U.S. Department of Justice

Office of Justice Programs

National Institute of Justice

National Institute of JusticeR e s e a r c h R e p o r t

DEP

ARTMENT OF JUSTICE

OF

FIC

E

OF JUSTICE PRO

GR

AM

S

BJA

N

IJOJJ DP BJS

OV

C

“Broken Windows” andPolice Discretion

U.S. Department of JusticeOffice of Justice Programs810 Seventh Street N.W.Washington, DC 20531

Janet RenoAttorney General

Raymond C. FisherAssociate Attorney General

Laurie RobinsonAssistant Attorney General

Noël BrennanDeputy Assistant Attorney General

Jeremy TravisDirector, National Institute of Justice

Office of Justice Programs National Institute of JusticeWorld Wide Web Site World Wide Web Sitehttp://www.ojp.usdoj.gov http://www.ojp.usdoj.gov/nij

“Broken Windows” andPolice Discretion

George L. KellingProfessor, School of Criminal Justice, Rutgers University

Research Fellow, John F. Kennedy School of Government, Harvard UniversitySenior Fellow, Manhattan Institute

October 1999NCJ 178259

Jeremy TravisDirector

Steve EdwardsProgram Monitor

This program was supported under award number 95–IJ–CX–0013 to George L. Kelling by the NationalInstitute of Justice, Office of Justice Programs, U.S. Department of Justice. Findings and conclusions ofthe research reported here are those of the author and do not necessarily reflect the official position orpolicies of the U.S. Department of Justice.

The National Institute of Justice is a component of the Office of Justice Programs, which also includes theBureau of Justice Assistance, the Bureau of Justice Statistics, the Office of Juvenile Justice and DelinquencyPrevention, and the Office for Victims of Crime.

iii

Foreword

The past two decades have seen growing aware-ness of the complexity of police work, an ex-amination of the use of discretion in officers’daily policing activities, and a better under-standing of the critical role community leadersplay in the vitality of neighborhoods.

Noted criminologist George L. Kelling hasbeen involved in practical police work sincethe 1970s, working day-to-day with officers innumerous agencies in all parts of the countryand serving as an adviser to communities, largeand small, looking for better ways to integratepolice work into the lives of their citizens.

In the context of the “broken windows” meta-phor, proposed by James Q. Wilson and Dr.Kelling in 1982 in The Atlantic Monthly, thisResearch Report details how an officer’ssensitive role in order maintenance and crimeprevention extends far beyond just arrestinglawbreakers—how discretion exists at everylevel of the police organization. Historically,police have asserted authority in many ways,often having nothing to do with arrest. Dr.Kelling takes a special interest in the use ofdiscretion to exercise the core police authority,enforcement of the law.

He wants to understand better why officersmake arrests in some circumstances and notothers, especially when they are dealing withthe more mundane aspects of policing—such

as handling alcoholics and panhandlers andresolving disputes between neighbors. And henotes that police officers themselves are oftenunable to articulate the precise characteristicsof an event that led them to act as they did.Kelling maintains that officers must and shouldexercise discretion in such situations. But giv-ing police officers permission to use their pro-fessional judgment is not the same as endorsingrandom or arbitrary policing. In his view, polic-ing that reflects a neighborhood’s values andsense of justice and that understands residents’concerns is more likely to do justice than polic-ing that strictly follows a rule book.

Police work is in transition within commu-nities. The police are more frequently involvedin creating and nurturing partnerships withcommunity residents, businesses, faith-basedorganizations, schools, and neighborhood asso-ciations. Their role in the justice process re-quires even greater commitment to developingpolicy guidelines that set standards, shape theinevitable use of discretion, and support com-munity involvement. We hope this ResearchReport will help inform the continuing debateover the proper exercise of police discretion inthis new era of policing.

Jeremy TravisDirectorNational Institute of Justice

v

Preface

thinking. Police culture and the professionhave changed dramatically as a result.

Nevertheless, discussions about substantivepolice work continue to lag. We now under-stand that telling officers only what they can-not do, which is so typical of police manualsand rules and regulations, has not improvedthe quality of policing. We know as well thatthe work world of police is too complex to tellofficers exactly what they should do in everycircumstance. The only alternative left for themanagement of most police work is to teachofficers how to think about what they shoulddo, do it, and then talk about it, so that theyimprove their practice over time and sharetheir emerging values, knowledge, and skillswith their colleagues and the profession.This report proposes a model for helpingpolice officers think about their work whilepracticing it.

During the late 1960s and early 1970s, FrankRemington, Herman Goldstein, and others ad-vanced the notion that police departments arecomparable to administrative agencies whosecomplex work is characterized by considerableuse of discretion. Moreover, they advocatedthe development of guidelines to shape policeuse of discretion. Their thinking and workwere ahead of their time; the field of policingwas simply not ready to consider seriously theimplications of this view. Policing was stillmired in the simplistic and narrow view of lawenforcement agencies as concerned primarilywith felonies—the front end of a criminal pro-cessing system.

Today, the ideas regarding the complexity ofpolice work and the ubiquity of discretion thatare inherent in research conducted about po-lice functioning during the 1950s through the1970s have permeated police and academic

vii

Contents

Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

A Short History of Police Accountability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

The Urgency of Influencing Police Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9Preventive Interventionist Policing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9Reopened Issues for Policing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Police, Law, and the Constitution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13What Is the Future of Guidelines Development? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

The Discovery of Discretion and Its Meaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Survey of Criminal Justice Agencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Study Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

The Police Guidelines Movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Development of Police Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33Recognizing the Complexity of Police Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34Acknowledging Police Use of Discretion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37Recognizing and Confirming How Police Work Is Conducted . . . . . . . . . . . . . . . . . . . . . . . . 38Advancing Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Putting Police Knowledge Forward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41Undergoing Development by Practicing Police Officers and Citizens . . . . . . . . . . . . . . . . . . 41Undergoing Clear and Broad Public Promulgation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42Emphasizing Police Adherence to a Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43Establishing Accountability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44Receiving Recognition as an Ongoing Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47

Appendix A: Order Maintenance Training Bulletin 96–1 . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

Appendix B: San Diego Police Department Policy Statement . . . . . . . . . . . . . . . . . . . . . . . . 53

1

Introduction

Police departments throughout the United States,as well as in Canada, England, and elsewhere inEurope, have begun to reemphasize the mainte-nance of public order as an essential aspect ofpolice work. Police activities to restore publicorder in New York City and its subway system,for instance, have received extensive publicityand professional attention. So, too, has the“broken windows” metaphor as it has beenclosely linked to New York City.1

The New York story, as well as others, raisesserious questions. To what extent is order main-tenance linked to the current decline of crimein the United States? Regardless of its efficacy,how proper is assertive police order mainte-nance? To what extent can police brutality be ex-plained by “turning the police loose” with ordermaintenance tactics? Many civil libertarians andadvocates for the homeless, for example, opposeorder maintenance because they believe it in-fringes on the liberties of selected populations(the poor, minorities, the homeless, and youths)and opens the door to abusive police practices.2

The debates about these issues have been vigor-ous and often rancorous.3

It is not the intent of this report to debate theseissues or even dwell on them at any length. Ihave addressed them elsewhere and will do soagain in the future.4 It is important, however,to note up front that order maintenance has thepotential for abuse. Vagrancy and loitering laws,for example, have been used to deny minoritiestheir rights and to abuse citizens, especiallyAfrican-Americans. But my concern here is notlimited to order maintenance activities; the con-cern here is how to manage properly what EgonBittner calls literal police work.5

Past and current police administrators have at-tempted to shape police work through the devel-opment of command and control organizations,recruitment, training, supervision, rules andregulations, rewards and punishment, specializa-tion, and routinized tactics like preventive patroland rapid response to calls for service. Certainly,these efforts to control officers have powerfullyinfluenced how American society is policed.Rationalized police organizations deploy well-equipped officers throughout their jurisdictionsin an organized fashion that allows them to re-spond to calls for service quickly. The old imageof policing on which the “Keystone Cops” werebased belongs to another age. Nevertheless, put-ting officers on the streets in a timely and orga-nized fashion and getting them to particularlocations rapidly is quite different from shapingpolice behavior once officers are out in the com-munity dealing with citizens’ problems, needs,and conflicts.

Police administrations’ limited ability to shapepolice street practice persists despite manage-ment’s preoccupation with control—an orienta-tion that largely grew out of efforts to minimizethe kinds of corruption, especially political cor-ruption, that plagued late 19th and early 20thcentury American policing.6 Yet, as valid as itsorigins were, police administrators’ preoccupa-tion with control, and the methods they adoptedto maintain it, have had tragic, unanticipatedconsequences. Most obviously, this preoccupa-tion has fostered a bitterly antimanagement cul-ture in many police departments. In this cultureofficers are alienated from the citizens theyserve, support a “stay out of trouble” (by doingnothing) mentality, and, while disapprovingof abuse and corruption, nonetheless protectdeviant officers in the name of occupationalsolidarity.7

2

James Q. Wilson noted some of the conse-quences of rules and regulations on officers’beliefs in 1968:

But there are at least two limits to the value ofnegative policies [things officers may not do].First, they leave untouched a large area ofnecessary discretion and, second, they areperceived as irrelevant and unhelpful restric-tions—as rules that “tell us what we shouldn’tdo” and thus “give the brass plenty of ropewith which to hang us,” but that “don’t tell uswhat we should do.”8

These beliefs—that rules tell officers only whatnot to do, will be used to “hang” officers, andfail to tell police what they should do in apositive way—have arisen in part from policeadministrators’ inattention to the substantivecontent of day-to-day policing.

At least three explanations can be given for thisinattention:

● Oversimplified, but robust, views of policework have dominated police conventionalwisdom for most of this century. The idea ofpolice as “crime fighters,” or merely “lawenforcement officers,” was the cornerstone ofan ideological view of police that dismissed,ignored, or was oblivious to actual policefunctioning.

● Many police chiefs who have been forthrightabout police work find that legislators, may-ors, and other officials often do not want tohear that police officials, not to mention patrolofficers, are involved in policy decisions abouthow problems should be managed.9 This ispartially because the general public or publicofficials simply do not understand the com-plexity of many problems.

● Politicians and media representatives often areso caught up in the simple-minded slogans of“wars on crime,” the “thin blue line,” and soforth, that they are not prepared (whether in-tentionally or not) to hear about the real worldof policing.

Some chiefs have demanded or implied thatwhen confronted with difficult or ambiguousproblems, line police should “do what has tobe done and cover your ass.” While often notexplicit, this message is nonetheless sent toofficers because they are frequently put intocomplex and troublesome situations where citi-zen demands for action are high and departmen-tal thinking about such situations is nonexistent.Although this defensive mentality does not char-acterize all police leadership, over time it hascharacterized enough leaders in enough depart-ments so that line officers have come to supportor tolerate a bitter antimanagement culture.10

Pushing harder and more stridently with currentcontrol mechanisms that exert little real controlover substantive work will not lead the way outof this quandary. Such specious thinking hasbeen in place since the 1950s (e.g., just a littlemore inservice training, a slightly tighter span ofcontrol, a few more general orders or rules, moremilitant internal affairs units, improved rewardsand punishments, improved or more representa-tive recruitment, greater militarization of recruittraining). Instead, police officials need to focuson the substantive content of police work; findand delineate the means to conduct police workmorally, legally, skillfully, and effectively; thenstructure and administer departments on thebasis of this literal work and not a fictionalizedview of police work. In other words, a cleardefinition and description of quality policing isneeded around which appropriate organizationsand administrations can be developed.

Concentrating on the substance of police work isnothing new; Herman Goldstein, Egon Bittner, thelate Frank Remington, and others have advocatedsuch a focus for decades. Today, the issue hasacquired a new urgency.

Although this report raises the broader issues ofdiscretion in all police work, it focuses primarilyon the more mundane aspects of policing, such asresolving petty conflicts, assisting and protectingchildren, managing drunks and the emotionally

3

disturbed—those activities that fill patrol officers’occupational lives. In addition, 11 principles to beconsidered in developing and implementing policeguidelines are explored:

● Recognizing the complexity of police work.

● Acknowledging police use of discretion.

● Recognizing and confirming how police workis conducted.

● Advancing values.

● Putting police knowledge forward.

● Undergoing development by practicing policeofficers and citizens.

● Undergoing clear and broad publicpromulgation.

● Prescribing what officers may not do.

● Emphasizing police adherence to a process.

● Establishing accountability.

● Receiving recognition as an ongoing process.

Moreover, this report takes a special interest inselective enforcement: helping officers articulatein professionally sound terms why they properlymake arrests in some circumstances (in the case ofpublic urination, for example) but not in others.

Notes1. Wilson, James Q., and George L. Kelling,“Broken Windows: The Police and Neighbor-hood Safety,” The Atlantic Monthly (March1982): 29–38; also see Kelling, George L., andCatherine M. Coles, Fixing Broken Windows:Restoring Order and Reducing Crime in OurCommunities, New York: Free Press, 1996.

2. Generally, civil libertarian groups, such as theNew York City Civil Liberties Union, make botharguments. Former San Jose Police Chief JosephMcNamara has been the most outspoken advo-cate of the points of view claiming that assertivepolice order maintenance tactics open the doorto abusive police practices. McNamara arguesthat former New York City Police Commissioner

William Bratton has a “flippant” attitude towardabuses such as those perpetrated by policeagainst Abner Louima (a Haitian immigrant) inNew York City, and that James Q. Wilson and I“opened the door” to such abuses in the original“Broken Windows” article. (See McNamara,Joseph, “Brutality in the Name of Public Safety,”Los Angeles Times, August 24, 1997, A1.)

3. At another level, the equation made in somequarters between police order maintenance ac-tivities (“broken windows”) and “zero tolerance”for disorderly behavior raises issues that go be-yond semantics. Without further comment, it isan equation that I have never made, find worri-some, and have argued against, considering thephrase “zero tolerance” not credible and smack-ing of zealotry.

4. Kelling and Coles, Fixing Broken Windows,Chapters 2, 4, and 5.

5. Bittner, Egon, Aspects of Police Work, Boston:Northeastern University Press, 1990: 4.

6. This does not imply that there were not otherreasons for police managers’ concerns for con-trol. Military organizations predisposed policetoward command and control issues, and the in-herent decentralization that goes with patrollingrequires that considerable effort be put intoestablishing control mechanisms.

7. For an interesting discussion of these issues,especially as a result of the Knapp Commissionin New York City, see Anechiarico, Frank, andJames B. Jacobs, The Pursuit of Absolute Integ-rity: How Corruption Control Makes Govern-ment Ineffective, Chicago: University of ChicagoPress, 1996.

8. Wilson, James Q., Varieties of Police Behav-ior, Cambridge, MA: Harvard University Press,1968: 279.

9. Goldstein, Herman, “Categorizing and Struc-turing Discretion,” Policing a Free Society,Cambridge, MA: Ballinger Publishing Com-pany, 1977: 93–130.

4

10. It is not unusual for police executives toblame unions for this antimanagement culture,but this is self-serving and deceptive becausethis thinking pervades many nonunionizeddepartments and existed long before unionsentered the police scene during the 1960s. See,for example, Kelling, George L., and Robert

Kliesmet, “Police Unions, Police Culture, theFriday Crab Club, and Police Abuse of Force,”in And Justice for All: Understanding and Con-trolling Police Abuse of Force, ed. William A.Geller and Hans Toch, Washington, DC: PoliceExecutive Research Forum, 1995: 187–204.

5

A Short History of Police Accountability

Police accountability is an issue that has beenwith democracies since modern police werecreated. Citizens should be concerned aboutcontrolling police: they have immense authority;they are armed and authorized to use force;and unlike the military, they are not sequesteredon bases—they are spread throughout thecommunity.

Modern American policing had its origins inEngland, where, in 1829, the questions ofwhether to create a police force and, if created,how to keep this force accountable were debatedby political, social, and philosophical elites formore than a century. Everything about policingwas “on the table” in this debate—the relation-ship of police to political authority, activitiesthat would constitute the business of policing,organizational structure and administrative pro-cesses, and the means by which police wouldobtain their goals.

During the late 18th and early 19th centuries, theEnglish stared across the channel at the French,or continental, model of policing, which reliedupon secret police and paid informers. Likewise,the English already had developed a model ofpolicing Ireland and their other colonies witharmed mounted police operating in groupsbivouacked to avoid contact with the nativepopulation.1

The English decided neither of these models wasacceptable. Early English policing emphasizeduniformed, unarmed, highly recognizable offi-cers who were diffused geographically through-out London and patrolled on foot. These officersnot only were granted authority from the Crownbut also had to earn the citizens’ trust. Investiga-tions conducted by plainclothes police, at leastinitially, were rejected; victims pursued criminal

investigations through some form of stipendiarypolice. Unobtrusive policing by public investiga-tors risked police meddling in political matters,and militarized policing risked further alienationof the general population. The primary businessof police was to prevent crime through presence,persuasion, and reduced opportunities.

In the United States, cities adopted the Englishmodel of policing. Although not originally out-fitted in uniforms—this was deemed Europeanelitism—American police were similarly dif-fused throughout cities, like their English coun-terparts, to prevent crime. For American police,the issue of who would control the police—urban political bosses or descendants of theoriginal Dutch and English settlers—was thedominant public concern. This struggle to con-trol police forces was to shape American policein remarkable ways.

During the first stage of evolutionary develop-ment (roughly from the 1850s to the 1920s),police were largely under the control of urbanpolitical machines; during the second stage(roughly the 1930s to the 1970s), police, withthe support of the progressives, evolved intovirtually autonomous urban agencies. HermanGoldstein has described police as evolving intothe least accountable branch of urban govern-ment during this era.2

To achieve this shift—from politically domi-nated and controlled to virtually autonomousforces—practically every aspect of Americanpolicing was reformed during the early decadesof the 20th century. Leading police thinkers (likeO.W. Wilson, Leonard Fuld, and Bruce Smith)were overwhelmingly preoccupied with issuesof control—that is, both wresting control ofpolice at all levels from political influences and

6

ensuring that only “professional” police influ-ence police. Bruce Smith, for example, wrotein 1929: “Without exceptions, all proposals forimprovement of organization and control havenecessarily been aimed at the weakening or theelimination of political influences.”3

In the name of eliminating corrupt political in-fluences from policing, these men attempted tochange the nature of the business from crimeprevention to reactive law enforcement. Theyrestructured police organizations, revised admin-istrative processes, developed new tactics, andredefined the relationship between police andcitizens—each, more or less successfully—allwith an eye toward gaining administrative con-trol of police, whether field commanders, super-visors, or patrol officers.4

By the 1950s, theorists wrote about the conductof police work—services line officers performin the course of their daily work—as being semi-automatic (i.e., police responses to incidentscould be, or should be, so controlled as to beanalogous to typing, piano playing, or rote ad-herence to a script). O.W. Wilson wrote in 1956,for example:

Administration has been defined as the artof getting things done. Police objectives areachieved by policemen at the level of perfor-mance where the patrolman or detective dealsface-to-face with the public—the complain-ants, suspects, and offenders—and the successof the department is judged by the perfor-mance of these officers.

Decisions that are advantageous to the depart-ment are most likely to be made by policemenwho have been selected in a manner to assuresuperior ability, who understand the policeobjectives and are sympathetic to them, whoare loyal to their department and capable ofoperating effectively, efficiently, and semi-automatically (i.e., with a minimum of con-scious self-direction, as in performance by askilled typist or pianist), and who have highmorale (i.e., the condition described by

military leaders as the right heart). (Emphasisadded.)5

Police work, in this view, as Egon Bittner oncesaid, could be conducted by persons who have“the ‘manly virtues’ of honesty, loyalty, aggres-siveness, and visceral courage.”6 The idea ofofficers thinking before they acted was to bediscouraged. They were to follow rules loyally.

The perception of police work as simple andunder administrative control was shattered, ofcourse, by research conducted in the 1950s bythe American Bar Foundation, which showedthat police work is complex, that police useenormous discretion, that discretion is at thecore of police functioning, and that police usecriminal law to sort out myriad problems. Theresearch suggested that the control mechanismsthat pervaded police organizations—especiallyrules and regulations, oversight, and militaristicstructure and training—were incompatible withthe problems that confront police officers dailyand the realities of how police services are deliv-ered. Aside from several who were scholars,few police administrators realized how “out oftouch” existing practices were with day-to-daypolice realities.

Paradoxically, a policing strategy that was over-whelmingly preoccupied with control, in thefinal analysis, failed to meet its most essentialcriteria. True, the strategy largely eliminated cor-rupt political influences in police departments.However, it left officers mostly to their owndevices in conducting the bulk of their work.

This state of affairs has not gone unnoticed ineither the legal or research community. The U.S.Supreme Court, for example, grew impatientwith the unwillingness or inability of policeexecutives to control criminal investigations—it was widely acknowledged since the 1930sWickersham Commission that police procedureembraced the practice of torture—and through aseries of decisions during the 1960s (the exclu-sionary rule, the requirement that offenders un-derstand their right to an attorney, and so forth),

7

established guidelines that shaped the futureconduct of criminal investigations.7 EgonBittner, based on his own research of policehandling of drunkenness and mental illness, haswritten both eloquently and indignantly aboutthe mismatch between “official” and real policework:

The official definition of the police mandateis that of a law enforcement agency. . . . Theinternal organization and division of laborwithin departments reflect categories of crimecontrol. Recognition for meritorious perfor-mance is given for feats of valor and ingenuityin crime fighting. But the day-to-day work ofmost officers has very little to do with all ofthis. These officers are engaged in what isnow commonly called peacekeeping and ordermaintenance, activities in which arrests areextremely rare. Those arrests that do occurare for the most part peacekeeping expedientsrather than measures of law enforcement ofthe sort employed against thieves, rapists, orperpetrators of other major crimes.

For the rich variety of services of every kind,involving all sorts of emergencies, abatementsof nuisances, dispute settlements, and an al-most infinite range of repairs on the flow oflife in modern society, the police neither re-ceive nor claim credit. Nor is there any recog-nition of the fact that many of these humanand social problems are quite complex, seri-ous, and important, and that dealing withthem requires skill, prudence, judgment, andknowledge.8

Notes1. Tobias, John J., “The British Colonial Police:An Alternative Police Style,” in Pioneers inPolicing, ed. Philip John Stead, Montclair, NJ:Patterson Smith, 1977: 241–261.

2. Goldstein, Herman, “Categorizing and Struc-turing Discretion,” Policing a Free Society,Cambridge, MA: Ballinger Publishing Com-pany, 1977: 93–130.

3. Smith, Bruce, “Municipal Police Administra-tion,” The Annals 146 (November 1929): 27.

4. Kelling, George L., and Mark H. Moore, “TheEvolving Strategy of Policing,” in Perspectiveson Policing, No. 4, Cambridge, MA: U.S. De-partment of Justice and Program in CriminalJustice Policy and Management, John F.Kennedy School of Government, HarvardUniversity, November 1988.

5. Wilson, O.W., “Basic Police Policies,” ThePolice Chief (November 1956): 28–29.

6. Bittner, Egon, Aspects of Police Work, Boston:Northeastern University Press, 1990: 6–7.

7. The Wickersham Commission was the firstnational survey of criminal justice and policepractices.

8. Bittner, Aspects of Police Work.

9

The Urgency of Influencing Police Work

Increasingly, police are under renewed and con-stant pressure from neighborhood groups andcity halls across the country—not to mentionState legislatures and the U.S. Congress—to “dosomething now” about eliminating the excessesof the drug market, getting guns off the street,and regaining control over public places.1 At-tempting to meet such demands has inevitablestrategic consequences for police departmentsand local government. The main consequence isthat police strategy shifts from a reactive and in-herently passive model to a preventive interven-tionist model that reopens policy issues aboutpolice handling of the homeless, drunks, drugdealers and users, the emotionally disturbed, andminor offenders that many believed had beenaddressed once and for all during the period fol-lowing the 1960s. This strategic change takespolice to the edge, or even over the edge, of

How New Haven Developed GuidelinesNew Haven’s policy guidelines embody important principles described in this report. Theguidelines focused on problems of disorder and were developed with officers and sergeants atthe request of then Chief Nick Pastore.

The guidelines have been distributed as a training bulletin in the New Haven Department ofPolice Service. (See Appendix A, “Order Maintenance Training Bulletin 96–1.”) The bulletinwas prepared as a city- and departmentwide document, rather than as a location-specific docu-ment. Consequently, it invites and outlines a problem-solving method to deal with location- orother problem-specific issues as they may occur in neighborhoods.

Of course, the value of guidelines depends on the skill of management in linking them to on-going police practice and administrative and supervisory procedures. Moreover, the examplesof guidelines included in this report are clearly not as good as they could be—none weredeveloped as part of a special “project” to develop guidelines. Likewise, the 11 points articu-lated elsewhere in this report are no doubt preliminary and inadequate. The art of developingguidelines has yet to be defined and will emerge only over time. Yet, guidelines are essentialto the development of police accountability and professionalism.

constitutional law—at least as it has been inter-preted for the past 30 years.

Preventive Interventionist PolicingWhen a shooting occurs in New Haven, Con-necticut, the Department of Police Servicesimmediately sends a team of Yale Child StudyCenter clinicians and police officers to helpchildren and families cope with the social andpsychological consequences of violence.2 (See“How New Haven Developed Guidelines.”)Likewise, when Community Patrol Officer JackFee, who is assigned to the Academy Homesin Boston, Massachusetts, discovers a hithertounenforced housing law that allows the courtto sentence a gang member with a history ofviolence, but no convictions, to a 1-year jailsentence for trespassing, this is preventive inter-ventionist policing.

10

Although it might not appear so, preventive po-licing that interrupts cycles of violence throughproblem solving is much more invasive of com-munity life than the reform—or 911—model thathas dominated policing in the United States sincethe 1950s. This inherent assertiveness can lead toserious misunderstandings of what constitutescommunity or problem-oriented policing.

For many people, thanks in part to how it hasbeen presented by many police leaders, commu-nity policing is viewed as “soft” policing compa-rable to community relations or, worse yet, socialwork. When asked to describe his efforts inAcademy Homes, Officer Jack Fee—knownthroughout the Boston Police Department as a“tough” cop—shrugged and said, half apologeti-cally, “I suppose I’m doing social work.” Thisperspective, that preventive policing is soft polic-ing, has created numerous obstacles for depart-ments attempting to get police rank and file toembrace community policing.

“Crime fighting” is inherently passive and reac-tive policing: patrol in a powerful car, receive a911 call, turn on flashing lights and siren, speedto the incident, tend to it as quickly as possible,and return to service. For the most part, thismethod of policing is in response to citizen-initiated calls to deal with incidents.3

Community policing is inherently proactive: scanfor problems; diagnose them; try to prevent themfrom occurring again; if they recur, try to limit thedamage and restore the victim/family/community’sfunctioning.4 This is implicit in New Haven’s ap-proach—tend to the incident, of course, but followup with action that limits the damage and helpschildren and families restore themselves to ad-equate functioning. Even police handling of murdershould go beyond criminal investigation.

This failure to understand the inherent asser-tiveness of community or problem-orientedpolicing, and equating it with soft policing, hasgreater consequences than just alienating linepolice officers. The equation of “community”and “soft” leads to public confusion when tough

police action is required to deal with severecrime or disorder problems. New York Cityprovides the best current example.

Critics have attacked New York Police Depart-ment (NYPD) community policing efforts sinceformer Commissioner William Bratton imple-mented assertive and enterprising (do not readcombative here) policing to deal with disorderand crime.5 Indeed, a major Los Angeles Timesarticle in 1995 argued that the NYPD wasremilitarizing and transforming itself into whatthe Los Angeles Police Department (LAPD)was, while the LAPD was attempting to demili-tarize itself and become the “officer-friendliesthat the NYPD once symbolized.”6 The easewith which the article made this argument pointsto the danger of using military metaphors, aswas done by NYPD officials, to describepolicing.

The Los Angeles Times reporter wrote: “In LosAngeles, officers are trying to shed the veryimage that the NYPD covets [military]. Copsare getting out of cars, walking beats, putting onshorts and riding bicycles, all to be closer to thepublic.”7 The reporter, who chose to focus on themilitary metaphors used to describe the NYPDanticrime problem-solving methods, ignored thefact that the NYPD cops were in reality doingall of these things—probably at a rate faster thanany other police department in the country.

The reporter’s view of community policing assoft is evident throughout the story. For ex-ample: “Los Angeles police are being encour-aged to spend more time talking to residents andmerchants about what troubles their lives, to em-phasize community contacts over arrests, to con-sider the underlying community pressures thatgive rise to crime rather than focusing exclu-sively on criminal behavior itself.”8 Most advo-cates of community policing could say, or havesaid, similar words. It does not follow, however,that if police officers talk to citizens and come tounderstand “what troubles their lives” this is theend of their responsibility; they are still police.

11

This debate about New York City, if anything, hasbecome even more heated in recent years. MayorRudolph Giuliani and current Police Commis-sioner Howard Safir are favorite targets of thosewho view assertive policing as “turning the policeloose.” To be sure, incidents like Louima andDiallo warrant public concern; however, the ideathat such events, as tragic as they were, are inevi-table consequences of restoring order ignoresboth the benefits of the current police strategy andthe reality that such incidents are not limited torecent history.9 Outrageous brutality and tragic ac-cidents occurred during the post-Knapp era, when“staying out of trouble” became the NYPD’sdominant ethos as well.10

The fact that police add options to their reper-toire of methods, try to limit damage, and restorefunctioning does not mean that conventional as-sertive law enforcement is disallowed as a legiti-mate police tactic. For example, understandingthe dynamics of New York City’s “squeegeemen”—unwanted car window washers who in-timidate drivers into giving them money—andtalking to them did nothing to deter their behav-ior. Careful and judicious use of law enforce-ment did: policing that included citations,warrant service, arrest, and jailing ultimatelywere required to solve the problem.11

The growing practice of police using civil law toachieve criminal law objectives shows just howenterprising community and problem-solvingpolicing can be. These tactical innovations, civilremedies such as Boston Police DepartmentOfficer Jack Fee applied, include efforts to dealwith domestic violence, drug trafficking, racialharassment, disorderly behavior, and weaponspossession. Civil remedies include injunctions,restitution, forfeiture, and civil fines that can beused either alone or in tandem with criminal law,as in the case of use of property forfeitures todeal with drug dealing. Civil initiatives not onlyadd to the police inventory of authority to dealwith problems but also, as Mary M. Cheh hasnoted: “[C]ivil remedies offer speedy solutionsthat are unencumbered by the rigorous constitu-

tional protections associated with criminal trials,such as proof beyond a reasonable doubt, trialby jury, and appointment of counsel.”12 This ishardly soft policing. In fact, civil initiatives raiseimportant strategic, legal, and constitutionalissues about use of police authority.13

The misrepresentation of community policingas soft plays into the hands of those who are op-posed to change in policing methods. Those whowant to get back to “tough” law enforcement—police riding around in cars, responding to 911calls, and occasionally conducting order mainte-nance “sweeps” (another terrible metaphor todescribe what police do)—trivialize what hashappened in New York City. Moreover, whensuccessful assertive policing is misconstrued ascombative or militarized policing, it becomesfodder for those who want to “turn the policeloose.”

Citizens, politicians, and vested interest groupshave always pressured police to cut legal andconstitutional corners—that is, “to do what hasto be done.” Whether the corner cutting is illegalsearches, manhandling suspects, or “testilying,”the consequence is always the same: a policeculture that is alienated from democratic valuesand the public and that tolerates police abuse ofcitizens and police corruption.14

Reopened Issues for PolicingDuring the 1960s, policing in the United Stateswas profoundly affected by converging social,political, and professional and occupationalforces. These forces included the triumph of pro-gressivism as a model for local government andthe installation of preventive patrol and rapid re-sponse as conceived by O.W. Wilson and otherpolice reformers as the model of effective, effi-cient, and honest policing. At the same time, therights revolution had a major impact on society’sresponse to mental illness, drunkenness, va-grancy, and disorderly behavior. These socialforces also created the “criminal justice system”with the police as its front end and a criminal

12

justice educational establishment that promul-gated the ideas inherent in the “system.”15

The 1960s shaped a surprisingly congruent po-lice strategy that, while not adhered to univer-sally, nonetheless constituted the official strategyof progressive midcentury American police.Policing became synonymous with law enforce-ment in which police should be out—or shouldget out—of the business of dealing with emo-tionally disturbed individuals, youths, and theproblems of drunkenness, disorderly behavior,and minor offenses.

In this respect, an unwitting alliance was struckbetween police and civil libertarians. For manycivil libertarians, the police always implicitlythreaten the use of authoritative power. In thislibertarian view, juvenile crime, mental illness,drunkenness, and disorder should be the soleprovince of social workers and advocates. Ifanybody should intervene, it certainly shouldnot be the police. For police, relief from theseresponsibilities seemingly bailed them out ofinherently ambiguous and sticky areas that hadgotten them into trouble in the past. Police rec-ognized the social control opportunities that po-licing disorder and drunkenness provided, yetthey were inherently “messy” businesses bothliterally and metaphorically: literally becausedrunks are often unkempt, dirty, and unpleasantto manage; metaphorically because considerablesocietal ambiguity exists about how to deal withsuch problems.

Narrowing the focus to serious crime simplifiedlife considerably for police. “Crime fighting”was easy to portray to the public and providedan occupational vision around which officerscould rally. The demand for order that grew incities during the 1980s and early 1990s, how-ever, thrust police back into the middle of the so-cial, legal, moral, and constitutional issues theyhad tried to avoid. The publication of “BrokenWindows” first popularized the link betweendisorder and fear and hypothesized the linksamong disorder, fear, crime, and urban decay

that gave voice and legitimacy to citizen de-mands that order be restored.16 Later work byWesley Skogan confirmed these linkages andprovided even greater momentum to the move torestore order.17

By the early and mid-1990s, highly publicizedprograms to restore order by New York City’sTransit Police Department in that city’s subwaysystem and the NYPD in public spaces gener-ally, in San Francisco’s “Operation Matrix,” andin other cities once again put the police in themiddle of the growing public debate about howto handle disorder.18 Policy and strategy issuesonce thought resolved again became issues forpolice. What was the proper balance betweenindividual rights and community interests?

Questions arose both within and outside policeranks about the proper police role in handlingjuveniles, drunks, the homeless, and disorderlycitizens:

● Is “zero tolerance” an apt phrase to describepolice order maintenance? What does itmean?

● Do police have a role to play in crime preven-tion by working with youths as other than lawenforcement officers?

● Given the lack of resources, what is theproper police role in dealing with drunks andthe emotionally disturbed?

● What are the bases of authority for police tohandle disorderly persons?

● Are new sources of authority for dealing withproblems available, and how are they bestdeveloped and deployed?

● What new forms of training are required toprepare police for working with troubledyouths and adults?

● What do community collaboration, coopera-tion, and accountability really mean in anoperational sense?

13

● How are police kept from protecting narrow,parochial interests—such as keeping “strang-ers” out of neighborhoods and communities?

● How do police refrain from doing what citi-zens should do for themselves or from usurp-ing the roles of private or other governmentalagencies?

These issues, many of which were once regard-ed as resolved in many police circles, demandcareful and thoughtful consideration. The rein-vention of policing profoundly challengescontemporary police leaders, policymakers,scholars, and academics to consider new meth-ods of active political and community controlof police discretion and the development ofnew police guidelines.19

Police, Law, and the ConstitutionA lieutenant in a moderately large police de-partment recently described the activities of aspecial antiviolence unit he headed that concen-trated on confiscating guns from youths. Thelieutenant was bright, capable, and intimatelyfamiliar with Lawrence Sherman’s research ongun confiscation in Kansas City and the NewYork City efforts to reduce gun carrying.20 In themiddle of the conversation about gun confisca-tion, the lieutenant acknowledged somewhatruefully, “We’re really pushing the FourthAmendment here.” I asked him, “What specialpolicy guidelines did the department provide toofficers to ensure that they acted appropriately,at least by departmental standards?” My ques-tion was prompted by my work in New YorkCity with the Transit (subway), Metro North(Grand Central Terminal), and Long Island Rail-road (Penn Station) police departments. I hadhelped the departments develop policies andtactics to deal with disorderly behavior that hadtaken them to the edge of the law as well. In-deed, Robert Kiley, chairman of the Metropoli-tan Transportation Authority, and I understoodvery well that we would be going to court assoon as those policies and tactics were imple-mented.21 From the beginning of our activities,

we prepared for the legal, constitutional, andmoral battles that would be forthcoming whenwe challenged existing practices and traditions.

The lieutenant’s response, however, confirmedthat although the department’s gun strategydeliberately brought it to the verge of illegality,neither the department nor the unit had devel-oped, or had considered developing, an explicitpolicy statement about its antigun tactics thatwould communicate its policies to the publicand guide and control either the unit’s—or itsofficers’—field practices. In all likelihood, thedepartment’s rules and regulations specified thatofficers would always operate legally and con-tained the routine homage to the Fourth Amend-ment and the law regarding search and seizure.Moreover, it is impossible to write a policy state-ment that covers every issue that officers wouldconfront, and it would be unwise to try. Theformer concern, the impossibility of writingan effective policy statement, reflects a widelyshared belief in policing—and certainly not acapricious belief—that practical circumstancesare so complicated that every exigency cannot becovered in a policy statement. Regarding the lat-ter concern—that it would be unwise to try—thisreflects what is often viewed as true in policing:some things the police do that take them to theedge of the law have the potential to be so con-troversial that they are best left unmentioned andup to the discretion of officers.

My concern about the lieutenant’s statement wasalso driven by my observations of patrol practicesduring the early 1980s in four police departmentsthat were known to confiscate large numbers ofguns. The great majority of guns were confiscatedthrough “traffic enforcement” at all four sites. Inthe large midwestern city where I spent the mosttime observing, police (routinely in two-personpatrol cars) would stop cars of “interest,” almostinvariably legitimately, on the basis of some traf-fic violation or faulty equipment. A car of interestalways contained one or more male youths. All ofthe youths were African-American, but nearlyeveryone in this area of the city was African-American except for some police.

14

In conversations with officers, I learned that theydid not stop cars carrying women because awoman could not be searched unless a femalepolice officer was present. More often than not,no female officer was on site, meaning one hadto be called in from elsewhere in the city—aprocess that consumed too much time. Likewise,the officers did not stop cars driven by malesover the age of 30 because “even if they werecarrying, they probably wouldn’t hurt anyoneexcept in self-defense,” according to one officer.

Once a car was stopped, typically, one officerapproached the vehicle and asked the driver tostep out. The other officer stood either in backof the car or on the passenger side. As the driverstepped out of the car, the officer asked for adriver’s license and for the man to step to therear of the car to be sure that they were out oftraffic. The officer then indicated the basis forthe stop—generally something like “you wereover the center line” or “your left tail light isout.” After some exchange, the officer wouldask: “You wouldn’t mind if my partner checkedyour trunk, would you?” Again, almost invari-ably, the driver acquiesced. And so it pro-gressed—the next request was to check insidethe car. (If passengers were in the car, they wereasked to get out and join the officer and driver atthe back of the car.) Depending on the circum-stances, the driver and passengers would be“patted down.”

What I observed, of course, was a street game,the rules of which almost everyone understood.The traffic offense was legitimate and worthy ofa ticket. Each party was negotiating: the citizento get out of a ticket and the officer to searchthe car. The request to search the trunk was theopening gambit in the negotiation, and if thedriver refused, the officer, depending upon his orher assessment of suspicious behavior, personalsafety, or some other justification, could have in-sisted on a search. A refusal, however, probablyclinched the ticket. It cannot be overemphasized,however, how politely and respectfully officersin this city managed these interactions. It was

clear from their behavior, as well as from theircommunications with me, that the officerswanted to avoid conflict.

In observing such transactions, one could nothelp but be impressed by the skill, courtesy,and reasonableness of most of the officers.They were dealing with a serious problem, guncarrying, and behaved in an exemplary fashion.Nevertheless, it is unclear just how voluntarymost of the searches I had observed really were.These were really pretext stops. In essence,the police established a quid pro quo: you letme search your car, and I let you out of a ticket.Moreover, I observed officers who interpreted arefusal to allow their search as a threat to theirsafety that justified a search. While such transac-tions or subsequent interpretations may not haveoccurred in Indianapolis or Kansas City duringthe research there, I am reasonably certain thatthe lieutenant mentioned above was referring topractices not dissimilar to those I had observedin the early 1980s when he discussed “pushingthe Fourth Amendment.”22

The point here is not to judge whether the offi-cers I observed were justified in what they weredoing—clearly, their performance was in linewith common practice in the department. Myconcerns, unrelated as they are to squeamishnessabout pushing the law, have to do with the fail-ure of police departments to develop clear policystatements regarding some of the serious andcomplex issues they encounter. (See “Democ-racy, Policing, and Discretion.”)

Defining quality police work, recognizing it,and managing it are internal police matters andshould be pursued to improve policing.23 Thatgood policymaking reduces litigation is an addedbenefit of competence.

The current demands placed on police and thestrategic consequences of responding to them innew and innovative ways underscore the inher-ent complexity of good police work and the per-vasive role discretion plays in policing. The is-sues described above—the inherent assertiveness

15

of the new models of policing, the individualrights versus community interests issue, and thelegal and constitutional issues that are raised bynew police tactics—provide further evidence ofjust how complex policing is.

While the complexity of police work and thediscretion required in policing may be widelyacknowledged, the impact of this awareness onpolice practice and administration is not clearlydistinguishable. Arguably, attempts to controland shape police behavior are still largely mired

Democracy, Policing, and DiscretionAlthough police discretion cannot be structured by simply proscribing certain actions or issu-ing policy statements, departments that fail to develop clear policy guidelines about complexissues face several serious concerns:

● Cutting moral, legal, and constitutional corners, regardless of the high-minded purposes fordoing so, creates and perpetuates the morally ambiguous nature of police work in its literalsense—that which line police officers do.a One of the primary consequences of such moralambiguity is an isolated police culture and its “blue curtain.”

● Regardless of the skill and grace of officers in “doing what has to be done,” cutting cornerssends improper messages to citizens about how problems should be solved. Both those in-volved with officers in the quid pro quo and those who demand that police “do what has tobe done”—whether citizens or politicians—are given a terrible and dangerous message aboutpolicing a democratic society.

● When failing to wrestle with the complex moral and legal issues of social policies, depart-ments risk litigation, the outcome of which can seriously jeopardize current and future de-partmental efforts to deal with serious problems. The U.S. Supreme Court’s inquiry intocriminal investigation was the result of poor police practices and, more important in this con-text, management’s failure to rein in and control detectives. In fact, police departments andcities throughout the United States are currently under serious legal assault from organiza-tions such as the New York and American Civil Liberties Unions and other libertarian andadvocacy groups.b

● Most problems can be solved, or at least managed, once they are properly understood and arange of alternative solutions has been explored. Perhaps the best example is graffiti in NewYork City’s subway system. Once thought insoluble, the problem yielded to creative andthorough work. This does not mean it was completely eradicated—New York’s graffiti artistsare a dedicated lot—but it is a problem that, with vigilant maintenance of subway cars, canbe managed and kept to a minimum.

Notes

a. Bittner, Egon, Aspects of Police Work, Boston: Northeastern University Press, 1990: 4. Alsosee Kelling, George L., “How to Run a Police Department,” City Journal 5 (4) (Autumn1995): 34–45.

b. Kelling, George L., and Catherine M. Coles, Fixing Broken Windows: Restoring Order andReducing Crime in Our Communities, New York: Free Press, 1996: 38–69.

16

in the organizational and control apparatus thatcharacterized the far simpler, mid-20th centuryview of policing: command and control, quasi-military training, factorylike models of supervi-sion, and extensive rules and regulations. Thesecontrol mechanisms, which are irrelevant tostreet policing, are based on the Taylor modelthat has dominated organizational thinking formost of the 20th century.24

During my work with the New York City TransitPolice Department, in subsequent work withother police departments, and in reviews of de-partmental rules and regulations, general orders,and other policy materials, I was struck by thedepartments’ virtual silence about actual policework. A review of numerous departmental writ-ten materials showed that the majority dealt withinternal administrative issues. A significantlysmaller number addressed “hot” issues: use offorce, hot pursuit, lineup, and arrest procedures.25

Frank Vandall, Professor of Law at EmoryUniversity, made a similar observation abouttraining materials in 1976: “An examination ofcontemporary police training materials revealsthat they fail to deal with the concept of discre-tion in law enforcement. They use such vaguephrases as ‘proper action,’ ‘necessary action,’and ‘cool thinking’ to gloss over the commondiscretionary problems involved in dealing withdiscretionary situations.”26

To give another example, I was asked recently toreview an eastern State’s recruit training materialsconsidered by many to represent state-of-the-arttraining for community policing. One of the firstsections I checked was “Traffic Stops.” After anintroduction that included pieties stating that thepurpose of traffic stops was not to give tickets but,rather, to educate citizens about traffic safety, thesection outlined, in great detail, how to make atraffic stop. Properly, the section communicatedconsiderable concern about officer safety. (In-deed, it dwelled on the subject to such an extentthat I could not help but infer that the purpose oftraffic stops was to keep officers safe.) But that isall there was. Not one word—and I checked other

sections that dealt with similar issues—was givento advise students about how they should use theirdiscretion to give or not to give tickets. Yet, theonly discretionary police acts that the majority ofcitizens will experience or observe will involvetraffic violations.

My own personal experience with a traffic stopis instructive here. On the New Hampshire Ex-pressway, I was stopped going 72 miles per hourwhen the limit was still 55. When the officerapproached me, he made the typical request formy license and registration but asked me as well:“Was there any reason why you were going 72miles per hour?” It was a question. There wasnothing sarcastic, confrontational, or entrappingabout it. I said: “No, I was just smelling the barnand eager to get home. I have no excuse.” Hereplied: “I just can’t give you 17 miles over thelimit—7, sure. Maybe 10. But not 17.” I under-stood and told him so. He wrote the ticket for64, which was 9 over the limit. I didn’t like get-ting the ticket, but I had no sense of injustice oranger. His opening question, which I heard as asincere request for information about whetherI had some justification for speeding, acknowl-edged his discretionary authority and acknowl-edged, as well, that I might have had a goodreason, and he was prepared to consider it. Therespect with which the officer treated me engen-dered my respect for him.

The gaps in police training and guideline devel-opment that are so typical, however, retard thedevelopment of police knowledge, impede thedevelopment of genuine professionalism, dimin-ish the quality of police services, invite the useof personal whim as the basis for discretionaryjudgments, and unnecessarily expose police of-ficers and departments to liability suits. Such cir-cumstances are not unique to American policing.Joanna Shapland and Jon Vagg identify similarcircumstances in England:

[S]enior officers did not regard area beat work asa specialism within uniformed work or even as aparticularly skilled task, in the way that they sawIRV [immediate response vehicles] and public

17

order work. Special training, meetings on goodpractice, and manuals of relevant skills wereconspicuously absent—in contrast to the posi-tion in many other areas of police work. ABOs[area beat officers] were thrown on their own re-sources and were often supervised by sergeantswho had joined the force at the time when pa-trols in panda cars [small cars used by patrol of-ficers] were being introduced. These sergeantshad now to learn what area beat policing wasabout at the same time as their constables werefinding out how to do it. The result was ironic.The police themselves had precious little todraw on as a model other than television imagesof Dixon of Dock Green [a previously popularEnglish television show about police].27

The issue here is the state of the art of guidelines,policies, rules and regulations, and training mate-rials and not necessarily how police actuallywork. I continue to walk, ride in patrol cars, andride bicycles with officers. The skill, sensitivity,and capability of many police officers—such asI have described above and have experiencedthroughout my career in police research—isstriking. (I have seen terrible policing as well.)

Two points stand out. First, police are almostuniformly unable to articulate what they do, whythey do it, and how they do it. Most, when que-ried, resort to the jargon of the field: “commonsense,” “proper action,” and so forth. Second,virtually all of their order maintenance, peace-keeping, and conflict resolution activities are un-official. Except for personal notes officers takedown, no records are kept of these activities. Itis only a slight exaggeration to say that the onlyway that the activities can become official is ifsomeone files a complaint against the officer.28

There are signs of hope. In some cities, policemanagers go beyond paying mere lip service tocomplexity and discretion and find alternativemodels of training, supervision, and control that re-flect the realities of how police work is carried out.The San Diego Police Department’s policy state-ment on undocumented persons, for example,

provides sophisticated guidelines about a verycomplex issue confronting police departments incities near national borders—the extent to whichlocal police should look for violations of nationalimmigration laws. (See Appendix B, “San DiegoPolice Department Policy Statement.”) This prob-lem could easily have been left to officer “discre-tion,” that is, to ignore it administratively and letofficers “work it out” in the field. Instead, the de-partment thought through its values, mission, andfunctions and elaborated a policy that put publicsafety and harmony above aggressive attempts toferret out undocumented aliens.

What Is the Future of GuidelinesDevelopment?The shifts in policing now under way cannotoccur unless police departments are reorganizedto manage literal police work. This is why manycommunity policing programs often fail to meettheir stated goals. For most police departments,organizational and administrative “business asusual” is still the rule. Most police departmentshave developed few materials about the day-to-day problems that confront and, at times, plaguepatrol officers.

Moreover, a body of police management litera-ture is being developed that reflects more realis-tically the way police work is carried out.29

Issues such as the role accountability of midlevelmanagers, supervision in a highly decentralizedorganization, and leadership have been broughtto the center of police attention. One such ex-periment that has received enormous publicityis the interactive control mechanism calledCOMPSTAT that William Bratton borrowedfrom the private sector and adapted for NYPDmidlevel managers (precinct commanders).30

The method, first described by Robert Simonof the Harvard Business School, combines man-agement, supervisory, and peer control into ahighly visible process that rivets the midlevelmanager’s attention on precinct problems.31 Asexciting and promising as this innovation is,

18

however, it is still unclear whether such methodswill be adapted to line police or how depart-ments can develop guidelines to shape lineofficers’ discretion—the most widespread andinvisible discretion in criminal justice.

Despite the complexity of police work, genericsets of guidelines about the substantive problemswith which police deal—for example, disorderlybehavior in a downtown area—can be devel-oped. Such guidelines can serve as the basis forpolice training, supervision, and practice; iden-tify competent police work and provide the basisfor officer accountability; help to articulate agenuinely professional police point of view;and, yes, even be used to defend police actionsin litigation.

Notes1. For a discussion of “doing something, now,”see Bittner, Egon, The Functions of Police inModern Society, Cambridge, MA: Oelgeschlager,Gunn and Hain, 1980.

2. Marans, Steven, and Jean Adnopoz, ThePolice-Mental Health Partnership: A Commu-nity-Based Response to Urban Violence, NewHaven, CT: Yale University Press, 1995.

3. For the best summary of this research, seeMary Ann Wycoff, “The Role of MunicipalPolice Research as a Prelude for Changing It,”Mimeograph, Washington, DC: Police Founda-tion, 1982.

4. Goldstein, Herman, Problem-Oriented Polic-ing, Philadelphia, PA: Temple University Press,1990.

5. Many have questioned whether Bratton wasreally an advocate of community policing. SeeKelling, George L., and Catherine M. Coles,Fixing Broken Windows: Restoring Order andReducing Crime in Our Communities, NewYork: Free Press, 1996. Also see Bratton,

William W., and Peter Knobler, Turnaround:How America’s Top Cop Reversed the CrimeEpidemic, New York: Random House, 1998.

6. Newton, Jim, “Police: New York Police ClaimTheir New Militarism Has Led to the City’sStartling Drop in Crime,” Los Angeles Times,December 24, 1995, A1.

7. Ibid.

8. Ibid.

9. The Louima incident involved the torture ofa Haitian immigrant by New York City policeofficers—a truly ugly episode. In the Dialloincident, 4 officers fired 41 shots and killed anunarmed African man.

10. The Knapp Commission was a 1972 investi-gation of corruption in the New York City PoliceDepartment that many believe led to a depart-mental preoccupation with police corruption andabuse.

11. Kelling and Coles, Fixing Broken Windows,141–143.

12. Cheh, Mary M., “Constitutional Limits onUsing Civil Remedies to Achieve Criminal LawObjectives: Understanding and Transcending theCriminal-Civil Law Distinction,” The HastingsLaw Journal 42 (July 1991): 1325–1413.

13. For a discussion of civil initiatives, see Cheh,“Constitutional Limits.”

14. “Testilying” refers to systematic police lyingduring court testimony.

15. See Kelling and Coles, Fixing BrokenWindows, especially Chapter 3.

16. Wilson, James Q., and George L. Kelling,“Broken Windows: The Police and Neighbor-hood Safety,” The Atlantic Monthly (March1982): 29–38.

19

17. Skogan, Wesley, Disorder and Decline: Crimeand the Spiral of Decay in American Neighbor-hoods, New York: The Free Press, 1990.

18. “Operation Matrix” was a joint police/socialwork effort to reduce disorder and deal withhomelessness, especially in parks.

19. Livingston, Debra, “Police Discretion andthe Quality of Life in Public Places: Courts,Communities, and the New Policing,” ColumbiaLaw Review 97 (3) (April 1997): 551–672.

20. Sherman, Lawrence W., James W. Shaw, andDennis P. Rogan, The Kansas City Gun Experi-ment, Research in Brief, Washington, DC: U.S.Department of Justice, National Institute ofJustice, January 1995, NCJ 150855.

21. For a detailed discussion of the legal issuesand how they were resolved, see Kelling andColes, Fixing Broken Windows, 38–69.

22. For a brief description of the gun confisca-tion program in Kansas City and an anecdotefrom Indianapolis, see Sherman, Shaw, andRogan, The Kansas City Gun Experiment, 5–6.

23. Goldstein, Herman, “Confronting the Com-plexity of Police Functioning,” in Discretion inCriminal Justice: The Tension Between Individu-alization and Uniformity, ed. Lloyd E. Ohlin andFrank J. Remington, Albany, NY: State Univer-sity of New York Press, 1993: 53; and Goldstein,Herman, “Police Policy Formulation: A Proposalfor Improving Police Performance,” MichiganLaw Review 65 (April 1967): 1123.

24. Kelling, George L., and William J. Bratton,“Implementing Community Policing: The Ad-ministrative Problem,” in Perspectives on Polic-ing, No. 17, Cambridge, MA: U.S. Department ofJustice and Program in Criminal Justice Policyand Management, John F. Kennedy School ofGovernment, Harvard University, July 1993.

25. Kelling and Coles, Fixing Broken Windows,182–185.

26. Vandall, Frank, Police Training for ToughCalls: Discretionary Situations (limited edition),Atlanta, GA: Emory University, Center forResearch in Social Change, 1976.

27. Shapland, Joanna, and Jon Vagg, Policing Bythe Public, London: Routledge, 1988.

28. Kelling, George L., “How to Run a PoliceDepartment,” City Journal 5 (4) (Autumn 1995):35–44.

29. See, for example, Sparrow, Malcolm K.,Mark H. Moore, and David M. Kennedy, Be-yond 911: A New Era for Policing, New York:Basic Books, 1990.

30. Bratton and Knobler, Turnaround.

31. Simon, Robert, The Mental Health Practi-tioner and the Law: A Comprehensive Hand-book, ed. Lawrence E. Lifson and Robert I.Simon, Cambridge, MA: Harvard UniversityPress, 1998.

21

The Discovery of Discretion and Its Meaning

In 1965, Frank Remington wrote, “The policeshould play a major role in fashioning andimplementing a proper law enforcement policyfor their community.”1 Remington drew an anal-ogy between the Federal Trade Commission, anadministrative agency with the responsibility todevelop enforcement policy, and local policedepartments, which have a similar responsibility.This was a radical departure from past thinkingabout policing. Legislatures and city councilsmade laws, and the police responsibility was toenforce them impartially. A generation of policehad been reared on the idea that law enforce-ment was what they did: individuals broke thelaw, the police arrested them. The police job wasjust that simple.

Implicit in Remington’s proposal was an alterna-tive view of the police as a de facto administra-tive agency of city government whose work wascharacterized by repeated use of discretionaryjudgment. This discretion existed not only at ex-ecutive levels but also at every level of the policeorganization, especially with the practicing po-lice officer. Remington’s views were based onresearch he directed during the late 1950s, spe-cifically the American Bar Foundation (ABF)Survey of Criminal Justice.

Until the 1950s and 1960s, surveys of police andcriminal justice agencies relied on official re-ports and statistics.2 Generally conducted on acity level by local elites aided by professionalconsultants like Bruce Smith, police surveysstudied the extent to which local police depart-ments followed the reform model of policing,especially their adherence to its organizationaland administrative principles. What police did—their literal work—remained out of public view.

Survey of Criminal Justice AgenciesIn 1953, Supreme Court Justice Robert H. Jack-son, believing criminal justice to be in a state ofcrisis, called for a major national study of crimi-nal justice agencies by the American Bar Asso-ciation (ABA). At first administered by the ABAand funded by the Ford Foundation, the surveyalso became the first project of the ABF, theABA’s newly created research arm. As first con-ceived, the study was designed to apply method-ologies used in previous local surveys: officialdata would be analyzed to learn the extent towhich agencies adhered to prescribed standards.

Not atypically, expert consultants devised theoriginal project and secured its funding. O.W.Wilson, the dominant police theorist of the 20thcentury, for example, compiled the checklist thatwas to be the foundation for the study. Otherexperts helped devise similar plans for examin-ing other criminal justice agencies.3 HermanGoldstein, who was a staff member of the origi-nal survey, described the early plans:

If the ABF survey had followed its originalplan for that portion of its inquiry relating tothe police, it would not have added much tothe sparse knowledge then available. Like thepolice field itself, the plan was heavily influ-enced by the prevailing perception of whatwas important in policing—the technical andadministrative aspects of running a policeagency. The detailed agenda for inquiry iden-tified 15 different categories of informationthat were to be systematically acquired, withall but two categories relating to the organiza-tion, administration, staffing, and equipping ofa police agency. The product would have beenan inventory of the degree to which the police

22

agencies conformed with the then-prevalentstandards for managing a police agency.4

Under the leadership of Remington and LloydOhlin (Emeritus Touroff-Glueck Professor ofCriminal Justice at Harvard Law School), how-ever, the survey went in a different direction.In some respects, the new direction was an out-growth of a serendipitous mix of the Universityof Wisconsin Law School’s “law in action”tradition as embodied by Remington and theChicago School of Sociology’s tradition of fieldobservation as represented by Ohlin, a Ph.D.from the University of Chicago.

Ohlin notes that Remington, the first staff direc-tor of the study, “decided from the outset thatthe ABF survey would be different from earlierstudies, much less concerned with official sum-mary statistics and more concerned with theanalysis of the criminal justice system in dailyoperation.”5 In other words, the survey wouldstudy law as it operated rather than law as it wasfound in the books and, as a consequence, wouldrely heavily on observational data of low-leveldecisionmaking. Field staff filled out the check-lists devised by the experts to be true to theoriginal proposal, but only for that purpose.6

Remington and Ohlin’s decision to reshapethe survey transformed it into one of the mostimportant social/legal research studies of the20th century.

Initially, three sites were selected for the study:Kansas, Michigan, and Wisconsin. Additionalsites were to be added later. After the surveybegan in 1956, however, researchers, confrontedby the enormity of the study and the data set,decided to limit their study to the original threesites.

Study FindingsThe study focused on the line personnel incriminal justice agencies—police, prosecutors,judges, and corrections officers—conductingtheir routine work. Conventional thinking wasturned on its head:

● Discretion was found to be used at all levelsof criminal justice organizations. The ideathat police, for example, made arrest decisionssimply on the basis of whether or not a lawhad been violated—as a generation of policeleaders had led the public to believe—wassimply an inaccurate portrayal of how policeworked.

● Low-level decisionmaking by line personnelin light of practical and real-life considera-tions was found to be a significant contributorto the crime control and problem-solving ca-pacity of criminal justice agencies. This wastrue not only for police but for prosecutorialand other decisionmaking personnel as well.

● Criminal law was used to solve many socialproblems, not just serious crimes.

● Behaviors designated as unlawful in criminalcodes, such as assault, were found to be ex-traordinarily diverse in nature and includedeverything from private debt settlement andspousal abuse to attacking strangers.

● The policies of each criminal justice agencywere found to have an impact on other crimi-nal justice agencies.7

Given the pervasive use of discretion in daily po-lice work, the most obvious question the surveyraised was why legal scholars, criminal justiceacademics, and practitioners had failed to noteit in the past—why were these findings such asurprise.8 The answer, of course, testified to thepower of reform ideology and its imagery. AsSamuel Walker notes:

The intriguing question, of course, is how 40crime commissions, which involved the bestminds of their day—Felix Frankfurter andRoscoe Pound, among others—could fail tosee such a crucial feature of the administra-tion of justice. The answer is that phenomenawhich are self-evident to one generation arenot necessarily evident to others. This high-lights the role of paradigms in scientificresearch. Paradigms describe observed phe-nomena, define problems, and guide research.

23

Phenomena that fall outside the prevailingparadigm either are not noticed at all or aredismissed as unimportant and not worthy ofinvestigation. So it was with discretion forearly crime commissions.9

Reflecting on these findings, Kenneth CulpDavis, emphasizing the central role of police inthe justice process, estimated that about half ofthe discretionary decisions made by criminaljustice agencies were made by police.10 Headded: “The police are among the most impor-tant policymakers of our entire society. And theymake far more discretionary determinations inindividual cases than any other class of adminis-trators; I know of no close second.”11

Throughout the 1960s and 1970s, scholars, fas-cinated by how low-level, low-visibility discre-tion was used by patrol officers, studied policefunctioning. Not surprisingly, study after studyconfirmed the ABF findings: police work iscomplicated, a small portion of police time isspent on criminal matters, and police use discre-tion throughout their work.12

Consequently, some scholars and policymakersbegan to study how to control or shape discre-tion. At this point, their thoughts began to di-verge. Some researchers focused on the abusesof authority that were reported in the survey’sfindings. A few advocated eliminating policediscretion altogether, at least in the decision toarrest. Joseph Goldstein stated this point of viewmost strongly:

The ultimate answer is that the police shouldnot be delegated discretion not to invoke thecriminal law. . . . [T]he police should operatein an atmosphere which exhorts and com-mands them to invoke impartially all criminallaws within the bounds of full enforcement. . . .Responsibility for the enactment, amendment,and repeal of the criminal laws will not, then,be abandoned to the whim of each policeofficer or department, but retained where itbelongs in a democracy—with elected repre-sentatives.13

Eliminating police discretion completely, evenin the decision not to arrest, has always been aminority view. Goldstein’s view has been mostintensely debated as it pertains to domestic vio-lence—a problem largely invisible to the publicprior to the ABF report. Prosecutorial discretionin the handling of domestic violence was alsohighlighted by early ABF reports.14 Since then,discussion, debate, and policy regarding policeand prosecutorial handling of domestic violencehave changed directions several times. A reviewof research on the impact of criminal justiceresponses to domestic violence testifies, morethan anything, to the complexity of the problemand the need for preventive problem-solvingapproaches.15

Notes1. Remington, Frank, J. “Police in a DemocraticSociety,” Journal of Criminal Law, Criminologyand Police Science (1965): 361.

2. Leonard, V.A., Survey of the Seattle PoliceDepartment, Seattle, WA, 1945.

3. Personal interview with Herman Goldstein,Madison, WI, September 2, 1994.

4. Goldstein, Herman, “Confronting the Com-plexity of Police Functioning,” in Discretion inCriminal Justice: The Tension Between Individu-alization and Uniformity, ed. Lloyd E. Ohlin andFrank J. Remington, Albany, NY: State Univer-sity of New York Press, 1993: 28.

5. Ohlin, Lloyd E., “Surveying Discretion byCriminal Justice Decision Makers,” in Discre-tion in Criminal Justice: The Tension BetweenIndividualization and Uniformity, ed. Lloyd E.Ohlin and Frank J. Remington, Albany, NY:State University of New York Press, 1993: 6.

6. Personal conversations with HermanGoldstein.

7. The findings were first used in Kelling,George L., “Toward New Images of Policing:

24

Herman Goldstein’s Problem-Oriented Polic-ing,” Law & Social Inquiry 17 (3) (Summer1992): 545.

8. The final reports were published in fivebooks, all edited by Frank Remington and pub-lished by Little, Brown and Company (Boston,MA): Lefave, Wayne, Arrest: The Decision toTake a Suspect Into Custody (1965); Newman,Donald J., Conviction: The Determination ofGuilt or Innocence without Trial (1966); Tiffany,Lawrence P., Donald M. McIntyre, Jr., andDaniel Rotenberg, Detection of Crime: Stoppingand Questioning, Search and Seizure, Encour-agement and Entrapment (1967); Dawson, Rob-ert O., Sentencing: The Decision as to the Type,Length and Conditions of Sentence (1969); andMiller, Frank W., Prosecution: The Decision toCharge a Suspect with a Crime (1970).

9. Walker, Samuel, Taming the System: TheControl of Discretion in Criminal Justice, 1950–1990, New York: Oxford University Press, 1992:7–8.

10. Davis, Kenneth Culp, Discretionary Justice:A Preliminary Inquiry, Baton Rouge, LA:Louisiana State University Press, 1969: 166.

11. Ibid., 222.

12. Wycoff, Mary Ann, The Role of MunicipalPolice Research as a Prelude to Changing It,Washington, DC: Police Foundation, 1982.

13. Goldstein, Joseph, “Police Discretion Not toInvoke the Criminal Process: Low-VisibilityDecisions in the Administration of Justice,” YaleLaw Journal 69 (4) (1960): 587.

14. Miller, Frank, Prosecution: The Decision toCharge a Suspect with a Crime, ed. Frank J.Remington, Boston: Little, Brown and Com-pany, 1970.

15. For a summary of the thinking in the area ofdomestic violence and an example of an indi-vidual author’s evolution in thinking, see Parnas,Raymond I., “Criminal Justice Responses toDomestic Violence Issues,” in Discretion inCriminal Justice: The Tension Between Individu-alization and Uniformity, ed. Lloyd E. Ohlin andFrank J. Remington, Albany, NY: State Univer-sity of New York Press, 1993: 175–210.

25

The Police Guidelines Movement

Other scholars have seen police discretion interms that have differed from those of JosephGoldstein. Frank Remington, Herman Goldstein,and James Q. Wilson, among others, were awareof the possibilities of abuse but understood boththe inevitability of discretion and the inherentopportunities for problem solving it offered po-lice and other criminal justice agencies. Wilson,for example, wrote in 1968: “The patrolman, inthe discharge of his most important duties, exer-cises discretion necessarily, owing in part to hisrole in the management of conflict and in part tohis role in the suppression of crime.”1 The issuewas to eliminate unnecessary discretion andenhance and shape necessary discretion.

During the mid-1960s, about the same time theAmerican Bar Foundation’s Survey of CriminalJustice reports were first published, PresidentLyndon Johnson created the Commission onLaw Enforcement and Administration of Justice,his answer to Barry Goldwater’s challengesabout crime during the 1964 presidential race.The Commission was set up to study crime andsociety’s response to it. Its findings, as presentedin its report entitled The Challenge of Crime in aFree Society, have shaped criminal justice think-ing during the past 35 years.

How the findings in the ABF’s Survey on Crimi-nal Justice and the Commission’s Challenge ofCrime in a Free Society support and contradicteach other is not at all clear. The history of thisera in criminal justice has not yet been written.2

Clearly the commission captured popular andprofessional thinking and eclipsed the ABF’swork—popularly, professionally, and sadly, itshould be noted, academically.

One reason why the ABF’s survey did not re-ceive the recognition it deserved was because it

examined an enormous amount of data and firstbegan presenting survey data in 1965, approxi-mately 9 years after the initial work began(1956) and almost simultaneously with the workand publications of the President’s Commission(1967). (The five volumes of the survey werepublished between 1965 and 1970.)

The ABF’s survey and the President’s Commis-sion came to be linked in a number of ways.Lloyd Ohlin, one of the survey leaders, becamean associate director of the Commission’s staff.Both Frank Remington and Herman Goldstein,also survey leaders, consulted with Commissionstaff and its director James Vorenberg. Ideas firstraised in the survey found their way into theCommission’s report, especially the use of dis-cretion and the idea of a criminal justice system.Yet, in the case of police, the President’s Com-mission kept itself anchored in the reform tradi-tion. The report emphasis is response time,administrative improvements, and police as thefront end of the criminal justice system. It fo-cuses police, as well as other criminal justiceagencies, on case processing—an orientationthat atomizes and obscures complicated neigh-borhood and community problems.

On balance, system improvement won the dayin the Commission’s report. Relations with thecommunity had to be improved and better per-sonnel had to be recruited, yet the basic policingparadigm went largely unchallenged: policebusiness was to ride around in cars and respondto calls for service. Police in the 1960s, whowere insular and suspicious of outsiders, espe-cially the liberal academics who staffed theCommission, nonetheless found their hand hadbeen strengthened. Police were now part of acriminal justice “system,” and their businesswas clearly defined as arresting and processing

26

offenders. The full impact of the ABF’s surveybecame apparent in the late 1980s and early1990s, when community and problem-orientedpolicing would refocus attention on the truecomplexity of policing.

Nonetheless, minor themes in the Commission’sreport pointed to the impact of the ABF effortsin its call for guidelines for controlling policediscretion:

Police departments should develop and enun-ciate policies that give police personnel spe-cific guidance for the common situationsrequiring exercise of police discretion. Poli-cies should cover such matters, among others,as the issuance of orders to citizens regardingtheir movements or activities, the handling ofminor disputes, the safeguarding of the rightsof free speech and free assembly, the selectionand use of investigative methods, and thedecision whether or not to arrest in specificsituations involving specific crimes.3

The truly remarkable outcome of the President’sCommission can be found in its companion re-port, Task Force Report: The Police. Chapter 2,which was authored by Frank Remington andHerman Goldstein, remains the definitive sourceon police policymaking and structuring discre-tion.4 The section entitled “The Need to Recog-nize the Police as an Administrative Agencywith Important Policymaking Responsibility”clearly is not only out of the “reform box” butalso so far ahead of the field that its full impactis yet to be felt.5 Remington and Goldsteinwrote:

There are two alternative ways in whichpolice can respond to the difficult problemscurrently confronting them:

(1) The first is to continue, as has been true inthe past, with police making important deci-sions, but doing so by a process which canfairly be described as “unarticulated improvi-sation.” This is a comfortable approach, re-quiring neither the police nor the communityto face squarely the difficult social issues

which are involved, at least until a crisis—likethe current “social revolution”—necessitatesdrastic change.

(2) The second alternative is to recognizethe importance of the administrative policy-making function of police and to take appro-priate steps to make this a process which issystematic, intelligent, articulate, and respon-sive to external controls appropriate in ademocratic society; a process which antici-pates social problems and adapts to meetthem before a crisis situation arises.

Of the two, the latter is not only preferable; itis essential if major progress in policing is tobe made, particularly in the large, congestedurban areas.6

Included in Remington and Goldstein’s formula-tion is a diagrammatic representation of thepolicymaking process. (See “Formulation andExecution of Police Policy.”)7

For the most part, outsiders to policing were thefirst to pick up on Remington and Goldstein’sline of thought and advocate the development ofstrong guidelines similar to those used by ad-ministrative agencies. These advocates includedscholars, reformers, lawyers, a few renegade po-lice chiefs, and others who believed that beyondthe guidance that courts, legislatures, prosecu-tors, and local officials could provide to shapeofficer discretion, police themselves had to de-velop their own professional guidelines. Guide-lines, as distinguished from administrative rulesin the minds of those advocating formal andpublic policymaking, would assist police in con-trolling crime, increase police responsivenessand accountability, allow for police input intopolicymaking and legal deliberations, and, asWarren LaFave suggests, encourage courts torespect police professional judgment.8

Convinced that the development of guidelineswas central to improving police performanceand accountability, the Police Foundation—created by the Ford Foundation in 1970 with a$30 million endowment—funded two early

27

Formulation and Execution of Police Policy

Decision to review policy

Identification of need for policy asdetermined by: court decisions, newlegislation, citizen complaints, analysisof crime and social problems, andanalysis of existing field practices

Referral by head of police depart-ment to planning and researchunit for study in cooperation withdivisions and staff specialists

Referral of findings to stafffor consideration

Consultation by staff with:chief political executive;neighborhood advisory committees;and prosecution, court, corrections,and juvenile authorities

Formulation of policy byhead of police department

Promulgation of policy—

To community through:Published policy statements and neighborhoodadvisory committee meetings

To personnel through:Training manual and orders

Execution of policy by fieldpersonnel: controlled throughsupervision and inspection

Evaluation of policy based upon:court decisions, new legislation,citizen complaints, analysis ofcrime and social problems, andanalysis of existing field practices

Adapted from: President’s Commission on Law Enforcement and Administration of Justice, The Challenge ofCrime in a Free Society: A Report by the President’s Commission on Law Enforcement and Administration ofJustice, Washington, DC: U.S. Government Printing Office, 1967: 105.

28

major efforts to develop guidelines. One was aprogram in rulemaking at Arizona State Univer-sity Law School under the direction of then fac-ulty member Gerald Caplan. The second was aneffort in the Boston Police Department (BPD)under the direction of Sheldon Krantz who, atthat time, directed the Center for CriminalJustice at Boston University.

Caplan’s Arizona State project helped 10 policedepartments develop model policies that would“channel” police discretion through “carefullyresearched and articulated rules—rules formu-lated in categories meaningful to a policeman,rather than law textbook groupings not relatedto the situations likely to be encountered on thenext tour of duty.”9 Krantz’s efforts in the BPDwere more focused, concentrating initially onthree areas of criminal investigation—searchwarrants, motor vehicle searches, and searchesincidental to arrest—and involved the use of taskforces composed of line as well as administra-tive staff. Later the efforts expanded to includeadditional investigative areas as well as drugenforcement issues.

Herman Goldstein was prescient when he wrotein 1967:

The overall picture, however, reflects a reluc-tance on the part of police administrators toestablish policies to fill the existing void. Thisreluctance is in sharp contrast to the strongtradition within police agencies for promul-gating a variety of standard operating proce-dures to govern the internal management ofthe police force.10

Academics, scholars, and lawyers might wantpolice to develop guidelines, but the insularityof police and their reluctance during the 1950s,1960s, and much of the 1970s to allow anyoutsider to get involved in the business of polic-ing doomed efforts such as the Caplan andKrantz projects.

Although the development of guidelines wasa promising means of strengthening police

practice, police would have little to do with suchefforts. In retrospect, Caplan remarked: “Thetiming was just not right. We could get a fewpolice staff members to get involved in eachdepartment we worked with, but the resistanceto outsiders in police departments was so strongthat we really couldn’t get much done.”11

With no reports and few products—model ruleswere developed but, for the most part, shelved—the project, like team policing in many cities,simply faded away. Likewise, although Krantzfound he could generate enthusiasm in task forcemembers, he could not generate any enthusiasmfor guidelines in either the BPD or other depart-ments. In retrospect, Krantz believes that a focuson routine police work, rather than the issuesstudied, would have been a more profitable ap-proach. Even at that level, however, the field justwas not ready.12

To be sure, a few chiefs of police understood theneed for guidelines and worked to develop them,but they were far from the norm, especially dur-ing the 1960s and 1970s. Working with Caplan,Chief Jerry Wilson in Washington, D.C., devel-oped guidelines for eyewitness identificationprocedures during the late 1960s to ensure po-lice practices that were both effective and legal.13

Likewise, Commissioner Patrick V. Murphy, in1971, initiated guidelines in New York City tomold police use of force, especially deadlyforce. Two of the most innovative chiefs of thisera were Robert M. Igleburger of Dayton, Ohio,and Frank Dyson in Dallas, Texas. While Caplanand Krantz focused on model rules and regula-tions that could be adopted by individual policedepartments, Igleburger and his administrativeassistant Frank A. Schubert, now associate pro-fessor of criminal justice at Northeastern Univer-sity, developed guidelines that were location andproblem specific.14 Funds for this effort wereprovided by the Police Foundation in 1972.

Akin to Krantz’s work, line police officers whohad to deal with specific problems were inte-grated into the decisionmaking process. Going

29

beyond earlier efforts, however, Igleburger andSchubert included in each task force citizenswho were affected by the problems. For ex-ample, during the early 1970s, Dayton experi-enced friction between students and residents ina working class neighborhood adjacent to theUniversity of Dayton. Not atypically, studentswho wanted to live off campus sought inexpen-sive rooms and apartments in the neighborhoodssurrounding the campus. The conflict arose be-tween working parents who wanted the neigh-borhood to quiet down by 10 p.m. or 10:30 p.m.,especially on weeknights, and students who, af-ter a day of classes and studying, wanted to en-gage in other activities until the early morninghours. The result, of course, was repeated callsto the Dayton police. Disturbance violations ofone form or another, especially noise violations,were the most common complaints.

As part of the police/public policymakingproject, Schubert, along with neighborhoodofficers, polled residents and students, met re-peatedly with both, and learned more about theneighborhood and its problems. Once the policeunderstood neighborhood standards about noiseand other issues that could cause conflict be-tween students and other residents, they devel-oped guidelines that police would use to enforcethose standards. The standards and guidelineswere then printed and widely distributed to allthe concerned parties. (See “Dayton, Ohio,Issues Guidelines Related to Student andNonstudent Residents.”)

As Schubert makes clear, issuance of the guide-lines was not a once-and-for-all determination.Each year the process was repeated to updatepolicies and to ensure that new students, resi-dents, and police understood and endorsedpolice policies regarding enforcement of banson noise and other forms of disturbance.15

Just as the time was not right for model rule-making during the 1970s, the time was not rightfor public policymaking in Dayton either. WhenIgleburger retired as chief of the Dayton PoliceDepartment in 1973, the program faded away.

When I consulted with a blue-ribbon citizengroup commissioned to examine the tragic deathof an innocent citizen that occurred during abotched special unit operation during the late1980s, no traces of police/public policymakingcould be found. Reports on the police/publicpolicymaking project were never published ordisseminated, although Igleburger and Schubertpublished one brief article in 1972 on shapingdiscretion through police/public partnerships.16

Generally speaking, academics, foundations,and others lost interest in guidelines as well,although publication of a few articles aboutcontrolling police discretion continued into the1980s, mostly in law journals. A few depart-ments developed guidelines, but this was not thegeneral rule.

The idea of model policies, rules, and regula-tions was somewhat revived by development ofthe accreditation movement in policing duringthe 1980s. However, the process had relativelylittle impact on how line officers should use orhave used their discretion. Research interestshifted to experiments in police tactics, largelyas a result of the early work of the Police Foun-dation in Kansas City, Missouri, Cincinnati,Ohio, and Newark, New Jersey. “Big” issues,such as preventive patrol and response time, or“sexy” issues, such as use of force and domesticabuse, largely dominated the attention of policeresearchers.

During the late 1970s and 1980s, as the PoliceFoundation’s financial resources dwindled, theNational Institute of Justice (NIJ) assumed lead-ership in guiding research, but shaping discre-tion was not the commanding issue that it hadonce been. Saying this should not diminish thevalue of much Police Foundation and NIJ-funded research: it contributed enormously tothe understanding of police work and reconsid-eration of the reform strategy. Nonetheless, acrucial movement in policing—developingguidelines to shape police discretion—lost itsimpetus because the field simply was not yetready for it during the 1970s. Police businesswas still police business.

30

Dayton, Ohio, Issues Guidelines Related to Student and Nonstudent Residents

The police general order read as follows:

The Department of Police has as one of itsprimary responsibilities the development andmaintenance of peaceable relations between citi-zens. Police officers are expected to act in an ap-propriate and effective manner so as to maintaina standard of public order which is generallyacceptable to the lifestyles of both the studentand non-student residents of the area in thevicinity of the University.

It should be recognized that divergent lifestylesfrequently generate conflict which is not condu-cive to the well-being of either students or non-students, and that compromises must be madeby all concerned. Other communities have at-tempted to deal with this campus-communitysituation by general suppression through use ofpolice power. The record seems to indicate thatthis response has not resolved the problems but,rather, increased the polarization of the parties.

I believe that we should seek to develop a policeresponse to the problems which will permit anevolutionary and improving relationship basedon respect and a concern by all for the mutualwelfare of the community. Such a police posturecan only be justified, however, if the situationshows improvement.

The following problems were identified bystudent and non-student residents meeting withpolice personnel as being matters of legitimatepolice concern.

Fires. The residents have indicated that theybelieve fires should not be permitted in the streetsas they constitute a serious threat to life andproperty. Residents expect police crews to act ap-propriately to prevent the lighting of fires and toput out existing fires.

Loud Noise After Midnight. The residents haveindicated that it is reasonable to expect that loudnoises should cease after midnight. Both studentsand non-students believe this compromise to be

fair and acceptable. I expect police crews to actso as to protect the rights of persons to peace andquiet in their residences after midnight.

Traffic Blockages. The residents recognize thatlarge numbers of students in attendance at partiesmay require the temporary use of the streets. It isimportant, however, that vehicles be permitted topass through these crowds so that residents cangain access to their property and so emergencyvehicles can pass. Police crews are expected toensure vehicles are not being blocked unless bar-ricades have been erected in accordance with ourblock party procedures.

It should be emphasized that the burden for com-plying with the standards discussed in this memoshould be on the residents, both student and non-student. The problem is unlikely to be resolved bypolice force; however, in the event we need to re-spond to situations such as those discussed above,we will act to implement the identified standardas the situation requires.

We are committed to persuasion as the primarymethod of dealing with these problems because itis constructive, and not destructive as is the con-frontational approach. Conflict management per-sonnel are to be called up to assist whenever it isfelt they can be of aid to the crew.

The Second District Sergeants and beat crewsare encouraged to confer with residents, ProjectSouth (telephone number), and the University Stu-dent Government so as to make use of their ser-vices in dealing with problems in a “non-police”fashion to the greatest extent possible.

The officers are to feel free to consult with theConflict Management Director, their superiors,or myself if they have any questions about thispolicy. It is the responsibility of the Supervisors toassure all officers have read and understand thisorder and respond accordingly.

Director of Police

31

Notes1. Wilson, James Q., Varieties of Police Behav-ior, Cambridge, MA: Harvard University Press,1968: 278.

2. Kelling, George L., “Toward New Images ofPolicing: Herman Goldstein’s Problem-OrientedPolicing,” Law & Social Inquiry 17 (3) (Summer1992): 539–559; and Kelling, George L., andCatherine M. Coles, Fixing Broken Windows:Restoring Order and Reducing Crime in OurCommunities, New York: Free Press, 1996.

3. President’s Commission on Law Enforcementand Administration of Justice, The Challengeof Crime in a Free Society: A Report by thePresident’s Commission on Law Enforcementand Administration of Justice, Washington, DC:U.S. Government Printing Office, 1967: 104;guidelines were also strongly endorsed by thereport of the American Bar Association’s Com-mittee on the Urban Police Function (AmericanBar Association, “The Urban Police Function,”Approved Draft, Institute of Judicial Administra-tion, New York, 1973) and the report of theKerner Commission, the National AdvisoryCommission on Civil Disorder created to studyurban rioting (Kerner Commission, Report of theNational Commission on Civil Disorder, NewYork: Bantam Books, 1968).

4. Remington, Frank, and Herman Goldstein,“Law Enforcement Policy: The Police Role,”Task Force Report: The Police, Washington, DC:U.S. Government Printing Office, 1967: 13–41.

5. Wilson, Varieties of Police Behavior, 285.In the final chapter, “Conclusions and PolicyImplications,” which anticipates the problemsinherent in developing police discretionaryguidelines, Wilson writes: “Politics, in thebroad sense of community involvement inpolicymaking, will be used to achieve whatadministrative rationalization cannot.”

6. Remington and Goldstein, Task Force Report:The Police, 18.

7. President’s Commission on Law Enforcementand Administration of Justice, The Challenge ofCrime in a Free Society, 105.

8. See LaFave, Warren, “Police Rule Making andthe Fourth Amendment,” in Discretion in Crimi-nal Justice: The Tension Between Individualiza-tion and Uniformity, ed. Lloyd E. Ohlin andFrank J. Remington, Albany, NY: State Univer-sity of New York Press, 1993: 211–277; andCaplan, Gerald M., “The Case for Rule-makingby Law Enforcement Agencies,” Law and Con-temporary Problems 36 (4) (1971): 500–514.

9. President’s Commission on Law Enforcementand Administration of Justice, The Challenge ofCrime in a Free Society, 502.

10. Goldstein, Herman, “Police Policy Formula-tion: A Proposal for Improving Police Perfor-mance,” Michigan Law Review 65 (April 1967):1123–1146.

11. Personal telephone conversation with GeraldCaplan, January 1994.

12. Personal interview with Sheldon Krantz,Washington, DC, July, 13, 1996; there wereother attempts at administrative rulemaking,including a collaboration between Texas officialsand the International Association of Chiefs ofPolice. See Ferris, Saadi, “Administrative Rule-making: Police Discretion,” Police Chief (June1974): 70–71.

13. Wilson, Jerry V., and Geoffrey M. Alprin,“Controlling Police Conduct: Alternatives to theExclusionary Rule,” Law and ContemporaryProblems 36 (4) (1971): 488–499.

14. This section paraphrases a discussion inKelling and Coles, Fixing Broken Windows, 191–192.

15. Personal interview with Frank A. Schubert,Boston, January 24, 1994.

16. Igleburger, Robert M., and Frank A.Schubert, “Policy Making for the Police,” Ameri-can Bar Association Journal 58 (March 1972):307–310.

33

that managers dump problems into officers’ laps,stick their heads in the sand, and then, when theinevitable legal challenges arise, say that dealingwith homelessness is a hopeless task—a positionthat police managers took earlier with graffiti—and then return to business as usual. Fortunately,things did not work out this way. Although itrequired a replacement of the transit police lead-ership to implement the policies and tactics de-veloped by the study group, the outcome wasrestoration of order and astonishing drops inthe level of serious crimes, most notably a nearly75-percent reduction in the number of robberies.2

Beyond the problem-solving aspects of the expe-riences, however, I returned to the literature ondiscretion to review the state of the art of policedepartment rules and regulations and generalorders. The study group anticipated that advo-cates for the homeless and civil libertarianswould sue the transit police once order mainte-nance efforts were initiated, and that selectiveenforcement would be a major issue. Moreover,advocates made it clear that they would challengeour attempts to control panhandling using FirstAmendment free speech as the pivotal issue.

With awareness that the courts would look overour shoulders, it was absolutely essential thatwe deal with how we would shape discretionaryguidelines and why panhandling, not just so-called aggressive panhandling, should be bannedin the subway. Even technical legal issues wereconsidered because we had to be prepared todemonstrate a “compelling governmental issue,”if courts found panhandling to be a protectedform of speech. Justifications had to be devel-oped for police actions in the subway that wouldmake the typical knee-jerk response—“throwthe bums out” from the right and “criminalizingthe homeless” from the left—untenable to thegeneral public.

My interest in complexity, discretion, andpolicymaking was restimulated by my experi-ences working with New York City’s Transit Po-lice Department during the late 1980s and early1990s.1 Confronted with growing numbers ofhomeless people living in the subway system—a mishmash of interrelated problems if thereever was one—police managers proposed obvi-ously contrived “cleaning operations” that wereto be “supported” by line officers. Maintenancepersonnel with high-power hoses were to go intothe subways where there were concentrations ofhomeless people to “clean” the subways. Police,in support of these cleaning operations, wouldeject the homeless. Dean Esserman, then legalcounsel for the transit police and now chief ofthe Stamford, Connecticut, Police Department,dubbed the proposed operation “commandocleaning.” At my urging, David Gunn, thenPresident of the New York City Transit Author-ity, and Robert Kiley, then Chairman of theBoard of the Metropolitan TransportationAuthority, rejected the plan for what it was—a sham.

Instead, David Gunn created a police studygroup, which I advised. It consisted of line offi-cers, supervisors, and managers of the transitpolice as well as civilian employees of theNew York City Transit Authority. The studygroup, using a variation of Herman Goldstein’sproblem-solving methodology to address theissue, conducted patron and focus group surveysthat both helped to define the nature of thesubway’s problems and provided public reviewof suggested police tactics.

It was quite apparent that line officers desper-ately wanted administrative guidance andauthority but were skeptical that it would beforthcoming. Their initial view, put simply, was

Development of Police Guidelines

34

In short, we were making public policy about aspecific set of problems in a particular set of lo-cations that was similar to the process identifiedby Remington and Goldstein and implementedin Dayton. Reflecting back on the process, itbecame clear that our efforts to develop andimplement policies to restore order in the sub-way attempted to meet at least 11 standards.The policies had to:

● Recognize the complexity of police work.

● Acknowledge that police will use discretion.

● Recognize and confirm how police work isconducted.

● Advance a set of values that may be appliedto the substantive work issue at hand.

● Put forward existing research, facts, or dataabout the substantive issue at hand.

● Undergo development by practicing policeofficers and citizens.

● Undergo public promulgation in a mannerclear to officers, the general public, commu-nity stakeholders, and the courts.

● Include rules about what officers should notdo.

● Emphasize police adherence to a process(application of knowledge, skills, and values),rather than any predictable outcome, becauseoutcomes of police interventions are oftenwildly unpredictable regardless of officers’skills, intent, and values.

● Establish accountability standards that iden-tify competent and/or excellent performance,violations of organizational rules, and incom-petent or uncaring work, including perfor-mance within organizational rules.

● Receive recognition as an ongoing andcontinuing process.3

These standards had broader applications andwere presented to line personnel in training ses-sions and experiments with their development inseveral police departments.

Recognizing the Complexity ofPolice WorkThe complexity of police work has two dimen-sions: the complexity of the situations or prob-lems confronting police, and the complexity ofpolice responses to those situations.4 Panhan-dling is an example of how complex a seeminglysimple act can be. It is the subject of consider-able litigation between civil liberties groups andpolice departments across the country, a clearexample of external groups using litigation andthe courts to limit police discretion. (Advocacygroups do so with good reason—police haveused vagrancy, loitering, and panhandling lawsto harass citizens and discriminate againstgroups in the past.)

Until recently, “loitering for the purpose of pan-handling” laws were the state of the art and hadappropriately replaced antiquated vagrancy andoverly broad loitering laws. Recently, court sup-port has drifted from laws against loitering forthe purpose of panhandling to statutes that banaggressive panhandling. These include lawsagainst touching someone while in the act ofbegging, blocking free passage, and beggingnear automatic teller machines, to give just threeexamples. Many libertarians still object to suchrestrictions.5 This shift toward aggressive pan-handling laws, of course, rests on the decision ofsome courts to elevate panhandling to the statusof political commentary about economic injus-tices, and other issues, thereby making the prac-tice a First Amendment free speech issue.

It is unlikely that panhandling can be completelyeliminated from city streets. Because some beg-gars are no threat to anyone, it seems appropriateto limit legislation and police action to aggres-sive panhandling. Yet, in many cities, municipalofficials are so certain that civil liberties groupswill take the city and police to court if theyattempt to enforce local panhandling laws thatthey no longer enforce their laws against loiter-ing for the purpose of panhandling.

35

Those who advocate aggressive panhandlinglaws (or tolerate them in lieu of completelydecriminalizing all begging behavior) and thecourts that try to steer cities toward specificforms of panhandling legislation assume thataggressive panhandling behavior can be clearlydefined in legislation, thus eliminating policediscretion in this matter. This further assumesthat aggressiveness can be defined solely on thebasis of the act itself. This is a tragic misunder-standing of street realities that will result inendless street hassles for police officers anddepartments that understand the social andeconomic costs of unchecked panhandling.

It is wrong to give meaning to behavior solely onthe basis of an act itself because the meaning ofbehavior is to be found in its context. Identicalbehavior can be passive or aggressive, dependingupon its context. At least five contextual ele-ments give meaning to behavior: time, location,number or aggregation of events, condition ofthe victim/observer relative to the perpetrator/ac-tor, and the previous behavior/reputation of theperpetrator/actor. A person standing at the top ofsubway stairs, holding a paper cup, and beggingduring the 8 a.m. rush hour is not much of athreat. At 10 p.m., the same panhandler, behav-ing in the same manner as observed during therush hour, is a genuine threat to a solitary elderlywoman returning home from work. The time andthe location and condition of the woman com-bine to turn the beggar’s behavior into a threatthat anyone in the woman’s position would feeland could result in hassles and inconveniencesfor the woman as well as her reluctance to ridethe subway.

To the contrary, 20 panhandlers in HarvardSquare in Cambridge, Massachusetts, are littlethreat to citizens or, in turn, to commerce.Harvard Square is so vibrant that it can absorbenormous levels of deviance without substantialthreat to citizens or commerce. Twenty pan-handlers in the central downtown area ofWatertown, Massachusetts, however, wouldthreaten many citizens—especially the elderly

and infirm—and could wipe out its commerce.In this situation, the number of panhandlersinteracts with the location to give the behaviorits meaning. (See “Police Must Put Behavior inContext.”)

The second dimension of complexity is officers’responses to situations or problems. In the mid-1980s, David Bayley and Egon Bittner pub-lished two papers that systematically viewedpolice work, its skills, and tactical choices.6

Based on research into police handling of do-mestic disputes and self-initiated traffic stops,Bayley and Bittner examined the teachableskills of patrol work. In the second article,Bayley “presents the results of an attempt to de-scribe as fully as possible the actions of patrolofficers in handling highly problematic situa-tions.”7 Bayley and Bittner identify the tacticalchoices police can make at three stages of en-counters with citizens: contact, processing, andexit. For example, a contact option for an officerduring a traffic stop is to ask the driver whetherhe knew why he was stopped; a processingoption is to warn disputants about what actionspolice might take if their dispute persists; andan exit option is to arrest a disputant.

Each tactical choice by the police, each citizen’sresponse, counterresponses by each, and changesin other variables in the context (for instance,intervention of strangers) create a fluid, ever-changing encounter.

A number of generic problems can arise overthe course of an encounter that may take officersbeyond “common sense” or “command pres-ence” as they weigh the context of the situationand the tactical choices available to them.Bayley and Bittner’s research arose out of adesire to link officer behavior with outcomes.In the realm of policymaking, the goal is to gofrom hypothesized good practice—the wisdomor craft of the field as embodied in goodpolice officers, case studies, and research—toprescribed “ways of thinking” about how toinitiate a contact, process it, and close it that

36

are based on an officer’s assessment of theevent, its situational context, and his or herchoices.

Indeed, officers do this implicitly all the time.The officer who sees a drunk harassing a preg-nant woman with a young child at a busy inter-section may opt to use force (contact choice) bygrabbing the man by the shoulders, walking himaway, and then admonishing him, rather than

Police Must Put Behavior in ContextVirtually any specific behavior gains meaning by the context in which it occurs. Most policeofficers intuitively understand this and make their decisions about whether and how to interveneon their assessment of an act, or series of acts, within a context. Thus, for example, as part of aproblem-solving exercise dealing with squeegee men, I spent time with a New York City neigh-borhood foot patrol officer who was perhaps as assertive in dealing with the problem as anyonein the NYPD. One night, because of some publicity after Mayor Rudolph Giuliani’s election, theofficer’s district commander initiated a crackdown on squeegee men in his district.

I had come to the district to meet with the officer and to interview some squeegee men but dis-covered they were either in jail or had withdrawn from the streets because the word was outabout the crackdown. The officer asked that we visit a particular intersection, where we cameupon an African-American man, in his 50s or so, with a squeegee and a bucket of water. Theofficer called the man by name, and the man walked over to our car. He greeted the officer byname as well. The officer then advised him to “get off the streets, there’s a crackdown.” Theofficer then introduced the man to us; we talked for a short time and then exchanged goodbyes.

I questioned the officer about this exchange because he was known to be enterprising regardingsqueegee men in his neighborhood; here he was warning such a person about a crackdown.“Joe is no problem,” the officer responded. “He really washes windows. He’s courteous; he askspeople if they want their windows washed, backs off if they give any sign that they don’t, anddoesn’t interfere with traffic. He’s a veteran and can’t get by on his pension. I just don’t wanthim to spend a night in jail because the district commander is upset.” For the officer, the contextgave meaning to the event—the location, the reputation of the man, and the man’s behaviorwhen approaching cars shaped how the officer used his discretion. This is the essence of goodorder maintenance and peacekeeping.

Skillful order maintenance activities acknowledge that squeegeeing, like prostitution, pan-handling, and drug dealing, will never be eliminated, but good policing can determine theconditions under which and how such activities can take place and bring them into line withcommunity standards. Was the officer taught how to use discretion in this fashion while he wasin the academy? Did he turn to police guidelines about issues to be considered when decidingwhen and how to intervene in situations? Most likely he learned about discretion on the streetsfrom colleagues and, if lucky, from supervisors as well as from his own experiences.

simple admonishment, because he fears for thesafety of the woman and child. The condition ofthe man (drunk and lurching), the condition ofthe woman (in late pregnancy and with anotherchild), and the time of day and location (on thecurb during a period of heavy traffic) are thecontextual variables that determine the immedi-ate use of force. Moreover, this incident, whichdid happen, could be followed through Bayleyand Bittner’s model with the changing nature

37

of the context interacting throughout the process-ing stage (the impact of comments by thedrunk’s friends) and the final outcome (use offorce and arrest).

Because officers have not been taught to expli-cate their implicit scanning and diagnosticprocess, few of them can describe their handlingof such events beyond saying they relied oncommon sense. With some prodding, officers canquickly move beyond such trivialization of theirwork. Ask the officer, “Would you have behaveddifferently if the drunk was harassing two powerforwards from the Rutgers basketball team?” andhe probably would reply, “I probably would havehad to protect the drunk from the youths ratherthan vice versa.” In other words, it is possible tochange the context of a scenario in ways thatofficers will immediately recognize and, in turn,cause them to alter their responses.

In any policymaking or guideline developmentprocess it is necessary to include, if not beginwith, an explicit statement about the complexityof both the problem and the range of officers’responses. How Bayley and Bittner’s modelshould be included in a total guideline develop-ment process will become clearer in what fol-lows because it is only the first of 11 elements.The second is the acknowledgment of discretion.

Acknowledging Police Use of DiscretionThe preceding discussion provided a contextthrough which officers can think about how theydeal with situations and problems and that will,hopefully, lead to development of training mate-rials, ultimately identify the links between tacti-cal choices and outcomes, and allow officers totalk about their professional practices in a waythat gives due credit to the wisdom, skill, andvalues that are implicit and unacknowledged inpolice work.

In any policy statement or guideline, depart-ments must explicitly authorize discretion andselective enforcement in the handling of situa-tions and problems. Such continued restate-ments are important, despite their redundancy,because citizens, prosecutors, courts, lawyers,and legislatures must clearly understand that theissue is not whether police officers use discre-tion. The real questions are how officers usediscretion and how their use of it is shaped.Linking discretion to scanning, diagnosis, andtactical choices makes it clear that use of discre-tion represents neither arbitrariness nor the per-sonal inclination of officers. (See “PrescribingWhat Officers May Not Do.”)

Prescribing What Officers May Not DoDiscretion is limited. Officers and departments cannot do particular things. Officers may usediscretion in deciding whether and how to intervene in a panhandling incident, but they maynot intervene on the basis of race. Officers may use their discretion to allow youths to stay in apark after curfew, but they may not authorize public or underage drinking (they may know thatsome of it goes on and be less than vigilant about catching it, but they cannot authorize it).Officers may work with neighborhood groups to control access to a neighborhood, but theymust not discriminate against ethnic groups in the process.

In virtually every policy area, there are clear limits that must be prescribed to police officers,citizens, and vested interest groups. Including such limits in policy statements can make valuesexplicit, remind officers that the exercise of judgment and choice is circumscribed, help themto say “no” to inappropriate demands from citizens or interest groups, and provide wide administrativeauthority to discipline unethical and inappropriate use of discretion.

38

Inside police departments, assertions such as“I have a right to use my discretion” can maskrenegade police claims to do virtually anything,or nothing, in the name of discretion. As Bayleyand Bittner point out, tactical choices can belimited regardless of the complexity of the event:

Situations can be ranked along a continuumfrom the cut-and-dried to the problematic. Forexample, American officers have few doubtsabout what to do when a man is found drunklying on the ground in the winter. He must bepicked up and taken to a shelter. The choicesare also fairly limited in serious traffic acci-dents, alleged housebreaking, and assault witha deadly weapon witnessed by a police officer.This is not to argue that some choices are notinvolved in such cases—officers can turn ablind eye or overreact—but rather that the ap-propriate responses are clearly recognized byeveryone involved—patrolman, public, andcommand officers. The appropriate actionmay not be easy to take, but it is obvious.8

When officers, regardless of their motivations,fail to do what they should, “discretion” is noexcuse.

Public officials’ assertions that officers rely onpersonal inclination in their work can mask theofficials’ ignorance or serve as political ammuni-tion in legislative or court processes. In litigationchallenging police order maintenance efforts,defense attorneys often will attempt to swayjudges by demeaning police motives and actionsas expressions of the officers’ personal inclina-tions. Currently, many city attorneys and depart-ments are ill prepared to counter such argumentsbecause they have not intelligently articulatedtheir position regarding discretion. For example,one city recently asked me to help defend itspolicies regarding homeless encampments. Be-cause I knew selective enforcement would bean issue, I asked to see all relevant departmentalgeneral orders, training materials, and rules andregulations that were used to shape how officershandled this problem. No such materials existed.Under such circumstances, it is not the fault of

those challenging the police or the judge in thematter if the department or the city has not doneits homework.

If discretion is properly linked to complexity,as suggested above, it will become increasinglyclear that discretion is, or at least ought to be,shaped by professional police and departmentalknowledge and skill, not personal inclination.

Recognizing and Confirming How PoliceWork Is ConductedThe nature of police work can be explicitly de-tailed in departmental guidelines or policies. It isimportant that everyone involved in developingor using these guidelines either understands oracknowledges policing realities, which includethe following:

● The majority of police work is conducted byan officer working alone or with a partner.

● The officer must make decisions outsidethe purview of supervisors or a commandsystem.

● The officer must make decisions based oninternalized knowledge and skill.

These circumstances are not incidental to Ameri-can policing. In both England and the UnitedStates, police originally were diffused through-out society rather than either barracked or de-ployed in units for tactical reasons (to increasetheir presence) and for control and accountabil-ity purposes. Officers operating alone and em-bedded in the community would be stronglyinfluenced by citizens and forced to seek theirapproval—an essential element in early 19thcentury policing given how distrusted policewere in both countries.

There can be exceptions to officers workingalone—special units, group activities, and soforth—when immediate command or supervi-sion is present, but standard police operationscall for an officer to work alone or with a partnerwithout overseers (in a literal sense). The fullconsequences of diffusing officers throughout

39

communities may have been mitigated by useof automobiles and radios. Nevertheless, mythsthat officers can be supervised or commandeddirectly must be undone because they perpetuatethe current misconceptions that police work canbe organized and administered like a factoryprocess. Such organizational control over offi-cers simply does not exist.

Advancing ValuesThe term “values” has two meanings in thecontext of police discretion: (1) overall values tobe articulated and attained, and (2) underlyingvalues and morals that shape discretion and limittactical choices. The distinction between attain-ing values and attaining goals is based on policeorganizations’ need to maintain vision and per-spective. The values to be attained in the NewYork City subway system included providingsafe, rapid, comfortable, reliable, and fear-freetransportation. A number of departments andagencies were responsible for attaining thesevalues. The police had and shared particularresponsibilities, including graffiti control, pro-tection of passengers and staff, maintenance oforder, and protection of citizen rights. In per-forming these responsibilities, goals were set:eliminate graffiti by 1989, limit and regulatepanhandling, reduce robberies, protect school-children during rush hours, and so forth.

When departments articulate the ultimate valuesto be attained, officers can focus on loftier goalsthan the trivial, messy, and distasteful tasks thatmake up much of daily police work. Where suchvalues are not articulated, it is difficult for offi-cers to maintain the idealism and vision thatmany bring with them as young recruits. Per-spective on the proper police role in society alsosuffers. Values in this sense inspire officers, es-pecially when they are wrestling with nitty-grittyproblems, considering tactical choices, and plan-ning overall strategy. In this light, dealing withpanhandling in New York’s subway system is notseen as “throwing the scumbags out” but, rather,as protecting citizens, reducing fear, puttingpolice in contact with hardened criminals, and

preventing crime—all of which contribute toquality transportation, especially for the workingclasses who cannot afford other options.

The external benefit of getting and keeping po-lice officers focused on the higher values inher-ent in their work is that doing so communicatesto the public a true vision of the purpose of po-licing in a democratic society.9 In controversialpolicy areas, putting values forward is not onlyimportant in its own right but also does not con-cede the moral high ground to policy opponents.In developing policing policies for New York’ssubway system, we believed that advocatesfor the homeless, with their insistence that thehomeless should be allowed to camp in tun-nels—as dangerous and unhealthy as they are—were cynically exploiting the issue to make apolitical point. Protection of the constitutionalrights of all citizens, including the needy home-less who lived in the subway system, was a keyvalue to be articulated. While the homeless en-joyed the same rights as everyone else, they alsohad to abide by the rules.10 (It was common prac-tice for transit police to waive the fare for home-less persons who would ask the officers if theycould use the subway to get to a shelter. Theofficers knew that this was often a scam, but,nonetheless, the rules of the subway were hon-ored in the request and civility was maintained.Moreover, many persons were desperate andneeded free transportation to get to help, andit was appropriate for officers to assist them.Under certain conditions police would evencall a van to get groups of persons to help or toshelters.)

Values in the latter sense refer to the mainte-nance of moral and constitutional values inpolice work. Virtually every police departmentnow has some statement of departmental values,which, while undeniably proper, arguably areof little relevance to day-to-day policing. Whilethe high value put on human life shapes policiesregarding use of deadly force in every police de-partment, deadly force is “sexy” stuff—shootingpeople—and has little to do with routine policework.

40

Guidelines and policy statements can be at thecore of linking values to police work. Moral andconstitutional issues are rife in police work, andit is almost inevitable that one will confrontthese basic issues in any attempt to solve prob-lems or develop policies. A good example oflinking values to policy in routine matters isfound in the San Diego Police policy statementregarding undocumented persons that was dis-cussed earlier. In the context of a discussionabout what role the local police departmentshould play in the enforcement of immigrationlaws, the statement “the San Diego PoliceDepartment recognizes and values the diversityof the community it serves” gains fresh andspecific meaning.11

Likewise, the New York City transit police hadto consider a mayoral edict (i.e., that no onecould be ejected from public facilities if thetemperature dropped to 32 degrees Fahrenheit)in developing policies regarding when rulebreakers would be ejected from the system.The value that underpinned this decree was, ofcourse, the value of human life—that someoneejected from a public facility could freeze todeath. The question was whether the Metropoli-tan Transportation Authority and the transit po-lice were going to honor this proclamation (as aState agency, the transportation authority is notunder the mayor’s direct administrative control,although the mayor has enormous influence overits policies and practices). The initial response ofalmost everyone in the study group was reflex-ive: of course, the rule would be honored. Toconsider the widest possible range of options inthe effort to restore order, however, members ofthe study group ultimately agreed that ejectionof disorderly persons had to be considered inpractically every circumstance, including badweather. Given the volatility of the issue, thestudy group understood that Chairman RobertKiley of the Metropolitan Transportation Au-thority would make the final decision. We wereobliged to wrestle with the issue and makerecommendations to him.

The discussions about whether people should beejected from the subway system during freezingweather were fraught with emotion. One mem-ber vehemently maintained there was no waythat he was going to order someone ejected todie in the cold. The issue was complicated bythe fact that the majority of the homelesspersons who violated the rules were seriousalcohol and/or drug abusers, mentally ill, orboth. Clearly, this was a vulnerable population inneed of special consideration. At the same time,people were dying of hypothermia in the sub-way. Staying in the subway, as some advocatesurged the homeless to do, exposed them to life-threatening risk. The group’s final recommenda-tions to Kiley were that rule violators were to beejected regardless of the weather; that at tem-peratures below 40 degrees, bus transportationshould be made available for those who wishedto go to a shelter; and that in cold weather policeshould be especially scrupulous about takinginto custody any persons who appeared to beincapable of helping themselves or who wereimproperly clothed and refused bus transporta-tion to a shelter.

Value issues need not be so grand as life anddeath. What about youths in parks on hot sum-mer nights in densely crowded neighborhoods?Should they be allowed to stay beyond a 10 p.m.curfew if they are quiet and don’t drink? Whatpotential values would be served? What valueswould be eroded? What about street vending?What rights do vendors have or should theyhave? Do the current means of obtaining licensesinadvertently discriminate against particulargroups or persons? Are there ways to control ormanage vending that do not deprive poor personsof the opportunity to earn money in otherwisehonest ways? How are the rights of vendors bal-anced against the rights of shopkeepers who arepaying high rent and other forms of overhead?

How should civil initiatives be used? What self-imposed limits should police develop in the useof civil initiatives as a tactical choice in solvingproblems? Does their use at times inappropri-ately circumvent criminal due process?

41

These are issues that police themselves shouldaddress because, if they do not, it is only a mat-ter of time until others, most likely the courts,will do so. Even if the courts or other authoritiesdo address these issues, it would be better if thepolice position were presented proactively afterit has been carefully carved out in the real worldand adapted in such a way that routine policework makes concrete the abstract values thatneed to be articulated when the relevance of thevalues is not always apparent.

Putting Police Knowledge ForwardOver the past several decades, research and casestudies have enriched the factual and empiricalbase from which police operate. Much of thisknowledge is available not only in traditionalforms—academic publication in books andjournals and in research publications of theNational Institute of Justice—but also, morerecently, over the Internet. The Police ExecutiveResearch Forum, for example, currently pub-lishes case studies of problem solving on theInternet. Such research and case studies arevaluable for at least three reasons. First, they putforward a model of systematized thinking aboutpolice business. Second, they provide factualevidence about the impact of particular policeactivities. Finally, they provide the intellectualjustification for police practices, especially whenthey are challenged legally.

In New York’s subway system, for example, twobodies of research were of special importancein developing policies. The research abouthomelessness was valuable because it helpeddistinguish the subway’s problems, which wereprimarily problems of disorder, from the generalproblems of homelessness. Moreover, researchabout the links between disorder and fear wasrelevant to police managers and officers, whoneeded to be convinced that order maintenancewas important police business, and also to themedia and general public.

The transit police studying the problem, how-ever, did their own research as well. Theycounted the homeless, checked records to deter-

mine the criminal histories of rule violators,surveyed middle level managers about theirperceptions of the problems, talked with personsviolating subway rules, photographed locationsto determine the extent of rule violations,conducted focus groups with passengers, andsurveyed riders through the MetropolitanTransportation Authority. Consequently, theliterature reviews and independent researchwere presented not only in policy statements butalso in training materials, educational materialsintended for the general public, and media re-leases. It was also essential in the litigation thataccompanied policy implementation.

Undergoing Development by PracticingPolice Officers and CitizensI have been involved with practicing policeofficers in problem-solving exercises from theearly 1970s in Kansas City (to determine howincreased numbers of police should be used inthe South Patrol District—a problem that gaverise to the Kansas City Patrol Experiment)through the 1990s in New York City around theproblem of squeegee men. Some of the exerciseshave been quite formal, others less so. Thesincerity with which officers approach suchexercises has been impressive, but, to be sure,they view such exercises with great skepticism.Task forces, they have said, are used to undercutunions, or officers’ comments and observationswill be ignored. Often, officers harbor unspokenresentment that a civilian—a social worker fromHarvard at that—should be involved in helpingpolice think about what their work is and howthey should do it. When such posturing is fin-ished, however, officers have worked hard to un-derstand problems, consider solutions, and carrymessages out to their colleagues—an invaluableasset in attempts to gain officer support ofnew efforts. Some task force members remaincranky—in some departments I have deliber-ately requested that the most obdurate officersbe placed on a serious task force—but I haveobserved those same officers often either jointhe study process or allow their “crankiness” tobe managed by the other officers.

42

Regarding “big ticket” issues, such as restoringorder in the subway, it is also worthwhile tobroaden input through targeted departmental sur-veys. For example, we anticipated that we wouldhave to develop wide support for order mainte-nance activities among transit district captains.To do so, we had to understand their concernsand seek their input in developing policies.Meeting with all district commanders provedto be unworkable. Consequently, a draft policywas distributed to district captains along witha questionnaire to ascertain whether the policyaccurately reflected the problem, recommendedappropriate responses, and offered realisticimplementation plans. The data gathered werethen used to shape the evolving plan.

Persons who do the work and understand thedaily issues provide rich experiences, data, andinsight during problem-solving or guideline-developing exercises. Many officers need helpin articulating their points of view. Some ideasmust be validated. More than once, I have heardan interesting point of view put forward by anofficer who was pounced on by an administratorwho said “we can’t do that.” Yet, it later becameapparent that the idea could be implementedand, moreover, provided the key to unlockingthe solution to the problem.

The role of citizens in guidelines development isless clear. Generally, it is a good idea to involvecitizens. What form citizen involvement shouldtake, and whether all policy issues require orshould have citizen input, also is less clear; po-lice departments are gaining experience in howto approach this issue. In the New York subways,citizen input came from civilian MetropolitanTransportation Authority employees and focusgroups. As a general principle, citizens shouldbe involved in policymaking, and the burden ofproof always should be on those who wish toexclude citizens from a role in policymaking,not on those who wish to include them.

Clearly, depending upon the issue, citizensshould consent to and support police policies.The notion that police can operate in a demo-

cratic society authorized solely by the law wentout with the riots of the 1960s, although policemust relearn this lesson in some communities.Additionally, we want citizens to cooperate withpolice when called upon to do so.

Finally, in many issues police need the activepartnership and collaboration of citizens tosolve problems. In such cases, citizen involve-ment in policy development seems to be a pre-requisite for collaboration. A myriad of issues,of course, surface when citizens become in-volved in policymaking. Which citizens? Shouldthey be representative? Of whom? To whatextent should opponents of policies—say home-less advocates who are opposed to antiaggres-sive panhandling ordinances—be involved inpolicy development? How do police reconcilecompeting interest groups in communities?Such questions aside, citizen involvement caneducate individuals about problems, involvethem in decisionmaking, define the citizen rolein problem solving, gather citizen support forpolice action, and help citizens understandpolice limitations.

Undergoing Clear and Broad PublicPromulgationPolice policies should be clearly written andtargeted at practicing police officers and theaffected citizens, which should include politicalleaders, community stakeholders, and themedia. Guidelines should be communicated tothese target groups in ways that encourage theirdiscussion and review.

Many people outside policing would be shockedby the current business-as-usual approach togeneral orders and rules and regulations, oftenabout highly controversial and politicallycharged issues, that many police departmentspursue. Frequently drafted by and for lawyers(staff lawyers of police departments and cityattorneys), they are filled with legal and bureau-cratic jargon. Practicing police officers are notinvolved in their development and do not reviewthem before they are promulgated. Important

43

general orders are distributed many times in payenvelopes (and promptly thrown out as junkmail). Their reading at role call is often perfunc-tory, often accompanied by the reader’s sneers oreye rolling, and delivered in a fashion that dis-courages discussion or questioning. It is no won-der that in department after department I haveobserved officers treat such policies with disdain:they are written to protect the city and manage-ment from liability; they have “someone’s nameon them”—they are written in response to a par-ticular officer’s mistake; they are used to “get”officers when they make honest, but highly pub-lic, mistakes; and so forth.

While our experience in dealing with disorderin New York City’s subways was a “big ticket”problem—one that required a shift in the basicoperation of the department—how the policy waspromulgated was instructive. All documents wereeither developed by or reviewed by police offi-cers in the study group. As drafts emerged, studygroup members were encouraged to share themwith other officers and to solicit feedback abouttheir relevance and clarity. Once policies hadbeen developed, in addition to training and videomessages played at roll call and made availablein all districts, a leaflet was developed and dis-tributed to all officers. The leaflet explained thenew policies and answered critical questionsabout what was being done, why it was beingdone, what moral and constitutional issues wereraised, and what impact the new policies wereexpected to have. Officers were encouraged tocarry copies of the leaflet and to distribute themto citizens who might question police policies.

Externally, promulgation was widespread. Signswere posted delineating subway rules; op-edpieces were published; political leaders, advo-cacy groups, and the media were consulted withand briefed; and other agencies that would be af-fected by subway policies and had been includedin planning were also briefed. In addition, leaf-lets made to look like desk appearance tickets(citations) were printed for distribution to ruleviolators during the 2-week period prior to

implementation of the new policies. In essence,they informed rule violators that they would besubject to ejection, citation, or arrest if they per-sisted in violating the new rules after a particulardate.

To be sure, this was a “big” problem and enor-mous resources were spent in developing andimplementing policies. Nonetheless, the prin-ciples hold true even for routine matters: theynot only involve citizens and police in develop-ing policies but also put those policies forwardin ways that are clear and forthright.

Emphasizing Police Adherence toa ProcessOur understanding of the links between policeactions and citizen reactions and outcomes is soextraordinarily primitive that we cannot realisti-cally hold officers accountable for outcomes.An officer handling a person on the threshold ofcommitting suicide might apply all the craft ofthe police trade and the result could still be asuicide. Likewise, a unit handling a hostagesituation might negotiate with considerable skill,yet a hostage might be killed. Even in minorsituations, an officer might gently reprimand anaggressive panhandler and still wind up with anugly incident.

The goal is to train officers to assess situationsaccurately and choose responses that are consis-tent with the state of the art of their occupation.Moreover, police should articulate both thecomplexity of situations and the reasons for theirtactical choices—that is, to end their reliance onjustifying their actions as “common sense” or“appropriate action.” In effect, good policy state-ments prescribe not what an officer should do—this is impossible in the real world; rather, theyprescribe how an officer should think about aproblem and choose among alternative solutions.Good policy statements should say: “These arethe factors you must consider when you ap-proach a situation. These are the values thatshould guide you. These are the choices youhave at your disposal.”

44

Good policy statements should provide the lan-guage that officers use to describe their work forboth development of ongoing police knowledgeand supervisory purposes. So little police workis conducted under the eye of supervisors thatthe only way to oversee most routine policework is for officers to talk about their work withtheir superiors. Talking about police work allowssupervisors to mentor officers in problem solv-ing (e.g., “What alternatives did you have at thismoment?”) and, over time and with the assis-tance of case studies and other research, beginthe process of linking police thinking and tacticsto outcomes. (A note regarding mentoring offi-cers: this approach to supervision does notassume that sergeants “know best” about howofficers should have handled situations. Ser-geants should help officers review their work toensure they considered all the proper factors intheir decisionmaking process.)

Establishing AccountabilityGood policy guidelines allow supervisors andadministrators to distinguish between officerswho operate within the rules and yet behave inan uncaring and incompetent fashion, and offi-cers who bend or break the rules and yet behavein a caring manner. In one example, an officer ina midwestern city was dispatched to respond toa call from a two-story flat. A man who livedalone upstairs had called the police at about11 p.m. on a winter night to report that twoyoung girls, ages 7 and 9, who lived downstairshad been left alone for several days. The re-sponding officer investigated and found thereport to be true. The two girls had been alonefor two nights and did not know when theirmother would return.

The police department had established a clearpolicy for dealing with such incidents. Theofficer was to contact an oncall social workerand turn case responsibility over to the socialwelfare department. The officer called. Thesocial worker, after listening to the officer’s ac-count, suggested that because the upstairs neigh-bor obviously was concerned about the girls’

welfare, as evidenced by his call, the policeofficer should ask him to take the girls into hishome for the night. The social worker wouldcome first thing in the morning to pick up thegirls. The officer refused to do this. To settle theresulting impasse, the officer bundled the twogirls into his squad car, stopped at a McDonald’sfor food, and took the girls to the police stationwhere they spent the night. By doing this, theofficer broke departmental policy by not trans-ferring the case to the social welfare departmentand departmental rules by transporting the girlsin his squad car.

In reviewing the case with the officer, it wasclear that he put a set of values about children,strangers, and the protection of children abovedepartmental policies and rules. (When I firstasked the officer why he acted as he did, he said,simply, “Well, it was just common sense.”)Happily, to the officer’s knowledge, everythingturned out satisfactorily. Nobody reviewed thematter with him, and the children were pickedup by social welfare workers the next morning.

In my conversation with the officer, I played outtwo alternative scenarios. In one, the officer’scar was hit by a drunken driver while he wastransporting the children, injuring them. I askedthe officer what authorized him to use his discre-tion as he had; in other words, how would he de-fend himself if charged with violating policy andbreaking rules? The officer knew of no basis inguidelines, policies, or traditions in the depart-ment that he could use to justify his actions.In fact, he firmly believed that the departmentwould discipline him if the scenario I hadspelled out had occurred.

In the second scenario, an officer comes to thescene, observes the same conditions, but followsthe welfare worker’s suggestion. The girls aremolested by the man. The officer believed thatwhile he would have been disciplined for violat-ing departmental guidelines and rules if he trans-ported the children and his cruiser was hit by adrunk driver, the fictitious officer who left thegirls with the upstairs neighbor would not have

45

been. Why? The officer who placed the childrenwith the man had observed the rules and despitethe tragedy could not be held responsible. He,the real officer, had broken rules and would beheld accountable if the children had been in-jured. (It is by no means certain that the officerwould have been disciplined by the department.)

Nevertheless, the importance of putting forwardvalues in policy guidelines becomes apparent.The officer had strong personal and professionalvalues about protecting young children fromvictimization that he held above specific policiesand rules. One simply does not put young chil-dren into the hands of strangers except in themost dire circumstances, no matter what policydictates. The officer who cared enough to thinkthrough the situation and use his discretion wasclearly in need of prior authorization to do thisas well as support in case the children wereaccidentally injured. The officer in the scenario,however, who obeyed the policies and rules, didso unthinkingly and uncaringly and put the chil-dren at serious risk. The message is, of course,that following the rules ought to be no protectionfrom charges of incompetence and disciplinewhen clear guidelines mandate the values thatshould undergird police actions.

In another example, Madison, Wisconsin, PoliceOfficer (now Sergeant) Joe Balles recounts howhe used his discretion and violated a court orderthat barred a father from access to his child. Thefather, known to Officer Balles, was involved ina custody dispute with the mother of the child.The father called Balles directly late one night toreport that the mother was seriously drunk andunable to tend to the child. Balles tried but wasunable to find suitable care for the infant for theremainder of the night. Based on his familiaritywith the father, the mother’s condition, and thecomplete absence of resources to care for theinfant, Balles left the child with the father for therest of the night and made arrangements to pickup the baby the next morning. Balles knowinglyacted against the court order. As in the first case,

contextual exigencies made the officer’s re-sponses entirely reasonable. Happily, the fatherturned the child over to authorities in the morning.

Policy guidelines can ferret out officers whofollow the rules but underperform and avoidresponsibility as well as identify officers whobreak rules to protect important values or achievedepartmental goals. As noted above about values,guidelines can put forward positive expectationsabout officer performance and set achievementstandards.

Receiving Recognition as anOngoing ProcessPolicy development is ongoing. It is a repetitive,never-ending aspect of police work. Just as thestudents in Dayton changed every year, requiringthe development of a new contract between po-lice, students, and neighborhood residents, so thecomposition of every neighborhood constantlychanges. Changing conditions, laws, traditions,and standards require continual updating ofpolice guidelines.

Notes1. Kelling, George L., and Catherine M. Coles,Fixing Broken Windows: Restoring Order andReducing Crime in Our Communities, New York:Free Press, 1996, especially Chapter 4, “TakingBack the Subway: New York City’s Quality ofLife Program”: 108–156.

2. Ibid.

3. This list is a variation of one first presented inFixing Broken Windows. The variations from thefirst list testify to the evolving nature of mythinking.

4. Because complexity is the most complicatedelement in policymaking, this section will beconsiderably longer than any that follow.

5. This issue is discussed in considerable detail,including legal details, throughout Kelling andColes, Fixing Broken Windows.

46

6. Bayley, David H., and Egon Bittner, “Learn-ing the Skills of Policing,” Law and Contempo-rary Problems 47 (4) (Autumn 1984): 35–59;and Bayley, David H., “The Tactical Choices ofPatrol Officers,” Journal of Criminal Justice 14(1986): 329–348.

7. Bayley, David H., “The Tactical Choices ofPatrol Officers,” Journal of Criminal Justice 14(1986): 329.

8. Bayley, David H., and Egon Bittner, “Learn-ing the Skills of Policing,” Law and Contempo-rary Problems 47 (4) (1984): 36–37.

9. For a discussion of such values, see Moore,Mark H., Creating Public Value: StrategicManagement in Government, Cambridge, MA:Harvard University Press, 1995.

10. A friend described an incident he had seenrecently. He had observed a person approach apolice officer in a subway car to complain abouta disheveled man who literally reeked. The per-son had asked the officer to eject the disheveledman. The officer patiently explained that, al-though he understood the man smelled bad, thiswas not a basis for ejecting him from the sub-way. The disheveled man was obeying the rulesand, therefore, had a right to use the subway.The officer quietly did something very impor-tant: he not only defended the man’s right to bethere but also encouraged the complaining per-son to be more tolerant.

11. San Diego Police Department, Field Opera-tions, Department Procedure 6.18, December 20,1987.

47

Conclusion

One of the lessons learned over the past threedecades, a lesson incorporated into communitypolicing, is that in a democratic society effectivepolicing can be achieved only with communitysupport and involvement. Sir Robert Peel under-stood this when he sent bobbies into the toughestareas of 19th century London with a mandate topersuade people to behave. Police demeanor wassuch that officers gained the sympathy and sup-port of the population. We still have much tolearn from this, especially since policing teetersnear the edge of militarism in so many locationsas part of the “war” on drugs and drug dealers.Viewing the police, as the late Frank Remington

and others have done, as an administrativeagency obliged to develop guidelines publiclythat will shape its inevitable use of discretionoffers one more way to develop community sup-port and involvement in policing urban America.This viewpoint not only will improve the qualityof policing but will also improve public under-standing and support of police. But guidelinesdevelopment must not be seen as a one-shotdeal—write the manual and send it out. It mustbe understood to be an integral, ongoing part ofpolicing. At its essence, developing guidelines isthe process of creating a community consensusabout our moral and legal basis for urban life.

49

Appendix A

Order Maintenance Training Bulletin 96–1

NEW HAVEN POLICE DIVISION OF TRAINING AND EDUCATION, NEW HAVEN,CONNECTICUT

I. PURPOSE. The purpose of this training bulletin is to define the order maintenance policies andprocedures of the New Haven Police Department (Department of Police Service), New Haven,Connecticut.

II. POLICY. It is the policy of the New Haven Police Department to work with neighborhoodresidents and others who use public spaces to maintain order legally, humanely, respectfully andequitably.

Maintaining order is an honorable and historical function of police departments. Although somecritics of police order maintenance activities have labeled such police activities as harassment, main-taining order on streets, in parks, and in other public spaces is no more harassment than is trafficenforcement. Whether citizens are using the street for transportation, recreation or commerce, it isexpected that they will do so with respect for others’ safety, and within the limits established by thelaws of the State of Connecticut and the ordinances of the City of New Haven.

The purpose of maintaining order is: (a) to prevent crime and reduce citizen fear; (b) to facilitatepublic discourse and activities; (c) to create an atmosphere tolerant of diversity; and (d) to improveand restore the quality of life in neighborhoods. The problems being referred to as order maintenanceproblems include, but are not limited to, abandoned cars, prostitution, noise, graffiti, public drinkingand disorderly conduct such as aggressive panhandling.

The New Haven Police Department recognizes that its order maintenance activities are discretionaryat all levels of the department, from chief through all personnel. Discretion, however, does not implypersonal inclination. Discretion is the application of the professional knowledge, values, and skill ofpolice departments and officers to particular problems and incidents. The starting point of all profes-sionalism for police, however, is the law.

Our order maintenance activities will always operate within the law. Having said this, however, itmust also be asserted that while the legality of our activities is basic, it must be augmented by publicsupport. Moreover, this support must be constantly renewed. This renewal is an ongoing process andintegral to our relationship with citizens through Management Teams, community organizations,neighborhood and commercial associations, schools, churches, synagogues and other organizations.

Because order maintenance is, and has been, a central function of police, it is expected that police willroutinely maintain orderly conditions throughout the city and, more specifically, in the geographicalareas for which they are responsible. While specific neighborhoods may develop priorities (for ex-ample, put heavy emphasis on the behavior of prostitutes), police will address all order maintenance

50

problems in all neighborhoods. This is no different than dealing with serious crimes: the fact thatburglary may be a neighborhood priority does not mean that officers will ignore armed robbery orassault.

III. THE LAW (Sources of Authority). Generally, the legal basis for order maintenance is found inbreach of peace, public disturbance, disorderly conduct, obstructing, harassment, intoxication byalcohol and/or drugs, criminal mischief, public indecency and loitering in or about school grounds.

IV. PHILOSOPHY of ORDER MAINTENANCE PRACTICES. The New Haven Police Depart-ment will always use the least forceful means possible to achieve its purposes. While we will nothesitate to cite or arrest offenders, our approach, at all levels of the organization, will be to attempt toget citizens to obey laws and ordinances as unintrusively as possible.

The first level of intervention, whether by managers, supervisors, or by police officers, will be toeducate the public about civility, the consequences of incivility, and the laws that oblige citizens tobehave in particular ways. This can be done in neighborhood meetings, in schools, or in interactionswith citizens. Some citizens do not fully understand their obligations, and if those obligations—forexample, regarding a noisy car or public drinking in parks—are patiently explained, they will adhereto the law.

The second level of intervention will be to remind citizens of their responsibilities if they are disor-derly—that is, that they are breaking the law and subject to penalties if they persist. This too can bedone in a variety of ways. It could be done by visiting a problem location and warning people that iftheir behavior continues they will be subject to penalties. Similarly, owners of locations that arechronic problems could be so warned by individual officers.

The final level of intervention will be law enforcement—the use of citation and arrest.

Having said that the least intrusive means of intervention will always be used should not be read tomean that in every incident police must start with education. Since police deal with incidents that havehistories (for example, with problems), it may well be that in a particular incident the offenders mighthave a history of outrageous behavior which warrants forceful action at the outset of the encounter(for example, warning or citation).

V. CRITERIA for the EXERCISE of POLICE DISCRETION. Managers, supervisors, and offi-cers, as well as Management Teams, will use at least the following five factors to determine the levelof intervention to be used. While articulating them here might seem extraordinarily formal, policeofficers traditionally have used these criteria daily as they make decisions on the street. We are simplymaking explicit here what has been implicit in the past.

a. Time: Disorder has important chronological aspects. We acknowledge this culturally throughthe creation of holidays (the Fourth of July, the Freddy Fixer Parade, Gay Pride Day, St.Patrick’s Day, for example) and other periods when we are more tolerant of behavior andentertainment (for example, Friday and Saturday nights, New Year’s Eve.)

b. Location: Different neighborhoods have different thresholds for various kinds of activities.Certainly, one can be more tolerant of noise levels in downtown New Haven than in residentialareas. Some forms of disorderly behavior are absolutely inappropriate around schools (public

51

urination by adults, for example) and would be the basis not for education or warning, but forstrong legal condemnation.

c. Condition of the Offender: Here we would be concerned about whether a person is intoxicatedor under the influence of drugs, or behaving in a variety of inappropriate or disturbed ways. Ill-ness, and behavior associated with illness, would be another variable affecting police response(seizures and post-seizure responses, for example). Obviously, we are not talking about mattersof social class, race, homelessness, etc., when we refer to the condition of the offender. Thefocus is behavior. Likewise, we would be less concerned about a person who urinated publiclyif the person attempted to find a solitary location and maintain a sense of modesty than some-one who flagrantly exposed him or herself in a highly visible location.

d. Condition of the Victim/Witness: Clearly, we would be more concerned about aggressive pan-handling, for example, which targeted vulnerable persons—children, the elderly, people withdisabilities—than we would about similar approaches to sturdy youths. Similarly, as a matter ofpolicy, we would always be more concerned about the impact of forms of disorderly behavioron children.

e. Numbers, Volume, or Aggregation: One panhandler is one type of problem, ten panhandlers isanother. Similarly, virtually every form of disorder has quite different meanings dependent onthe number and concentration of people committing the act(s).

These factors, and others, will be primary in the determination of police response to disorder whetheron a departmental, substation, or individual officer level. New Haven police officers are expected touse their discretion wisely and proudly.

VI. PROCEDURES and PROBLEM-SOLVING. The vast majority of order maintenance activitieswill be conducted informally by officers who encounter disorderly behavior on the street. Most oftencitizens will be educated or warned. Occasionally, especially when education or warning is ignored orwhen behavior is unusually outrageous, arrests will be made. The basis of all such law enforcementactivities will be probable cause.

Some activities will be more formal, however, especially when neighborhoods are confronted withintransigent problems that require the coordinated efforts of neighborhood police, the community,citizens and, at times, special units. When more formal, coordinated, order maintenance projects areconducted, it is expected that Management Teams, as well as supervisory personnel, will use aproblem-solving approach. This approach will include:

a. Identification of neighborhood priorities and the ranking of problems within those priorities;

b. Clear explication of the nature of the problem (problems are not always what they seem);

c. Consideration of tactical options—including the roles of citizens, other agencies and institutionsand police (this may vary widely depending upon resources of neighborhoods);

d. The legal and moral implications of each tactical option;

e. The expected results of each option;

f. The selection of an option;

g. The identification of early indicators of success and/or failure;

h. A preliminary implementation of the selected option;

52

i. Monitoring for early indicators of success and/or failure;

j. Full implementation or return to “b” (above) if early indicators are negative; and

k. Termination of effort (goals obtained).

When formal problem-solving is conducted, brief written records will be kept by the officer, DistrictManager or Management Team.

See Connecticut Law Enforcement Officers’ Field Manual (Red Book) for specifics.Effective date 1 January 1996.

53

Appendix B

San Diego Police Department Policy Statement

ALL PERSONNEL UNDOCUMENTED PERSONS

FIELD OPERATIONS D.P.6.18 12/20/87 3.8

I. BACKGROUND

This Department Procedure is intended to clarify the Police Department’s policy with respect to thehandling of undocumented persons.

The San Diego Police Department recognizes and values the diversity of the community it serves. Thepurpose of this policy is to ensure the safety and well being of all persons, regardless of their immi-gration status. The primary responsibility for the enforcement of Federal immigration laws rests withthe Immigration and Naturalization Service and the U.S. Border Patrol. Although State and localpeace officers have the authority to assist in enforcing immigration laws, it is the policy of the SanDiego Police Department that officers shall not make an effort to look for violations of immigrationlaws.

San Diego Police personnel will focus on detecting and apprehending individuals involved in criminalactivity.

II. ADULT UNDOCUMENTED PERSONS

A. San Diego Police officers are responsible for the enforcement of all laws, Federal, State andlocal, and the safety and protection of all persons. Therefore, officers have a duty to contact anyperson(s) when there is a reasonable suspicion to believe they are involved in criminal activity.(Refer to SDPD Procedure 4.1.)

B. If upon investigation “probable cause” to arrest exists, unrelated to the person’s immigrationstatus, officers may arrest for the offense.

1. If the subject is booked in the City or County jail and determined to be illegally in theUnited States, a hold for INS/Border Patrol shall be placed on the prisoner.

a. Immigration documents identified as evidence in a criminal investigation will beimpounded per Department Procedure 3.2.

b. See Health and Safety Code section 11369 for notification requirements when the arrestis drug related and the person is suspected to be undocumented.

C. Officers are authorized to release subject(s) to INS/Border Patrol if there is no “ProbableCause” to arrest but there is “Reasonable Suspicion” that criminal activity unrelated to immi-gration status still exists. After investigation determines the detainee(s) is/are in violation of8 United States Code Section 1304(e) officers are authorized to notify INS/Border Patrol and

54

release the detainee(s) at the scene of contact, within a reasonable time. (“Reasonable suspicion”defined in Department Procedure 4.1.)

1. The duration of the stop or detention prior to the detainee’s release to INS/Border Patrolshall be in accordance with Department Procedure 4.1, paragraph 4a, which provides asfollows:

A person stopped pursuant to this procedure may be detained at or near the scene of the stopfor a reasonable time. Officers should detain a person only for the length of time necessaryto obtain or verify the person’s presence or conduct, or an account of the offense, or other-wise determine if the person should be arrested or released.

Such factors as remoteness and safety considerations for the person(s) detained may extendthe reasonable time of detention prior to INS/Border Patrol release. However, such deten-tions should ordinarily be no longer than twenty (20) minutes.

2. Officers are generally prohibited from transporting detained undocumented persons to apolice facility for the sole purpose of releasing them to INS/Border Patrol.

3. Undocumented persons may be transported if they voluntarily consent in order to completeor further an investigation.

D. After investigation if it is determined the person(s) is/are not involved in criminal activity unre-lated to immigration status, the person should be released, regardless of immigration status.

III. EXCEPTIONS

A. Certain criminal situations, because of their inherent danger to citizens of the United States andundocumented persons as well, require immediate action by San Diego Officers. Officers areauthorized to detain and release undocumented persons to INS/Border Patrol when contactedunder the following conditions:

1. Drop House—a house or building being utilized as a transfer/holding facility for personsengaged in smuggling undocumented persons.

2. Load Vehicles—vehicles engaged in smuggling undocumented persons.

3. Drug House—house or building being used to facilitate narcotics trafficking.

IV. SITUATIONS WHERE BORDER PATROL INVOLVEMENT IS PROHIBITED

A. Officers are prohibited from releasing undocumented persons to INS/Border Patrol under thefollowing conditions:

1. They are victims or witnesses of a crime, unless a determination has been made byinvestigators to hold them as material witnesses.

2. When contacted during family disturbances.

55

3. Generally, during the enforcement of minor traffic offenses (interactions and nonbookablemisdemeanors).

4. When the person(s) are seeking medical treatment.

V. MIGRANT CAMPS

A. The majority of residents living in migrant camps work in the surrounding area and are in thiscountry legally. Officers are generally prohibited from detaining undocumented persons in thissetting for INS/Border Patrol unless there is probable cause to arrest for a crime not related toimmigration violations. If probable cause to arrest exists, officers will follow the procedureslaid out in this policy.

VI. DOCUMENTATION OF DETENTIONS

A. All undocumented persons who are arrested will be booked or cited and appropriate reportsprepared. When undocumented persons are detained and released to INS/Border Patrol, officerswill prepare a detention report in every case.

1. Detention reports involving undocumented persons must list in detail the reasons for thedetention, including all facts that led to the “reasonable suspicion” that the subject wasinvolved in criminal activity unrelated to immigration status.

2. In the event officers locate a drop house, load car or drug house containing multiple undocu-mented persons, they will document the event by preparing an ARJIS–8 on each subject andan ARJIS–9 detailing the circumstances of detention.

3. The name(s) and identification number(s) of the agent(s) who take custody of the detaineeswill be included in the appropriate (ARJIS–8/ARJIS–9) report(s).

VII. ASSISTANCE TO LAW ENFORCEMENT AGENCIES

A. Officers are directed to provide necessary assistance to all law enforcement agencies includingthe United States Border Patrol when requested to do so.

B. An emergency statute, California Government Code Section 55069.75, taking affect on October4, 1993, was enacted to guarantee “continued federal support for local enforcement activities.”It provides as follows:

“. . . no local law shall prohibit a peace officer or custodial officer from identifying and report-ing to the United States Immigration and Naturalization Service any person, pursuant to federallaw or regulation, to whom both of the following apply:

(a) The person was arrested and booked, based upon the arresting officer’s probable cause tobelieve that the person arrested had committed a felony.

56

(b) After the arrest and booking in subdivision (a), the officer reasonably suspects that theperson arrested has violated the civil provisions of the Federal immigration laws.”

VIII. UNDOCUMENTED JUVENILES

A. 300 W&I Dependent Children

1. Under 13 years of age

Children in this category will be transported to Hillcrest Receiving Home if a parent orguardian cannot be contacted. Hillcrest will determine the status and disposition of thechild.

2. Thirteen years of age or older

If the juvenile’s parent or guardian can be located, the juvenile will be released to themregardless of the family’s immigration status.

3. If the juvenile’s parent or guardian cannot be contacted, the juvenile will be released toBorder Patrol. Transportation to an INS/U.S. Border Patrol facility is authorized for thispurpose.

4. An ARJIS–9 report will be submitted detailing the circumstances of the detention and thedisposition of the juvenile.

B. 601 W&I Status Offenses (i.e., curfew, truants and runaways)

1. Under 13 years of age

If the parents reside in a foreign country, the juvenile will be transported to HillcrestReceiving Home.

2. Thirteen years of age or older

It is incumbent upon the Police Department to return juveniles without parental supervisionto their parents, guardians or school officials. If the parents or guardians are in the UnitedStates and can be contacted, the juvenile will be released to them.

If the juvenile’s parent or guardian cannot be contacted, the juvenile will be released toBorder Patrol. Officers are authorized to transport the juvenile when Border Patrol is unableto respond or there would be an excessive time delay.

3. A juvenile contact report (ARJIS–8) will be completed detailing the circumstances of thedetention.

57

C. 602 W&I Minor Offenses

1. Under 13 years of age

If the parent or guardian cannot be contacted, a court order is required before Juvenile Hallwill accept them. In these cases, personnel at Juvenile Hall will be responsible for obtainingthe court order. Officers will stand by until a disposition is reached by Juvenile Hall.

In cases where a court order is not issued, the arresting officer should contact the Division’sJuvenile Detective (day or night). The Juvenile Detective, with the assistance of the JuvenileDistrict Attorney, will coordinate the placement of the juvenile in Juvenile Hall or HillcrestReceiving Home.

2. Thirteen years of age or older

If a juvenile is arrested and the parents or guardian cannot be contacted, the juvenile will beplaced in Juvenile Hall.

3. A Juvenile Contact Report (ARJIS–8) will be completed detailing the circumstances of thearrest.

D. 602 W&I Serious Offenses

1. Juveniles arrested for serious and/or violent crimes shall be placed in Juvenile Hall.

2. Officers shall photograph and fingerprint the juvenile taken into custody. The photographand fingerprints should be attached to the investigator’s copy of the juvenile contact report.

3. All arrests of undocumented juveniles shall be documented on a juvenile contact report(ARJIS–8).

For more information on the National Institute of Justice, please contact:

National Criminal Justice Reference ServiceP.O. Box 6000

Rockville, MD 20849–6000800–851–3420

e-mail: askncjrs@ncjrs.org

To access the World Wide Web site, go tohttp://www.ncjrs.org

If you have any questions, call or e-mail NCJRS.

About the National Institute of Justice

The National Institute of Justice (NIJ), a component of the Office of Justice Programs, is the research agency ofthe U.S. Department of Justice. Created by the Omnibus Crime Control and Safe Streets Act of 1968, as amended,NIJ is authorized to support research, evaluation, and demonstration programs, development of technology, andboth national and international information dissemination. Specific mandates of the Act direct NIJ to:

● Sponsor special projects, and research and development programs, that will improve and strengthen thecriminal justice system and reduce or prevent crime.

● Conduct national demonstration projects that employ innovative or promising approaches for improvingcriminal justice.

● Develop new technologies to fight crime and improve criminal justice.

● Evaluate the effectiveness of criminal justice programs and identify programs that promise to be successful ifcontinued or repeated.

● Recommend actions that can be taken by Federal, State, and local governments as well as by private organizationsto improve criminal justice.

● Carry out research on criminal behavior.

● Develop new methods of crime prevention and reduction of crime and delinquency.

In recent years, NIJ has greatly expanded its initiatives, the result of the Violent Crime Control and Law EnforcementAct of 1994 (the Crime Act), partnerships with other Federal agencies and private foundations, advances intechnology, and a new international focus. Some examples of these new initiatives:

● New research and evaluation are exploring key issues in community policing, violence against women, sentencingreforms, and specialized courts such as drug courts.

● Dual-use technologies are being developed to support national defense and local law enforcement needs.

● The causes, treatment, and prevention of violence against women and violence within the family are beinginvestigated in cooperation with several agencies of the U.S. Department of Health and Human Services.

● NIJ’s links with the international community are being strengthened through membership in the United Nationsnetwork of criminological institutes; participation in developing the U.N. Criminal Justice Information Network;initiation of UNOJUST (U.N. Online Justice Clearinghouse), which electronically links the institutes to theU.N. network; and establishment of an NIJ International Center.

● The NIJ-administered criminal justice information clearinghouse, the world’s largest, has improved itsonline capability.

● The Institute’s Drug Use Forecasting (DUF) program has been expanded and enhanced. Renamed ADAM(Arrestee Drug Abuse Monitoring), the program will increase the number of drug-testing sites, and its roleas a “platform” for studying drug-related crime will grow.

● NIJ’s new Crime Mapping Research Center will provide training in computer mapping technology, collect andarchive geocoded crime data, and develop analytic software.

● The Institute’s program of intramural research has been expanded and enhanced.

The Institute Director, who is appointed by the President and confirmed by the Senate, establishes the Institute’sobjectives, guided by the priorities of the Office of Justice Programs, the Department of Justice, and the needs ofthe criminal justice field. The Institute actively solicits the views of criminal justice professionals and researchersin the continuing search for answers that inform public policymaking in crime and justice.