept is not always well- understood by most elected and other government officials, its essence is increasingly clear to neighborhood residents who step forward to work with the police officer. Because of the lack of understanding by politicians, problem-oriented policing is vulnerable; those who co-opt the label and turn it into something else can kill it. The relationships built through working with the community to solve problems have allowed the police to survive some tough times as a result of things like controversial shootings, etc. Dan Reynolds: Developing a relationship with the public, sharing information, working with others. Problem-oriented policing has enabled these things because the problem-solving process requires these things be done in order to gather information. And this process of gathering and sharing information has helped break down some stereotypes the police had of some members of the public. Many police officers began to realize that dispatchers aren't the only or best source of information about what's happening on the street. We are now sharing information with the community more freely than we ever did. Rana Sampson: Problem-oriented policing has made police officers and administrators think more about the substance of their work. That sounds obvious, but we've been so blinded by so many other things that happen in the community, and by administrative responsibilities, that we haven't developed the expertise needed to address problems. Our conventional instruments were so blunt–citations, arrests, etc. The potential is immense. The analysis of problems by police officers in some projects is impressive, and their responses, creative. Herman Goldstein recognized how creative officers are. Problem-oriented policing allows for that creativity; it is no longer just something that is exercised when the sergeant isn't looking. When a police department takes a problem-oriented policing approach, it turns police work upside down by asking whether the current response is working. It calls for a constant reexamination of what we do, including our relationship with the community. So much of police work previously was bluffing. Problem-oriented policing professionalizes policing in a real sense. Problem-oriented policing has the capacity to redefine policing away from the view that it's merely the entry point to the criminal justice system. That's incredibly significant, and recognizing it helps us have a real impact on problems. This hasn't occurred to any great extent yet, but it is beginning. Police chiefs and the U.S. Department of Justice haven't fully engaged with this fact. It also calls for a whole new education of judges and prosecutors about the role of the police. Malcolm Sparrow: The visible successes have been at the beat level, where problem-solving appears as a natural companion to community policing. Compared with other regulatory professions, the police have led the way in the early articulation and implementation of the problem-oriented approach. The police, however, have since run into a specific obstacle, which is their general failure to construct the managerial systems that are required to run problem-solving at higher levels, and as the core of police operations. Darrel Stephens: I'm impressed with even the small proportion of police officers who have made problem-oriented policing part of their work. A lot of good work has been done. The concept is broad-based in some police departments, and the results have been impressive. It's been demonstrated in lots of different departments. The implementation of the concept at present is wide, but not deep. 3. What aspect of the problem-oriented policing movement over the past 20 years has most disappointed you? John Eck: As Gary Cordner has pointed out, the linkages between analysis and response are the weakest. Many people have difficulty conceptualizing information. This problem is not restricted to the police; it's a human problem. Other professions have similar difficulty translating research knowledge into practice. Some efforts, like exploring the proper structure for a police agency, are likely to have only a marginal impact on problem-oriented policing, and on improving police service. I am also very much disappointed with the police profession, national organizations and academia for not advancing problem-oriented policing beyond where we were in 1985. We have many more police agencies and officers involved in problem-solving, and there are many exceptional efforts at addressing problems. But for the most part, no one has taken what was done by Herman Goldstein or by the Newport News Police Department and expanded upon it in any substantial manner. Academics have criticized various ways problem- oriented policing has been implemented, and this is good, but with rare exception, they have not attempted to build a better mousetrap. National agencies have promulgated much of the original work, but have not looked for ways of improving problem analysis. Police agencies have adopted aspects of a problem-oriented approach, but have focused more on the management of the organization than on trying to understand the problems their officers face. Bob Heimberger: The lack of understanding of the concept, including what outcomes are desired. The concept often gets confused with efforts to improve community relations. The follow-up in training programs didn't occur. We needed more resources from funding agencies to help with evaluation. But many auditors from funding agencies knew less about the methodology than the department receiving the funds. Anything can be made to look successful. It has been too easy to get and keep the money. David Kennedy: Since the early 1990s, it has been frustrating that, after 10 years, most problem-solving efforts remained small local efforts by line officers, not intermediate or communitywide sorts of problems. Not much has been done on serious crime, mostly on disorder and fear. There has been little engagement by police management, and little structural change in police agencies. There is a strong need to improve agencies' data gathering and analysis capacities. There is a need to get mid-managers more involved. With the exception, perhaps, of the San Diego Police Department, problem- oriented policing has not become normal policing yet. These frustrations prompted many of the features of the Boston funding proposal to NIJ (Operation Cease-Fire). We wanted to raise the bar. Some of these things are changing a bit, but they remain largely true today. Gloria Laycock: The time frame it takes to get these ideas developed and implemented in police agencies. One of the strengths of the concept is its simplicity, so it's hard to understand why it's so difficult for some people and agencies to do it. Some police agencies are very unimaginative, and that's disappointing. Those who struggle with the concept tend not to appreciate the value of data, the time it takes to make use of it, and the patience required to reflect on the real nature of problems. That's largely a product of police training–they want to get on with things quickly. Police chiefs have been slow to take this up, and they have fewer excuses than line officers. It's hard to get chiefs to concentrate on the concept long enough to make it work; they seem preoccupied with local political problems, which is in itself understandable, but points to a fairly basic problem in the whole system of policing. Problem-oriented policing is fundamentally rational; it is the scientific method applied to policing. It should be self-evident that it is needed. Unfortunately, there isn't enough fundamental rationality in policing. Nancy McPherson: The lip service paid to the concept for political reasons, and lack of a real commitment to the principles of problem- oriented policing. There has not been enough technology developed or implemented to support line people in their problem-solving efforts. Dennis Nowicki: The lack of understanding of what this is all about. Public policymakers and the uninvolved public develop their view of policing from watching TV cop shows. We haven't communicated the core elements sufficiently and consistently enough for them. Our deficiencies in selling the concept mean that even many police officers don't have a common understanding of what is expected of them. Dan Reynolds: I thought we would be doing more problem-oriented policing projects, that there would be more outcomes. Problem- oriented policing is harder to sustain than I imagined, to get beyond the sense that this is a fad. When we started into problem-oriented policing in the early '90s, I thought implementation would take one year. Yet we're still teaching it. We measure our progress, in part, by how many problem-oriented policing projects we've undertaken, though I realize this is a dubious measure. We probably underestimated the resistance of the police culture, and we should have attended more to changing the culture before seeking complete implementation of problem-oriented policing. Reactive policing is so much easier. Police officers are trained to prefer order to disorder, and problem-solving seems, to some officers, to be creating disorder, to be upsetting the balance of things. My expectations of problem-oriented policing have changed over time. I'm not as frustrated about the pace of change now. It's interesting to hear how new these ideas are still to some people, and in some places in the country. Some of us sort of assumed that if we knew this stuff, everyone else did. That just isn't true. In a strange way, perhaps the complexity of the drug problem has forced the police to become more sophisticated in analyzing the related problems and looking for new solutions. It certainly has fueled the money that has gone into training, research and technology, much of which has supported problem-oriented policing. It may have been a blessing in disguise for the police profession. Rana Sampson: Most chiefs have not taken the time to understand Goldstein's concept. Most chiefs probably haven't even read Goldstein's book once, and it is worth reading many times. So police organizations haven't invested in learning it. Many have taken the easier path of community policing projects and Compstat. Malcolm Sparrow: The practice of problem-solving seems to have stalled, partly because it has not been sufficiently distinguished from its frequent companion (community policing), and has therefore been viewed by many police agencies as a question of professional style for beat-level officers, and not a central challenge for the departmental management structure. So larger problems tend not to get addressed in a problem-oriented fashion. Problem-solving can be done at the field level without making systematic or structural changes to the police organization, or to its various administrative and managerial systems. There has been a tendency to simplify and reduce the problem-solving concept, and to focus on particular innovations rather than the systems and managerial behaviors that produced them. This tendency is by no means unique to the police. Some problems that the police must address don't lend themselves to the sort of community partnership responses envisioned by community policing, and for those kinds of problems, problem- solving has been less well-developed. Those problems nonetheless are amenable to problem-solving interventions. Problem-solving seems to happen more naturally at both the bottom and top levels of police organizations. But it's not satisfactory that it happen at only these levels, since most problems we care about are intermediate-size problems, calling for intermediate-level responses, organized and coordinated within the middle layers of police organizations. Darrel Stephens: I'm fairly disappointed it hasn't become more of what police officers do on a daily basis. Most problem-solving that gets done is additional activity. Both the community and the police see the criminal justice response as the primary one to crime problems, and other matters are viewed as mere annoyances. The public has maintained a high level of confidence in the criminal justice system. Police work is still defined narrowly, in spite of efforts to expand it. A few chiefs have gone very far in implementing problem-oriented policing, but most have been more limited in their efforts. It is still not mainstream policing. It remains a small proportion of the total investment in policing. Community leaders still cling to the idea that government and the police can and will solve all their problems, and they tend not to want to take responsibility for solving their own problems. There have been a few exceptional efforts. On the whole, most city managers and mayors remain ignorant of this concept. In some places, good problem-solving occurs outside of, and in spite of, the local government framework. In such places, there is little political buy-in. Buy-in tends to be tied to the energy and commitment of the police chief, not the elected officials. Problem-oriented policing likely will remain concentrated among certain kinds of police chiefs, and in certain communities. A good economy has relieved some of the pressure to do policing differently. There is less fiscal pressure on the criminal justice system. The New York experience has been detrimental to the concept; it's had a bigger impact than I imagined it would. They are doing some analysis of problems in New York, but then applying conventional responses–presence, pressure, intimidation by the police. 4. What information technology have you seen that has enhanced problem-oriented policing? John Eck: There is lots of information technology, now that industry has discovered marketable products. Mapping software is abundant. A lot of this information technology will not be particularly helpful, but some will be. I can't predict where this development will go. I love maps and feel that computer mapping can improve the way police address problems. However, the major barriers to improving problem- oriented policing are not technological. They have to do with how we conceptualize crime and disorder problems. The risk with technology is that it often allows us to continue doing what we always did poorly, rather than spurring us to new ways of seeing things. Bob Heimberger: The websites of the COPS Office and the Community Policing Consortium made research and the exchange of information easier. Laptops in police cars raised the awareness that police officers need data on the streets. But there has been too much flying by the seat of the pants in developing this technology; there is no good central repository of information. Gloria Laycock: Computer technology like computerized mapping can help the police manage the enormous volumes of data they possess. This has been the best contribution of technology. Mapping, however, is actually a bit of a red herring. It can even be unhelpful. I worry that people are becoming obsessed with maps and their pretty colors, without thinking much about what information they contain or what can be learned from them. The technology itself becomes what is fascinating, rather than the knowledge to be gained from it. So technology can at times inhibit the development of problem-oriented policing, because it stops people from thinking. Nancy McPherson: Crime mapping, at least to get people focused on hot spots and series-of-crime analysis. The Seattle Police Department is now trying to get some recently declassified Central Intelligence Agency technology that will enable things like link analyses. Dan Reynolds: The Internet, with its many websites that allow us to share and gather information about common problems, and potential strategies and solutions. The National Criminal Justice Reference Service's website, for example, is a good source of information. Every police officer in our department now has access to computer technology–if not for all functions, at least to be able to gather data from our records systems and from other information databases. Malcolm Sparrow: I don't think information technology has been used the way it needs to be. Geographic mapping has helped in hot- spot analysis, as have software programs like STAC (Spatial and Temporal Analysis of Crime), but they are specific and limited forms of analysis. We need a much broader array of analytical techniques, built on top of flexible data access systems such as data warehouses. I describe in greater detail the needs and the possibilities for applying information technology to problem-solving in my book Imposing Duties, particularly chapter 4. Darrel Stephens: There have been lots of attempts, but not much result. Computerized mapping and data systems to allow police to assess calls for service have been useful. There's been far more promise than delivery. 5. How well has research supported problem-oriented policing? John Eck: This is highly variable for individual researchers and institutions. Some are very good. Few researchers are genuinely interested in applied policing, though. It's difficult for a young academic to balance a research agenda with a firm grounding in reality. Research contributions likely will have to come from the academic community, or they won't come at all. Private industry has a stronger history of engaging in sound in-house research than does the public sector, including the police. We need to preserve diversity of thinking in policing. To some degree, the federal funding programs have promoted homogeneity of thought, and that's something to be careful of. We shouldn't expect brilliant ideas to come out of government. It isn't what it does best. Tying federal money to ideas about policing tends to promote fads more so than it promotes solid ideas. Smaller experiments with a variety of ideas that others can adopt voluntarily tend to work better. Bob Heimberger: The real influence has come from individuals rather than research institutions. Bill Spelman and John Eck's writings have been very influential. Lots of researchers are engaged in the subject more for personal and financial interests than to improve policing. David Kennedy: First, one must distinguish the research on problem- oriented policing from the research in problem-oriented policing. There has been some good operational research on substantive problems done out of universities such as Rutgers. Among the SACSI sites funded by NIJ, most are new to this methodology. I don't see much research relevant to problem-oriented policing coming out of some of the other top criminology schools, however. Among researchers, this sort of work is not professionally valued. This is true more for cultural than principled reasons. Policy analysis is looked down on by the social sciences. There is no place to publish it in the respected journals. Some see this type of research as too basic. It requires field research to gather relevant data, rather than relying on existing data sets. There remain cultural barriers between scholars, cops and the community. Doing problem-oriented policing well requires a level of understanding of problems and agencies' capacities that is often lacking. This is needed for researchers to understand what value might realistically be produced. What many researchers know how to do isn't always useful to problem-oriented research. Basic social science training and the social science mindset are about two fundamental questions: How did this happen?–which explores causal factors–and, What does this look like?–which is descriptive. Problem- solving research is about useful interventions. Gloria Laycock: I'd like to think the work we've done in the Home Office in the U.K. has been useful, even though all of it wasn't always done under the rubric of problem-oriented policing. In the United States, NIJ and the COPS Office have led the way, though in those agencies, problem-oriented policing has become muddled with community policing. They've spent lots of money in the area, but haven't yet produced the results out the other end. PERF has supported the POP conference, but I think they could have done a lot more. 6. What are the next steps the profession should take to advance problem-oriented policing? John Eck: If the public demands better police service, the police will likely provide it. But demand is a very localized thing, and demand isn't always expressed in a very clear or sophisticated fashion. Where there are good police leaders, they can help translate that demand into better policing. The police need to take criminology more seriously, and criminologists need to shift some of their focus away from explaining why people become criminals, and toward reducing criminal opportunities. Environmental criminology has a lot to offer the police. Bob Heimberger: We need to improve our understanding of the concept, including the management issues for police commanders. I'm hesitant, however, to say just provide more training. We need closer follow-up by funding and research agencies to keep initiatives on track. There needs to be more research on substantive community problems to improve the police understanding of complex problems. There remains a lot of misunderstanding among police about problems like homicide. More money for more police on the streets isn't necessarily the answer. David Kennedy: The profession needs to promote examples and models of what it wants done. So it needs a deliberate process of supporting, assisting and guiding this work. SACSI is such an effort, and the COMPASS program incorporates information technology. We need to create the next round of stories. We should create a list of substantive problems we want addressed, like domestic violence, robbery, burglary, and child abuse, and then target the best agencies and researchers in the problem-oriented policing field, and provide them with lots of coaching. We need to think about the desirable institutional features of police departments, like their internal research capacity and their problem identification abilities. We need a journal to publish this kind of work. The Journal of the American Medical Association started out as doctors writing stories. There has been a large failure of leadership among police executives. Gloria Laycock: Documentation of what works is needed, teasing out the specific mechanisms by which certain interventions to certain problems have proven effective. A more systematic training effort is needed to get the concept into the consciousness of police officers. There also needs to be a valid and reliable performance regime developed. Measuring what matters is crucial here. I do not think the current trend in focusing at the individual officer level will work. I would look at the precinct or district level, and hold district commanders accountable for reducing the number or seriousness of hot spots in their areas. “Solving” trivial problems–single issues–would be discouraged. Nancy McPherson: Focus on accountability at the commander level. While New York's Compstat and Los Angeles' FastTrack models are not the direction the Seattle Police Department wants to go, we are looking at command accountability processes. We need processes that get beyond just Part I crimes and the public humiliation of commanders. Organizations need to carefully craft these processes; they can take them down the wrong path quickly. Line officers often say they are not supported by higher-ups, so this needs serious attention. We need to get police investigators more engaged in problem-solving. Performance evaluations should emphasize collegial, coaching feedback. We need improved training for police supervision. We should promote integrated criminal justice systems, including linked records and data with courts and other police agencies. Dennis Nowicki: We need to aggressively educate the public on what effective policing has come to be. We need to give police officers and the community the information tools they need to do problem- oriented policing. All local governments need to develop a problem focus. This evolution is underway; we're starting to see neighborhood services, prosecutors, and probation and parole get involved. We need problems to become the basic units of work in policing and other city services, and to make that idea real. We need organizational structures and systems to become better aligned with the problem-oriented policing concept. Remember, this process of change toward problem- oriented policing has not been sustained for 20 years within any one agency. There is lots of inconsistent change within police departments. We don't have critical mass yet throughout the country, but it does exist in a few departments. The organizational structure of police agencies tends to reflect the individual personalities of the chiefs. The potential end of COPS Office funding will create a small crisis in the field; it will give some people and agencies an excuse not to engage in problem-oriented policing. This may not happen, but if it does, it could be devastating. Dan Reynolds: There is a need to focus on reorganizing police departments more substantially. We should consider creating units and functions we might not have had before. We need to continue to decentralize police operations to enable officers to work more closely with the community. We need stronger information technology units within police departments. We need to exploit technological means by which we can get information out to the public about crime and problems, to solicit their support and assistance. Reverse-911 technology, through which we can place directed automated phone calls to residences or businesses about specific crimes or problems, is an example of this sort of technology. Rana Sampson: There is a need to do things on several different levels. We need more victimization research. We need to collect a body of research on how to affect particular problems. The National Institute of Justice needs to rethink how to research problems and get information out in a timely fashion. There are people who can do this. The federal government needs to invest in situational crime prevention research, and make it accessible to police officers. Ninety- nine percent of police officers have never seen a Research in Brief. There is a need to better understand the research audience–line police officers. We need more written about how the police are not just the entry to the criminal justice system. Police departments need to invest in their internal research capabilities. Smaller agencies should pool resources toward this end. Research units can't and shouldn't do all the work in problem-oriented policing, but they should do problem-solving of a different type than beat-level problem-solving, and should engage in some experimentation. There is a need to do work on situational crime prevention and victims. We don't even know how many crimes occur, because most aren't reported. We need to take a broader view of police responsibilities, in addition to solving reported crimes. There is a need for more research into criminals' perceptions of their work–like Scott Decker and Richard Wright's work on burglars. We need better research on deterrence and opportunity reduction, picking up on the work of Marcus Felson and Ron Clarke. We need to better understand criminogenic places. Urban planners know very little about crime–police and urban planners need to get better connected. University faculty need to get more engaged in local crime problems. Police, social service, code enforcement, and mental health agencies must learn to share information, especially regarding the repeat nature of people and places. Cities need to develop an understanding of, and an explicitness about, the responsibilities of businesses that consume a disproportionate volume of police services–like convenience stores and shopping malls–social responsibilities of companies not to create crime opportunities. There is a need for more and better training for crime analysts, to make them more than people who input data, and better training for officers to analyze the data. Malcolm Sparrow: We need to prescribe more definitively what administrative arrangements work best to support problem-oriented policing above the beat level. I would like to see problem-oriented policing draw more from the fields of intelligence analysis. The concept of intelligence-led policing, which one finds primarily in the United Kingdom and Canada (and emerging in some places in the United States), represents some movement toward organizing police operations around analysis of the issues they face, but this movement still remains somewhat disconnected from the development of problem-oriented policing. The International Association of Law Enforcement Intelligence Analysts, led by Marilyn Peterson, is trying to connect problem- oriented policing with intelligence-led policing. I have been studying the operations and management of a number of other risk control operations, such as environmental protection, customs, occupational safety and health, and tax collection. Like the police, these agencies are essentially regulatory enterprises. Unfortunately, the police don't view themselves as “regulators,” so they miss some opportunities to connect with and learn from a broad range of colleagues in other regulatory bodies. I focus on those parallels in The Regulatory Craft [Sparrow, forthcoming]. The central thesis of that book is that problem-solving has enormous potential across the entire regulatory side of government, including policing, and, for reasons we are only beginning to understand, this extraordinary potential has not yet been realized. Darrel Stephens: I'm not sure. The concept will advance among those who have had positive experiences with problem-oriented policing. Without external pressure, though, it may not happen, especially if doing conventional police work is easier. When the economy turns down, and crime goes back up, there may be more pressure to adopt the problem-oriented approach. External pressure is more powerful than internal desire. References Alpert, G., and M. Moore (1998). “Measuring Police Performance in the New Paradigm of Policing.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Babcock, L. (1996). “The Evolution of Community Policing in Newport News, Va.” In Themes and Variations in Community Policing: Case Studies in Community Policing. Washington, D.C.: Police Executive Research Forum. Barrett, G. (1996). “Community-Oriented Problem-Solving in Santa Barbara, Calif.” In Themes and Variations in Community Policing: Case Studies in Community Policing. Washington, D.C.: Police Executive Research Forum. Bayley, D. (1991). “Book Review: Problem-Oriented Policing.” American Journal of Police 10(4). Bazemore, G., and A. Cole (1994). “Police in the 'Laboratory' of the Neighborhood: Evaluating Problem-Oriented Strategies in a Medium-Sized City.” American Journal of Police 13(3). Bennett, T. (1994). “Community Policing on the Ground: Developments in Britain.” In D. Rosenbaum (ed.), The Challenge of Community Policing: Testing the Promises. Thousand Oaks, Calif.: Sage Publications. Berry, G. (1999). “Merseyside Police: Initial Evaluation of the Problem-Solving Approach.” Report to the Merseyside Police. Geoff Berry Associates. Berry, O. (1996). “Community Policing in Philadelphia.” In Themes and Variations in Community Policing: Case Studies in Community Policing. Washington, D.C.: Police Executive Research Forum. Bichler, G., and R.V. Clarke (1997). “Eliminating Pay Phone Toll Fraud at the Port Authority Bus Terminal in Manhattan. In R. Clarke (ed.), Situational Crime Prevention: Successful Case Studies (2nd ed.). Albany, N.Y.: Harrow and Heston. Bittner, E. (1990). “Some Reflections on Staffing Problem-Oriented Policing.” American Journal of Police 9(3). Block, C.R., M. Dabdoub and S. Fregly (eds.) (1995). Crime Analysis Through Computer Mapping. Washington, D.C.: Police Executive Research Forum. Boston Police Department (1998). “Operation Cease-Fire.” Submission for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. Brame, R., and A. Piquero (1998). “The Police and the Control of Crime and Disorder.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Bratton, W. (1995). “The New York City Police Department's Civil Enforcement of Quality-of-Life Crimes.” Journal of Law and Policy 3. Bratton, W. with P. Knobler (1998). Turnaround: How America's Top Cop Reversed the Crime Epidemic. New York: Random House. Brodeur, J. (1998a). “Introduction.” In J. Brodeur (ed.), How to Recognize Good Policing: Problems and Issues. Washington, D.C.: Police Executive Research Forum and Sage Publications. — (1998b). “Tailor-Made Policing: A Conceptual Investigation.” In J. Brodeur (ed.), How To Recognize Good Policing: Problems and Issues. Washington, D.C.: Police Executive Research Forum and Sage Publications. Bureau of Justice Assistance (1993a). Problem-Oriented Drug Enforcement: A Community-Based Approach for Effective Policing. Washington, D.C.: U.S. Department of Justice. — (1993b). A Police Guide to Surveying Citizens and Their Environment. Washington, D.C.: U.S. Department of Justice, Bureau of Justice Assistance. Buslik, M., and M. Maltz (1998). “Power to the People: Mapping and Information Sharing in the Chicago Police Department.” In D. Weisburd and T. McEwen (eds.), Crime Mapping and Crime Prevention. Crime Prevention Studies, Vol. 8. Monsey, N.Y.: Criminal Justice Press. Bynum, T. (forthcoming). Using Analysis for Problem-Solving: A Guidebook for Law Enforcement. Washington, D.C.: U.S. Department of Justice, Office of Community-Oriented Policing Services. Capowich, G., and J. Roehl (1994). “Problem-Oriented Policing: Actions and Effectiveness in San Diego.” In D. Rosenbaum (ed.), The Challenge of Community Policing: Testing the Promises. Thousand Oaks, Calif.: Sage Publications. Clarke, R.V. (ed.) (1992). Situational Crime Prevention: Successful Case Studies. New York: Harrow and Heston. — (ed.) (1993). “Editor's Introduction.” In Crime Prevention Studies, Vol. 1. Monsey, N.Y.: Criminal Justice Press. — (ed.) (1994a). Crime Prevention Studies, Vol. 2. Monsey, N.Y.: Criminal Justice Press. — (ed.) (1994b). Crime Prevention Studies, Vol. 3. Monsey, N.Y.: Criminal Justice Press. — (ed.) (1996). Preventing Mass Transit Crime. Crime Prevention Studies, Vol. 6. Monsey, N.Y.: Criminal Justice Press. — (1997a). “Problem-Oriented Policing and the Potential Contribution of Criminology.” Report to the National Institute of Justice. — (1997b). Situational Crime Prevention: Successful Case Studies (2nd ed.). Guilderland, N.Y.: Harrow and Heston. — (1998). “Defining Police Strategies: Problem-Solving, Problem- Oriented Policing and Community-Oriented Policing.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Clarke, R.V., and M. Felson (eds.) (1993). Routine Activity and Rational Choice: Advances in Criminological Theory, Vol. 5. New Brunswick, N.J.: Transaction Books. Clifton, W. (1987). “Convenience Store Robberies in Gainesville, Florida: An Intervention Strategy by the Gainesville Police Department.” Photocopy of Gainesville Police Department internal report. Cole, D. (1999). “Discretion and Discrimination Reconsidered: A Response to the New Criminal Justice Scholarship.” The Georgetown Law Journal 87. Coles, C., and G. Kelling (1999). “Prevention Through Community Prosecution.” The Public Interest 136 (Summer). Community Policing Consortium (1999). “Assessing and Evaluating: An Inside Look at the Westminster Police Department.” Community Policing Exchange, Phase VII, No. 29 (November/December). Cordner, G. (1998). “Problem-Oriented Policing vs. Zero Tolerance.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Cosgrove, C., and M.A. Wycoff (1999). “Investigations in the Community Policing Context.” In C. Solé Brito and T. Allan (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work, Vol. 2. Washington, D.C.: Police Executive Research Forum. Couper, D., and S. Lobitz (1991). Quality Policing: The Madison Experience. Washington, D.C.: Police Executive Research Forum. Das, D.K. (1992). Some Recent American and British Contributions to Police Literature. Book review. Justice Quarterly 9(4) (December). Diulio, J. (1990). “Book Review: Problem-Oriented Policing.” Journal of Policy Analysis and Management 10(3). Eck, J. (1997). “Preventing Crime at Places.” In Preventing Crime: What Works, What Doesn't, What's Promising. Washington, D.C.: U.S. Department of Justice, Office of Justice Programs. — (1998). “A Problem-Solving Manual for the Police and the Public.” Unpublished draft, Nov. 1. — (1999). “Problem-Solving Detectives: Some Thoughts on Their Scarcity.” Paper prepared for Leadership Sessions to Support Problem-Oriented Policing. Eck, J., and W. Spelman (1987). Problem-Solving: Problem-Oriented Policing in Newport News. Washington, D.C.: Police Executive Research Forum. Eck, J., and D. Weisburd (eds.) (1995). Crime and Place. Crime Prevention Studies, Vol. 4. Monsey, N.Y.: Criminal Justice Press, and Washington, D.C.: Police Executive Research Forum. Farrell, G. and K. Pease (eds.) (2001). Repeat Victimization. Crime Prevention Studies, Vol. 12. Monsey, N.Y.: Criminal Justice Press. Felson, M. (1994). Crime and Everyday Life. Thousand Oaks, Calif.: Pine Forge Press. Felson, M., and R. Clarke (1998). Opportunity Makes the Thief: Practical Theory for Crime Prevention. Police Research Series, Paper No. 98. London: Home Office Policing and Reducing Crime Unit. Fontana Police Department (1998). “Transient Enrichment Network.” Submission for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. Fraser, C., M. Scott, J. Heisey, and R. Wasserman (1998). Challenge To Change: The 21st Century Policing Project. Washington, D.C.: Police Executive Research Forum. Fremont, Calif., Police Department (1997). “A Project Analysis: Domestic Violence Revictimization Prevention.” Submission for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. Fyfe, J., J. Greene, W. Walsh, O.W. Wilson, and R.C. McLaren (1997). Police Administration (5th ed.). New York: McGraw-Hill. Geller, W., and G. Swanger (1995). Managing Innovations in Policing: The Untapped Potential of the Middle Manager. Washington, D.C.: Police Executive Research Forum. Gilling, D. (1996). “Problems With the Problem-Oriented Approach.” In The Politics and Practice of Situational Crime Prevention. Crime Prevention Studies, Vol. 5. Monsey, N.Y.: Criminal Justice Press. Giuliani, R., and H. Safir (n.d.). Compstat: Leadership in Action. New York: City of New York. Glazer, E. (1999). “How Federal Prosecutors Can Reduce Crime.” The Public Interest 136 (Summer). Glendale Police Department (1997). “Day Laborer Project.” Submission for the Herman Goldstein Award for Excellence in Problem-Oriented Policing. Goldstein, H. (1963). “Police Discretion: The Ideal vs. the Real.” Public Administration Review 23. — (1977). Policing a Free Society. Cambridge, Mass.: Ballinger Publishing Co. — (1979). “Improving Policing: A Problem-Oriented Approach.” Crime and Delinquency (April). — (1980). “Proposal for Further Development of a Problem- Oriented Approach to Improving Police Service,” May 12. — (1985a). Early Impressions of the Impact of Increased Sanctions on the Arrest, Prosecution, Adjudication, and Sentencing of Drinking Drivers in Madison, Wisconsin. Madison: University of Wisconsin Law School. — (1985b). “Community Policing: Some Hard Questions.” Speech at New York University Law School, Nov. 18. — (1987a). “Toward Community-Oriented Policing: Potential, Basic Requirements and Threshold Questions.” Crime and Delinquency 33(1) (January). —(1987b). “Comments of Herman Goldstein on Problem-Oriented Policing.” Paper presented at the Conference on Policing: State of the Art III, Phoenix, June 12. — (1990a). Problem-Oriented Policing. Philadelphia: Temple University Press. Also published in paperback by McGraw-Hill. — (1990b). “Does Community Policing Work?” The New York Times, Dec. 30. — (1991). “The Sine Qua Non of Policing.” Problem Solving Quarterly 5(1) (Fall/Winter). Reprint of speech by Herman Goldstein to Second Annual Problem-Oriented Policing Conference. — (1992). Notes for speech presented at the Third Annual Problem- Oriented Policing Conference, Oct. 10. — (1993a). “A Concept Paper on a Program To Support the Changing Response of Municipal Police to Crime, Disorder and Associated Fear.” Prepared for the U.S. Department of Housing and Urban Development, March 9. — (1993b). “Confronting the Complexity of the Policing Function.” In L. Ohlin and F. Remington (eds.), Discretion in Criminal Justice: The Tension Between Individualization and Uniformity. Albany, N.Y.: State University of New York Press. — (1993c). “The New Policing: Confronting Complexity.” Research in Brief (December). Washington, D.C.: National Institute of Justice. — (1994a). “Discussion paper: Examining the current status of problem-oriented policing, and thinking through an agenda for research and technical support for further development of the concept.” Memorandum to Craig Uchida, acting director, National Institute of Justice, July 11. Copy shared with author by Herman Goldstein. — (1994b). “Suggestions for inclusion in the NIJ research agenda relating to the police.” Memorandum to Jeremy Travis, director, National Institute of Justice, Sept. 29. Copy shared with author by Herman Goldstein. — (1995a). “Problem-Oriented Policing: The Rationale, the Concept and Reflections on Its Implementation.” Paper prepared for the Conference on Crime Control in Theory and Practice, Robinson College, Cambridge University, England, July 10. — (1995b). Personal correspondence from Herman Goldstein to Professor Jean-Paul Brodeur. Copy shared with author by Herman Goldstein. — (1995c). “Resolving the Present Conflict Between Image and Reality: An Overview of Emerging Changes in the Structuring of Police Service.” Presentation handout, Oct. 18. — (1996a). “Problem-Oriented Policing: Restructuring the Delivery of Police Service.” In Problem Oriented Policing: Towards Best Practice. Conference Proceedings. Address by Herman Goldstein to the Fourth Annual Crime Prevention Conference, Brisbane, Australia, June 11-12. — (1996b). “An Address by Herman Goldstein to the Summer Conference of the Association of Chief Police Officers.” Manchester, England, July 3. — (1996c). “Establishing Ownership of Inquiries and Responses to Problems in the Context of Problem-Oriented Policing.” Unpublished draft. — (1999). “SUNY Albany Conference on Zero Tolerance Policing Initiatives - Legal and Policy Perspectives.” Criminal Law Bulletin 35(4):370-73. — (n.d.). “Summary of the Basic Elements Incorporated in the Problem-Oriented Approach to Improving Police Service.” Goldstein, H., and C. Susmilch (1981). The Problem-Oriented Approach to Improving Police Service: A Description of the Project and an Elaboration of the Concept. Photocopy of Vol. 1 of the Project on Development of a Problem-Oriented Approach to Improving Police Service. Madison, Wisc.: University of Wisconsin Law School. — (1982a). The Drinking Driver in Madison: A Study of the Problem and the Community's Response. Photocopy of Vol. 2 of the Project on Development of a Problem-Oriented Approach to Improving Police Service. Madison, Wisc.: University of Wisconsin Law School. — (1982b). The Repeat Sexual Offender in Madison: A Memorandum on the Problem and the Community's Response. Photocopy of Vol. 3 of the Project on Development of a Problem-Oriented Approach to Improving Police Service. Madison, Wisc.: University of Wisconsin Law School. — (1982c). Experimenting With the Problem-Oriented Approach to Improving Police Service: A Report and Some Reflections on Two Case Studies. Photocopy of Vol. 4 of the Project on Development of a Problem-Oriented Approach to Improving Police Service. Madison, Wisc.: University of Wisconsin Law School. Goldstock, R. (1991). The Prosecutor as Problem-Solver. New York: Center for Research in Crime and Justice at New York University School of Law. Green Mazerolle, L., and R. Haas (n.d.). “'The Problem-Solver': The Development of Information Technology To Support Problem- Oriented Policing.” Unpublished paper. Green Mazerolle, L., and J. Roehl (eds.) (1998). Civil Remedies and Crime Prevention. Crime Prevention Studies, Vol. 9. Monsey, N.Y.: Criminal Justice Press. Greene, J. (1998a). “The Road to Community Policing in Los Angeles: A Case Study.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. — (1998b). “The Importance of Place in the Control of Crime.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Greene, J., and S. Mastrofski (eds.) (1988). Community Policing: Rhetoric or Reality? New York: Praeger. Greene, J.A. (1999). “Zero Tolerance: A Case Study of Police Policies and Practices in New York City.” Crime and Delinquency 45(2):171-87. Gresty, B., et al. (1997). “Problem-Solving Policing: Report of Findings of Project Team.” Internal report. Liverpool, England: Merseyside Police. Gunn, E. (1998). Par Four. New York: Walker and Co. Harcourt, B. (1998) “Reflecting on the Subject: A Critique of the Social Influence Conception of Detterence, the Broken Windows Theory, and Order. Maintenance Policing New York Style.” 97 Mich. L. Rev. 291. Harries, K. (1990). Geographic Factors in Policing. Washington, D.C.: Police Executive Research Forum. Hartnett, S., and W. Skogan (1999). “Community Policing: Chicago's Experience.” National Institute of Justice Journal (April). Washington, D.C.: U.S. Department of Justice, National Institute of Justice. Hawkins, C., Jr. (1998). “Ready, Fire, Aim: A Look at Community Policing in Edmonton, Alberta, Canada.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Also published in Police Executive Research Forum (1996). Themes and Variations in Community Policing: Case Studies in Community Policing. Higdon, R., and P. Huber (1987). How To Fight Fear: The Citizen-Oriented Police Enforcement Program Package. Washington, D.C.: Police Executive Research Forum. Hoare, M.A., G. Stewart and C.M. Purcell (1984). The Problem-Oriented Approach: Four Pilot Studies. London: Metropolitan Police, Management Services Department. Home Office (1998). “£250 Million To Develop an Effective Crime Reduction Strategy of the Future.” News release, July 21. Homel, R. (ed.) (1996). The Politics and Practice of Situational Crime Prevention. Crime Prevention Studies, Vol. 5. Monsey, N.Y.: Criminal Justice Press. — (ed.) (1997). Policing for Prevention: Reducing Crime, Public Intoxication and Injury. Crime Prevention Studies, Vol. 7. Monsey, N.Y.: Criminal Justice Press. — (1998). “Policing Alcohol Problems.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Hope, T. (1994). “Problem-Oriented Policing and Drug-Market Locations: Three Case Studies.” In R.V. Clarke (ed.), Crime Prevention Studies, Vol. 2. Monsey, N.Y.: Criminal Justice Press. Hornick, J., B. Burrows, I. Tjosvold, and D. Phillips (1990). An Evaluation of the Neighborhood Foot Patrol Program of Edmonton Police Service. Report. Edmonton, Alberta: Canadian Research Institute for Law and the Family. Jacobs, J. (1961). The Death and Life of Great American Cities. New York: Vintage Books. Jeffery, C.R. (1971). Crime Prevention Through Environmental Design. Beverly Hills: Sage. Jesilow, P., J. Meyer, D. Parsons, and W. Tegeler (1998). “Evaluating Problem-Oriented Policing: A Quasiexperiment.” Policing: An International Journal of Police Strategies and Management 21(3). Kelling, G.L., T. Pate, D. Dieckman, and C.E. Brown (1974). The Kansas City Preventive Patrol Experiment: A Summary Report. Washington, D.C.: Police Foundation. Kennedy, D. (1999). “Research for Problem-Solving and the New Collaborations.” In Viewing Crime and Justice From a Collaborative Perspective: Plenary Papers of the 1998 Conference on Criminal Justice Research and Evaluation. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. Kenney, D.J., and S. Watson (1998). “Reducing Fear in the Schools: Managing Conflict With Student Problem-Solving.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Kirby, S. (1997). “The Feasibility of Implementing Problem-Oriented Policing Within the Lancashire Constabulary.” Internal report. Lancashire, England: Lancashire Constabulary. Klein, M. (1998). “The Problem of Street Gangs and Problem- Oriented Policing.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Koller, K. (1990). Working the Beat: The Edmonton Neighborhood Foot Patrol. Edmonton, Alberta: Edmonton Police Service. Kramer, L., and P. McElderry (n.d.). Total P.O.P.: How a Police Department Adopted Problem-Oriented Policing as Its Comprehensive Strategy. Colorado Springs, Colo.: Colorado Springs Police Department. La Vigne, N. (1999). “Computerized Mapping as a Tool for Problem- Oriented Policing.” Crime Mapping News 1(1) (Winter). Washington, D.C.: Police Foundation. La Vigne, N., and J. Wartell (eds.) (1998). Crime-Mapping Case Studies: Successes in the Field. Washington, D.C.: Police Executive Research Forum. — (1999). “Crime Mapping for Problem-Solving.” In C. Solé Brito and T. Allan (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work, Vol. 2. Washington, D.C.: Police Executive Research Forum. Law Enforcement News (1999). “Thinking Out Loud: How Far Has CJ Research Come in 10 Years?” Vol. 25, No. 513 (June 15). Leigh, A., T. Read and N. Tilley (1996). Problem-Oriented Policing: Brit POP. Crime Detection and Prevention Series, Paper No. 75. London: Home Office Police Research Group. — (1998). Brit POP II: Problem-Oriented Policing in Practice. Police Research Series, Paper No. 93. London: Home Office Policing and Reducing Crime Unit. Leighton, B. (1994). “Community Policing in Canada: An Overview of Experience and Evaluations.” In D. Rosenbaum (ed.), The Challenge of Community Policing: Testing the Promises. Thousand Oaks, Calif.: Sage Publications. Lewin, R. (1992). Complexity: Life at the Edge of Chaos. New York: Macmillan Publishing Co. Maple, J. with C. Mitchell. (1999). The Crime Fighter: Putting the Bad Guys Out of Business. New York: Doubleday. Massing, M. (1998). “The Blue Revolution.” The New York Review, Nov. 19. Mastrofski, S. (1999). Policing for People. Ideas in American Policing Series. Washington, D.C.: Police Foundation. Mastrofski, S., and C. Uchida (1993). “Transforming the Police.” Journal of Research in Crime and Delinquency 30(3). McElderry, P. (1999). “A Shift in Mindset.” Subject to Debate 13(3). Washington, D.C.: Police Executive Research Forum. McElroy, J., C. Cosgrove and S. Sadd (1993). Community Policing: The CPOP in New York. Newbury Park, Calif.: Sage Publications. McEwen, T. (1999). “NIJ's Locally Initiated Research Partnerships in Policing: Factors That Add Up to Success.” National Institute of Justice Journal (January). McEwen, T., and R. Pandey (1998). “National Institute of Justice and Community-Oriented Policing Services Locally Initiated Research Partnerships Program: Abstracts of Participating Projects.” Unpublished document sent to author, Feb. 2. Merseyside Police (n.d.). “Incident Management Units: A Corporate Policy.” Internal report. Liverpool, England: Merseyside Police. Moore, M. (1998). “Problem-Solving and Community Policing.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Natarajan, M. and M. Hough (2000). Illegal Drug Markets: From Research to Prevention Policy. Crime Prevention Studies, Vol. 11. Monsey, N.Y.: Criminal Justice Press. National Crime Prevention Council (1997). Designing Safer Communities: A Crime Prevention Through Environmental Design Handbook. Washington, D.C.: National Crime Prevention Council. National Institute of Justice (1999). 1998 Year-in-Review Research Report. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. — (2000). Excellence in Problem-Oriented Policing: The 1999 Herman Goldstein Award Winners. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. Newman, O. (1972). Defensible Space: Crime Prevention Through Urban Design. New York: MacMillan. Nicholl, C. (1999a). Toolbox for Implementing Restorative Justice and Advancing Community Policing. Washington, D.C.: U.S. Department of Justice, Office of Community-Oriented Policing Services. — (1999b). Community Policing, Community Justice and Restorative Justice. Washington, D.C.: U.S. Department of Justice, Office of Community-Oriented Policing Services. O'Connor Shelley, T., and A. Grant (eds.) (1998). Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Oettmeier, T., and M.A. Wycoff (1998). “Personnel Performance Evaluations in the Community Policing Context.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Office of Community Oriented Policing Services (1999). COPS: 100,000 Officers Funded. Washington, D.C.: U.S. Department of Justice, Office of Community Oriented Policing Services. Office of Justice Programs (1998). Office of Justice Programs Fiscal Year 1998 Program Plans. Washington, D.C.: U.S. Department of Justice, Office of Justice Programs. Ohlhausen Research Inc. (1999). “A Community Policing Approach to Domestic Violence Prevention and Control.” Report prepared for the Regional Community Policing Institute at Eastern Kentucky University. Painter, K. and N. Tilley (1999). Surveillance of Public Space: CCTV, Street Lighting and Crime Prevention. Crime Prevention Studies, Vol. 10. Monsey, N.Y.: Criminal Justice Press. Pawson, R., and N. Tilley (1994). “What Works in Evaluation Research?” British Journal of Criminology 34(3). Peak, K., and R. Glensor (1996). Community Policing and Problem-Solving: Strategies and Practices. Upper Saddle River, N.J.: Prentice Hall. — (1999). Community Policing and Problem-Solving: Strategies and Practices (2nd ed.). Upper Saddle River, N.J.: Prentice Hall. Police Executive Research Forum (1989). “Taking a Problem-Oriented Approach to Drug Enforcement.” Interim report to the Bureau of Justice Assistance. Washington, D.C.: Police Executive Research Forum. — (1996). Themes and Variations in Community Policing. Washington, D.C.: Police Executive Research Forum. — (2000). National Evaluation of the Problem-Solving Partnerships (PSP) Project for the Office of Community Oriented Policing Services (COPS), U.S. Department of Justice. Washington, D.C.: Police Executive Research Forum. Police Foundation (1981). The Newark Foot Patrol Experiment. Washington, D.C.: Police Foundation. Prenzler, T. (1997). “A Problem-Oriented Approach to Preventing Sex Discrimination in Police Recruitment.” In R. Homel (ed.), Policing for Prevention: Reducing Crime, Public Intoxication and Injury. Crime Prevention Studies, Vol. 7. Monsey, N.Y.: Criminal Justice Press. Read, T. and N. Tilley (2000). Not Rocket Science? Problem-solving and crime reduction. Crime Reduction Research Series, Paper No. 6. London: Home Office Policing and Reducing Crime Unit. Remington, F. (1952). “A Proposed Criminal Code for Wisconsin.” 20 U. Kan. City L. Rev. 221. — (1954). “Criminal Law Revision: Codification vs. Piecemeal Amendment.” 33 Neb. L. Rev. 396. Reuland, M.M. (ed.) (1997). Information Management and Crime Analysis: Practitioner's Recipes for Success. Washington, D.C.: Police Executive Research Forum. Roberts, D. (1999). “Foreword: Race, Vagueness and the Social Meaning of Order-Maintenance Policing.” Journal of Criminal Law and Criminology 89(3) (Spring):775-836. Rosen, M.S. (1997). “A LEN Interview With Professor Herman Goldstein, the 'Father' of Problem-Oriented Policing.” Law Enforcement News 23(461). — (1999). “A LEN Interview With Professor George Kelling, Coauthor of 'Fixing Broken Windows.'” Law Enforcement News 25(511, 512). Rosenbaum, D., R. Flewelling, S. Bailey, C. Ringwalt, and D. Wilkinson (1994). “Cops in the Classroom: A Longitudinal Evaluation of Drug Abuse Resistance Education (DARE).” Journal of Research in Crime and Delinquency 31:3-31. Rosenbaum, D., and A. Lurigio (1998). “An Inside Look at Community Policing Reform: Definitions, Organizational Changes and Evaluation Findings.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Rosenfeld, R., and S. Decker (1998). “Consent to Search and Seize: Evaluating an Innovative Youth Firearm Suppression Program.” In T. O'Connor Shelley and A. Grant (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work. Washington, D.C.: Police Executive Research Forum. Roth, J. and J. Ryan (2000). “The COPS Program After 4 Years–National Evaluation.” Research in Brief. Washington, D.C.: National Institute of Justice. Rottman, D., and P. Casey (1999). “Therapeutic Jurisprudence and the Emergence of Problem-Solving Courts.” National Institute of Justice Journal (July). Ryan, J., and A. Doyle (1986). Operation Outreach: A Study of Runaway Children in New York City. New York: New York City Police Department. Sampson, R. (1999). “Systematic Social Observation of Public Spaces: A New Look at Disorder in Urban Neighborhoods.” American Journal of Sociology 105(3) (November):603-51. Sampson, R., and M. Scott (2000). Tackling Crime and Other Public-Safety Problems: Case Studies in Problem-Solving. Washington, D.C.: U.S. Department of Justice, Office of Community-Oriented Policing Services. Savannah Morning News (1999). “Shooting Blanks,” June 15. Saville, G. (1999). “Using Urban Design To Help Eradicate Crime Places.” In C. Solé Brito and T. Allan (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work, Vol. 2. Washington, D.C.: Police Executive Research Forum. Saville, G., and D.K. Rossmo (1995). “'Striking a Balance': Lessons From Problem-Oriented Policing in British Columbia.” In K. Hazlehurst (ed.), Perceptions of Justice: Issues in Indigenous and Community Empowerment. Aldershot, England: Avebury. Scott, M.S. and R.V. Clarke (2000). “A Review of Submissions for the Herman Goldstein Award for Excellence in Problem-Oriented Policing.” In C. Solé Brito and E. Gratto (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work, Vol. 3. Washington, D.C.: Police Executive Research Forum. Sherman, J. (1990). “Bottles and glasses 'used in most attacks in cities.'” The London Times, April 16. Sherman, L. (1991). “Book Review: Problem-Oriented Policing.” Journal of Criminal Law and Criminology 82(3). — (1998). Evidence-Based Policing. Ideas in American Policing Series. Washington, D.C.: Police Foundation. Sherman, L., D. Gottfredson, D. MacKenzie, J. Eck, P. Reuter, and S. Bushway (1997). Preventing Crime: What Works, What Doesn't, What's Promising. Washington, D.C.: U.S. Department of Justice, Office of Justice Programs. Sherman, L., C. Milton and T. Kelly (1973). Team Policing: Seven Case Studies. Washington, D.C.: Police Foundation. Sherwood, F. (1977). “It Happened in Santa Ana: A Study in Organization Change.” Los Angeles: The Police Executive Program (a cooperative venture of the Police Foundation, Law Enforcement Assistance Administration and University of Southern California). Silverman, E. (1999). NYPD Battles Crime: Innovative Strategies in Policing. Boston: Northeastern University Press. Skogan, W. (1998). “Community Policing in Chicago.” In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Skogan, W., S. Hartnett, J. Dubois, J. Comey, M. Kaiser and J. Lovig (1999). On the Beat: Police and Community Problem Solving. Boulder, Col.: Westview Press. Skolnick, J., and D. Bayley (1988). Community Policing: Issues and Practices Around the World. Washington, D.C.: National Institute of Justice. Solé Brito, C., and T. Allan (eds.) (1999). Problem-Oriented Policing: Crime- Specific Problems, Critical Issues and Making POP Work, Vol. 2. Washington, D.C.: Police Executive Research Forum. Sparrow, M. (1994). Imposing Duties: Government's Changing Approach to Compliance. Westport, Conn.: Praeger. — (forthcoming). The Regulatory Craft: Controlling Risks, Solving Problems and Managing Compliance. Washington, D.C.: Brookings Institution Press. Stockdale, J.E., C.M.E. Whitehead and P.J. Gresham (1999). Applying Economic Evaluation to Policing Activity. London: Home Office. Taft, P. (1986). Fighting Fear: The Baltimore County COPE Project. Washington, D.C.: Police Executive Research Forum. Taxman, F., and T. McEwen (1998). “Using Geographical Tools With Interagency Work Groups To Develop and Implement Crime Control Strategies.” In D. Weisburd and T. McEwen (eds.), Crime Mapping and Crime Prevention. Crime Prevention Studies, Vol. 8. Monsey, N.Y.: Criminal Justice Press. Taylor, R. (1999). “Crime, Grime, Fear, and Decline: A Longitudinal Look.” Research in Brief. Washington, D.C.: National Institute of Justice. Tilley, N. (1993). Understanding Car Parks, Crime and CCTV: Evaluation Lessons From Safer Cities. Police Research Group Crime Prevention Unit Series, Paper No. 42. London: Home Office. — (1999). “The Relationship Between Crime Prevention and Problem-Oriented Policing.” In C. Solé Brito and T. Allan (eds.), Problem-Oriented Policing: Crime-Specific Problems, Critical Issues and Making POP Work, Vol. 2. Washington, D.C.: Police Executive Research Forum. Toch, H., and J.D. Grant (1991). Police as Problem-Solvers. New York: Plenum Press. Travis, J. (1999). “Can we talk? Yes, let's!” Law Enforcement News 24(500). Travis, J., G. Lynch and E. Schall (1993). Rethinking School Safety: The Report of the Chancellor's Advisory Panel on School Safety. New York: Chancellor's Advisory Panel on School Safety. Vaughn, M.S. (1992). “Problem-Oriented Policing: A Philosophy of Policing for the 21st Century.” Review essay. Criminal Justice and Behavior 19(3) (September). Waldrop, M.M. (1992). Complexity: The Emerging Science at the Edge of Order and Chaos. New York: Simon & Schuster. Walker, S. (1993). “Does Anyone Remember Team Policing? Lessons of the Team Policing Experience for Community Policing.” American Journal of Police 12(1). Weick, K. (1984). “Small Wins: Redefining the Scale of Social Problems.” American Psychologist 39(1):40-49. Weisburd, D., and T. McEwen (eds.) (1998). Crime Mapping and Crime Prevention. Crime Prevention Studies, Vol. 8. Monsey, N.Y.: Criminal Justice Press. Weisel, D.L., and J. Eck (1994). “Toward a Practical Approach to Organizational Change: Community Policing Initiatives in Six Cities.” In D. Rosenbaum (ed.), The Challenge of Community Policing: Testing the Promises. Thousand Oaks, Calif.: Sage Publications. Weiss, A. (1997). “The Communication of Innovation in American Policing.” Policing: An International Journal of Police Strategies and Management 20(2). — (1998). “Informal Information Sharing Among Police Agencies.” NIJ Research Preview (December). West, M. (1995). “POP in Fresno: Effective Problem-Solving Techniques.” Police Chief 62(3):37-43. Willemse, H. (1994). “Developments in Dutch Crime Prevention.” In R.V. Clarke (ed.), Crime Prevention Studies, Vol. 2. Monsey, N.Y.: Criminal Justice Press. Williams, J. and R. Sloan (1990). Turning Concept Into Practice: The Aurora, Colorado Story. National Center for Community Policing monograph. East Lansing, Mich.: Michigan State University. Wilson, J.Q., and G.L. Kelling (1982). “Broken Windows: The Police and Neighborhood Safety.” The Atlantic Monthly (March):29-38. — (1989). “Making Neighborhoods Safe.” The Atlantic Monthly (February):46-52. Wycoff, M.A., and W. Skogan (1993). Community Policing in Madison: Quality From the Inside Out: An Evaluation of Implementation and Impact. Washington, D.C.: U.S. Department of Justice, National Institute of Justice. Young, J. (1998). Community Policing in Savannah, Georgia. In G. Alpert and A. Piquero (eds.), Community Policing: Contemporary Readings. Prospect Heights, Ill.: Waveland Press. Also published in Police Executive Research Forum (1996). Themes and Variations in Community Policing: Case Studies in Community Policing.