National Academies Press: OpenBook

Fairness and Effectiveness in Policing: The Evidence (2004)

Chapter: 3 The Nature of Policing in the United States

« Previous: 2 Criminal Justice Research on Police
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 47
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 48
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 49
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 50
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 51
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 52
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 53
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 54
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 55
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 56
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 57
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 58
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 59
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 60
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 61
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 62
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 63
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 64
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 65
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 66
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 67
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 68
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 69
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 70
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 71
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 72
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 73
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 74
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 75
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 76
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 77
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 78
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 79
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 80
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 81
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 82
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 83
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 84
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 85
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 86
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 87
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 88
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 89
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 90
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 91
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 92
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 93
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 94
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 95
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 96
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 97
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 98
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 99
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 100
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 101
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 102
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 103
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 104
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 105
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 106
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 107
Suggested Citation:"3 The Nature of Policing in the United States." National Research Council. 2004. Fairness and Effectiveness in Policing: The Evidence. Washington, DC: The National Academies Press. doi: 10.17226/10419.
×
Page 108

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 The Nature of Policing in the United States T he public view of policing--police officers patrolling streets and re- sponding to emergency crime calls; police detectives interviewing wit- nesses, examining forensic clues, checking records and interrogating suspects; police chiefs and officers struggling to escape from the threat of political interference--is not entirely inaccurate. But it owes far too much to a particular view of policing that is presented on television and in movies for dramatic effect, and far too little to statistics and research. In this chap- ter we rely on the latter to more accurately describe the landscape of polic- ing--in the United States, a varied and complex structure that we call the "police industry." Especially noteworthy characteristics of this industry in- clude the wide variety of functions that law enforcement organizations are asked to perform and the role of individual police officer discretion, and we give both special attention. Within these two broad themes we focus on issues of importance to the rest of the committee's work: the complexity of police-public encounters and the special question of what determines when, why, and how individual police officers exercise authority and force. Fi- nally, the chapter describes recent major innovations in policing that are referred to and evaluated throughout this report. These include new polic- ing strategies and the changing profile of rank-and-file police officers. Like much of this report, most of this discussion is limited to the findings of research on big-city police, because rural and small-town policing has been so underevaluated. Because governments at all levels are concerned with fostering further efficiency, effectiveness, and fairness, this chapter also includes an examina- tion of the innovation process itself. The decentralized nature of policing in 47

48 FAIRNESS AND EFFECTIVENESS IN POLICING the United States presents special assets, but it also complicates the process by which police innovate and learn. The locally funded and locally con- trolled world of policing challenges the ability of federal and state govern- ments to promote systematic and directed change. Efforts to increase the efficiency, effectiveness, and fairness of policing are also limited by what is known. As this report makes clear, there is a tremendous imbalance in what is known about certain aspects of policing and about "what works." Some aspects of policing have been heavily stud- ied, and others have been neglected. This limits our ability to recommend specific programs or policies in many areas. The report calls for further research more often than one might hope, because the knowledge base for policing is so limited on many key topics. ORGANIZATIONAL STRUCTURE OF AMERICAN POLICING It is useful to describe American policing as an industry because, as is the case with other goods and services, services are delivered by a number of different providers (Ostrom, Whitaker, and Parks, 1978:3). Unlike po- lice in most other industrialized countries, the structure of the American police industry is highly fragmented, in the sense that there is a multiplicity of loosely coordinated agencies at different levels of government. In recent years, the relationships between public agencies at different levels of gov- ernment have been further complicated by the existence of multijurisdic- tional task forces directed toward particular problems, such as drug traf- ficking (Geller and Scott, 1992; Jeffries et al., 1998). Estimates vary, depending on the definition that is used, but the most expansive estimate is that there are as many as 21,143 federal, state, and local law enforcement agencies (Roth et al., 2000). The Bureau of Justice Statistics (BJS), however, estimates that in 1997 there were approximately 19,160 public law en- forcement agencies (Reeves and Goldberg, 2000). For many years, experts in the field believed there were 40,000 agencies (President's Commission, 1967). The number of employees similarly varies from the 608,540 police officers and civilian employees counted by the Urban Institute in 1996 (Roth et al., 2000) to the Bureau of Justice Statistics survey, which found over 1,000,000 employees, about 775,000 of whom were police officers with arrest powers (Reaves and Goldberg, 2000). The main focus of this report is on local police departments. In 1999, BJS counted 13,500 agencies at this level of government, with about 557,000 employees overall, and about 436,000 sworn police officers (Hickman and Reeves, 2001). Public law enforcement agencies vary considerably in size, as measured by the number of sworn officers (Hickman and Reaves, 2001). A relatively small number of large (1,000 or more officers) agencies (62) provide polic-

THE NATURE OF POLICING IN THE UNITED STATES 49 ing services for 17.2 percent of the U.S. population (Brown and Langan, 2001). Large cities also have a disproportionate share of violent crime: in 2000, cities with over 1 million population accounted for 18.6 percent of all violent crime reported to the police (Federal Bureau of Investigation, 2001; Langan et al., 2001). At the same time, thousands of the smallest agencies, many with one or two officers, respond to very small numbers of violent crimes. Agency size is an extremely important factor in policing: Not only do the largest police departments face the most complex responsi- bilities, in large part because they are in urban environments with the most serious social problems, but they are also intricate bureaucracies and as a consequence extremely difficult to manage and to change. Because of these variations, it is virtually impossible to talk about policing as a homogenous institution. Law enforcement agencies at different levels of government have sub- stantially different formal responsibilities, as prescribed by law, and there- fore engage in different kinds of activities. Local police (municipal police departments, county police departments, and county sheriffs) have the broadest mandate: nearly all municipal and county police agencies enforce criminal laws, maintain order, and provide miscellaneous services to the public on a day-to-day basis. Because these agencies engage in routine pa- trol throughout the communities they serve and respond to requests for service, they are the agencies most visible to the public and also have the most direct contact with them. At the local level, municipal law enforcement agencies provide the lion's share of police services in the United States, with a few sheriffs providing general policing services. The 1999 BJS agency survey (Brown and Langan, 2001) indicates that there were approximately 13,524 municipal police forces in the United States. Only 46 of these departments employed more than 1,000 sworn officers, yet these same agencies constituted 34.1 percent of all local sworn officers in the country. Meanwhile, 771 local police de- partments employ just one sworn officer. Local law enforcement is also organized at the county level, primarily around the office of county sheriff. Fewer than 100 of the 3,100 or so county law enforcement agencies go by the designation "police depart- ment," while the rest are titled "sheriff's office" (or "department") (Reaves and Goldberg, 2000). The 1997 BJS survey indicates that there were nearly 175,000 sworn officers employed by the approximately 3,000 sheriff's of- fices throughout the nation (Bureau of Justice Statistics, 2000). One of the major distinctions between sheriff's offices and county and municipal po- lice departments is that the head of the agency (i.e., the sheriff) is nearly always an elected official, while both city and county police chiefs generally are appointed executive branch employees. Sheriff's departments are also unique in that they serve all three branches of the criminal justice system:

50 FAIRNESS AND EFFECTIVENESS IN POLICING law enforcement, courts, and corrections. About 98 percent of all sheriff's departments also provide bailiff and other services to the county courts, including serving summonses and other civil law matters, and about 80 percent also operate the primary jail in their counties (Brown and Langan, 2001). State law enforcement agencies fall into two basic categories. About half are primarily responsible for traffic enforcement on highways, while the other half have general law enforcement responsibilities throughout the state. In addition, many states maintain state-level bureaus of criminal in- vestigation with broad law enforcement responsibilities. In 1996 state po- lice agencies employed about 85,000 people, 56,000 of whom were sworn officers (Hickman and Reaves, 2001). Most states also maintain other or- ganizations with special and limited law enforcement powers, such as fish and game police, harbor police, and units that guard state buildings. There were an estimated 575 such agencies in 1993 employing 14,300 people (Roth et al., 2000). In addition, state agencies often cover jurisdiction anomalies or unincorporated areas and small towns that do not have their own police forces. Basic police services are provided to these areas by state law enforcement or county sheriff's departments, often under a formal con- tractual arrangement. At the federal level, there are an estimated 69 law enforcement agencies employing a total of about 88,000 officers "with arrest and firearm author- ity" (Reaves and Hart, 2000). The responsibilities of federal agencies are generally very specific and defined by federal law--for example, the Cus- toms Bureau enforces import and export laws. While the largest federal agencies--the Immigration and Naturalization Service, the Federal Bureau of Investigation (FBI)--are well known to the public, there are also many very small agencies, for example the Library of Congress police, which em- ploys only 147 law enforcement officers. Law enforcement services are also provided by a number of special district police, which are independent or semi-independent of other units of government. Of these, the most important are American Indian tribal law enforcement police. As a result of the historic and unique legal status of American Indian tribes, many tribal authorities operate their own police departments (and in some cases entire criminal justice systems) (Feinman, 1986; Luna, 1998; Wakeling et al., 1999). These agencies are not subject to many of the state and federal laws (e.g., equal employment opportunity requirements). Little research has been conducted on tribal policing; how- ever, a recent federal report found very high rates of criminal victimization and inadequate law enforcement protection in tribal areas (U.S. Attorney General, 1997; Bureau of Justice Statistics, 1999). Apart from tribal police, there are an estimated 1,316 other special district police departments, including 117 public school system police and

THE NATURE OF POLICING IN THE UNITED STATES 51 28 transportation system police agencies. (Bureau of Justice Statistics, 1998). There are between 581 and 1,316 campus law enforcement agencies; the varying estimates illustrate the difficulties in determining even how many law enforcement agencies there are in the country (Bromley and Reaves, 1998; Bureau of Justice Statistics, 1998). Law enforcement and the criminal justice system more generally are an expensive item in the national budget. Most of the cost of policing is borne locally. Overall, local government spending on policing totaled about $41 billion during 1996-1997, the most recent figures available. This represented a near tripling in spending since 1981-1982, when local policing cost about $14 billion. In 2000, the nation's largest local police departments almost equaled the entire allocation of 1981-1982: the 62 local agencies that serve cities with 250,000 or more population spent $13.1 billion operating their departments. In 1996-1997, local government spent just over three times as much on education, the largest function of local government, as it did on policing (U.S. Census Bureau, 1997). Estimates of the cost of policing must consider the important role that overtime pay plays in certain police pro- grams. Many departments pay police officers overtime to staff special events, extend the shifts of special units, and respond to emergencies, as well as to compensate for recurring understaffing. During the 1990s, fed- eral support for overtime in order to provide personnel for projects of na- tional interest played a growing role in agency finances (Bayley and Worden, 1998). As a result, spending, rather than personnel totals, best reflects the overall magnitude of police operations. Governance of Public Police Agencies As public-sector organizations, police agencies in the United States are subject to direct public control and influence. The bulk of these agencies serve municipalities and other local jurisdictions, and they are locally fi- nanced and locally controlled. The tradition of local control, which also prevails with respect to public education in the United States, generates both unique strengths and special problems for policing. On one hand, local political control represents a fundamental aspect of democratic self-government: the principle that public agencies are subject to control by officials elected by and responsible to the taxpayers. Because of this preference, the decentralized nature of policing has been remarkably resistant to long-standing recommendations for consolidation into larger units serving several communities (Ostrom, Parks, and Whitaker, 1978). Local political control is a vital aspect of the legitimacy of the police. By providing a means by which residents can control the law enforcement agen- cies that serve their communities, the policies and priorities of these agen- cies are responsive to public concerns.

52 FAIRNESS AND EFFECTIVENESS IN POLICING On the other hand, public control and influence--or what is often la- beled "politics"--has historically resulted in problems for law enforcement in the United States, including inefficiency, corruption, and abuse of power- less groups, especially minorities (Fogelson, 1977; Walker, 1977). At vari- ous times in history, these problems have led to efforts to reform the gover- nance of policing. One approach popular in the 19th century involved shifting control of big city agencies to state government. With only a few exceptions, control later returned to the municipalities. More often reform- ers attempted to insulate police departments from politics by protecting employees through civil service regulations or guaranteeing the chief execu- tive long or lifetime tenure in office. Consequences of Fragmentation The fragmentation of law enforcement agencies has long raised con- cerns about the lack of coordination of effort among agencies in the same geographical jurisdiction and a consequent loss of effectiveness and effi- ciency. These concerns were part of a larger concern about a "nonsystem" of criminal justice and lack of coordination among all components of the criminal justice system (Law Enforcement Assistance Administration, 1973). In response to the coordination, effectiveness, and efficiency concerns, pub- lic administration experts have for decades advocated the consolidation of law enforcement agencies in certain geographic areas. Little consolidation has occurred, in large part because of opposition from entrenched political and bureaucratic interests. In addition, research by Ostrom, Whitaker, and Parks (1978:321) found that small law enforcement agencies were not as inefficient as they had traditionally been portrayed. Apart from formal consolidation, considerable coordination and shar- ing of services exists along the lines described by Ostrom et al. (1978), and there is some reason to believe that such efforts have increased in the past 25 years. The International City Management Association (Fyfe, 1983) esti- mates that police communications and jail services are among the govern- ment services most commonly shared through contracting or joint agree- ments between local governments. New concerns about regional, national, and international criminal activity have led to the creation of regional task forces or strike forces related to specific forms of criminal activity, notably drugs (Jeffries et al., 1998). The heightened national concern about terrorism following the events of September 11, 2001, has increased public interest in coordinated law enforcement efforts and new roles for local police in combating a previ- ously faraway problem. While a number of new antiterrorism measures have been launched since September 11, as yet there is little systematic re- search to serve as a guide for these new ventures. Little is known, for ex-

THE NATURE OF POLICING IN THE UNITED STATES 53 ample, about the factors related to successful interagency coordination or successful antiterrorist strategies and tactics in a domestic U.S. context. As a result of the decentralized police industry, different levels of gov- ernment exert different degrees and forms of influence on law enforcement. Role of the Federal Government Contrary to popular belief, the federal role in shaping the character of policing is relatively small. Federal influence or control over policing is expressed through each of the three branches of the federal government. For the most part, and especially in comparison to the organization of policing in many other industrial nations, the federal government has had little to do with the conduct of American policing. In the United Kingdom, for example, local agencies receive half of their annual budget from the national government, and senior police managers receive extensive train- ing at a national police academy. The budgetary process and national train- ing are used there used to ensure conformance with national standards for local policing. Presidents of the United States have the power to influence law enforce- ment in very limited ways. Most directly, the president appoints the U.S. attorney general and the director of the FBI, thereby shaping federal law enforcement policy. The president also appoints the directors of the Office of Justice Programs and the National Institute of Justice, both charged with providing assistance to state and local criminal justice agencies, thereby influencing the policies and spending of these agencies. The U.S. Depart- ment of Justice (DOJ) can also take action to protect the rights of citizens. Most recently, the 1994 Violent Crime Control and Law Enforcement Act authorized the DOJ to sue police departments for a "pattern or practice" of violating the rights of citizens; before 1994, such broad powers were lim- ited to police agencies receiving certain government funds or to individual complaints filed with the DOJ Civil Rights Division. To date about eight departments have resolved such suits with consent decrees requiring a vari- ety of organizational reforms. These legal powers are a potentially impor- tant instrument of police reform (Chapter 7 considers them, and lawsuits more generally, in more detail). The president also sponsors federal legisla- tion that may impact law enforcement. Finally, presidents appoint justices to the Supreme Court, which ultimately shapes constitutional law as it af- fects policing. Federal financial support for state and local law enforcement funds represents less than 10 percent of all law enforcement spending. This per- centage closely parallels the role of federal spending with regard to public education in the United States (8.5 percent of all spending in 1996) (U.S. Census Bureau, 1999:166). Through its research arm, the National Insti-

54 FAIRNESS AND EFFECTIVENESS IN POLICING tute of Justice (and its predecessor the Law Enforcement Assistance Admin- istration), the federal government funds research on various aspects of the criminal justice system, including policing. Under the 1994 Violent Crime Control Act, the Office of Community Oriented Policing Services had spent $7.6 billion by November 2000 to support community policing (Ramirez et al., 2001b). The Office of Justice Programs, meanwhile, has supported dem- onstration projects designed to spur innovation and promote police effec- tiveness. In the end, however, the typical local law enforcement agency could carry out its core functions without federal support. Aside from financial support, the principal federal legislation affecting police departments are the various equal employment opportunity laws. The most important is Title VII of the 1964 Civil Rights Act, which outlaws employment discrimination on the basis of race, ethnicity, national origins, religion, and gender. Other federal laws, such as the Americans with Dis- abilities Act and the Pregnancy Discrimination Act, also affect the em- ployment practices of law enforcement agencies. It is important to note, however, that these laws are negative in orientation, prohibiting various practices. They are not affirmative in the sense of creating federal standards of employment for law enforcement. Decisions by the federal courts have significant implications for police policies and practices. The most important involve constitutional aspects of criminal procedure and cover such areas as searches and seizures (Fourth Amendment to the Constitution) and interrogations (Fifth Amendment). Some of the most important decisions have had important ancillary conse- quences in terms of prodding police departments to improve recruitment, training, and supervision (Walker, 1993). The federal courts are also cur- rently involved in handling Justice Department "pattern or practice" suits and enforcing consent decrees that require major organizational reforms on the departments in question (Davis et al., 2002). Role of State Governments State governments also have a less limited but still relatively small role in shaping the overall character of public law enforcement in the United States. Historically, there has been almost no state financial support for municipal policing. Beginning in the late 1960s, the states set minimum standards for the training and certification of police officers. States have long had certification or licensing standards for other professions, but re- quirements for the police actually were developed later than those for many other jobs. State certification requirements (e.g., length of preservice train- ing) are typically lower than those required by many large urban police agencies. A number of states have instituted procedures for decertifying officers guilty of misconduct, thereby denying them the right to be police

THE NATURE OF POLICING IN THE UNITED STATES 55 officers in that state--but not other states (Puro, Goldman, and Smith, 1997). More than federal law, state statutes impact police operations. State statutes define which employees have the official status of "peace officer" and describe the authority of police to arrest and to use force. Many states also have laws governing specific police actions, such as high-speed pursuits and handling domestic violence incidents. State laws also regulate the col- lective bargaining rights of organizations representing police employees. State laws regarding the appeal or arbitration of police officer discipline cases have an impact on accountability in local departments. States also codify the criminal statutes that the police enforce, setting procedural stan- dards for apprehending and prosecuting those who violate them. State courts play a significant role, partly because they make operational the gen- eral principles of police conduct enunciated by federal appellate courts. Role of Local Governments Local governments exercise the greatest control over American law en- forcement. This control is manifested in terms of budgetary support and the appointment or election of chief executives. Through the appointment pro- cess, local officials have a strong voice in shaping the general policies of police departments. For example, a mayor might dismiss a police chief be- cause the chief is perceived to be insufficiently aggressive about fighting crime or has been unable to eliminate corruption in the department. Local governments in over 100 cities and counties (representing at least one-third of the U.S. population) have also created civilian oversight agencies, which have various roles and responsibilities with respect to complaints by the public against police officers and, as a consequence, have become signifi- cant factors with respect to regulating police behavior (Walker, 2001). Private agencies provide a wide range of law enforcement services in the United States (Kakalik and Wildhorn, 1977). This already large and still rapidly growing private security sector can variously work to supplement, complement, and conflict with the activities of public police. The division of labor between public and private agencies is increasingly ambiguous (Bayley and Shearing, 2001). The exact size of the private security industry is not known, in part because of the difficulty in measuring an enterprise that consists of many small firms employing a large number of part-time and short-term employ- ees. One study estimated more than 400,000 employees in the private secu- rity industry in 1972, with just under 75 percent of these individuals being primarily employed in the industry (Kakalik and Wildhorn, 1977:18). Some private security employees are uniformed and easily recognizable. They provide a broad spectrum of security services, such as guarding build-

56 FAIRNESS AND EFFECTIVENESS IN POLICING ings or business districts, protecting shipments of valuable goods, patrolling neighborhoods, and responding to home security alarms. Other "white- collar" security workers conduct investigations, audits, and computer secu- rity projects that parallel to a certain extent the functions of public-sector police. Not only is the so-called private security industry large and com- plex, but also there are many arrangements whereby public agencies con- tract with private security providers and private groups contract with pub- lic agencies for services. Some police departments contract out a few of their activities to these firms, including prisoner services, training, court security, dispatching, and traffic and parking control. As Forst and Man- ning (1999:15) note, "The police no longer monopolize public safety." Other Forms of Policing Public and even private police represent only a part of the law enforce- ment and security apparatus of the United States. Because of its focus on larger municipal police departments--a focus imposed by the body of re- search open for review--a great deal of "law enforcement" in the United States falls outside the purview of this report. A broad range of organiza- tions exist at least in part to protect the public from criminal attack, doing so at least in part by enforcing criminal laws. They include, for example, the Occupational Safety and Health Administration. It enforces a body of law that is both civil and criminal. Its goal is to protect workers from the negli- gent or malicious acts of employers. Yet it does not think of itself as a police organization. Neither do state-level alcoholic beverage control commissions, even though they, too, enforce laws with an eye to protecting the safety of the public. Viewed from this perspective, the police are a subset of the broader class of security and law enforcement agencies. These activities are carried out by a wide range of public agencies at the federal, state, and local levels of government. It is also important to note (but this report does not further consider) that perhaps the most important source of community security is generated by individuals acting in their own self-defense, by exercising caution and good judgment, buying locks or dogs, and sometimes banding together with other community members in collective security efforts. Even the debate over civilians carrying licensed concealed weapons has taken on "policing" overtones, through their possible crime prevention effects (Kleck, 1991). Indeed, both self-help and private security came long before public policing. Public policing as we know it was invented only about a century and a half ago, and prior to that time, enforcement of criminal laws lay in the hands of private parties. An important issue for the future is the extent to which state and local police, private security forces, and even neighborhood watch groups and

THE NATURE OF POLICING IN THE UNITED STATES 57 citizen patrols should or could be more effectively integrated into a larger national security apparatus. Concerns about terrorism, drugs, and so on seem to push in this direction. By the same token, strong traditions that favor localism and instinctively distrust powerful central governments ad- vise us to resist the temptation. Because we are focusing on American polic- ing--understood to be primarily public agencies controlled at the local level--we have not addressed ourselves to these larger structural issues, but they will doubtless become increasingly important in the future. ACTIVITIES OF THE POLICE The police have many different responsibilities. Many of these are framed in very vague terms. It is not precisely clear, for example, what maintaining order might encompass. Interpreting these vague mandates and selecting priorities from among them requires a high level of discretion on the part of police executives and policy makers. In addition, there are po- tential conflicts between many of the various responsibilities. One of the core dilemmas of policing, for example, is that public demands for effective law enforcement may seem to conflict with the responsibility to protect individual civil liberties. Growing recognition of the complexity of police roles and responsibilities has led members of the profession (which includes both law enforcement officials and policy-oriented members of the aca- demic community) to reconceptualize the police role and redirect police department activities in ways that are more in accord with the realities of policing and the needs of the American people (Goldstein, 1977; Sparrow, Moore, and Kennedy, 1990). This section examines a number of police activities; these activities serve as the context in which police execute their many responsibilities. First, police encounter the public in a number of ways, including, especially, uni- formed patrol. Under this broad rubric, the police work to maintain order, provide service, control traffic, prevent or investigate crimes, and process information. Occasionally, the police provide what can be called special- ized services--wherein discrete organizational units, like SWAT (Special Weapons and Tactics) teams, perform unique functions. This section exam- ines what is known about all of these activities. Engagement with Citizens and the Community Uniformed Patrol The bulk of police work is conducted by uniformed officers assigned to patrol specific geographic areas (beats). Typically, roughly 60 percent of all sworn officers in city police departments are assigned to the patrol bureau,

58 FAIRNESS AND EFFECTIVENESS IN POLICING and these officers have the vast majority of police officer contacts with the general public. More than 90 percent of the local police agencies that em- ploy 100 or more sworn officers assign at least three-fourths of their patrol force to automobile patrol (Reaves and Hart, 2000). A substantial propor- tion of agencies employing 100 or more officers (and nearly all of the big- city agencies) assign officers to walk foot beats, with some devoting more than a third of their patrol force to such duty (Reaves and Hart, 2000). Most agencies also conduct some patrol on motorcycles, although the vast majority of agencies assign fewer than 10 percent of their patrol officers to motorcycle duty, and none assigns more than 40 percent (Reaves and Hart, 2000). In the last decade or so, many agencies have added bicycles to the list of conveyances via which their officers patrol. Today, substantially more than half of the agencies employing 100 or more sworn officers conduct routine bicycle patrol, with some assigning more than a quarter of their patrol units to this task (Reaves and Hart, 2000). For example, big-city departments have almost universally (98 percent, in one federal study) added bicycles to their inventory of patrol vehicles (Brown and Langan, 2001). The bulk of patrol officer contacts with the public involve responding to calls for service. Overall, 65 percent of all local police officers worked by responding to calls for service in 1999. In less specialized, smaller agencies, as many at 90 percent did so (Bureau of Justice Statistics, 2001d). Histori- cally, one of the most important changes in policing has been the growth of publicly initiated, as opposed to police-initiated, interactions between po- lice and community residents. This has been driven by the development and widespread adoption of communication systems designed to allow indi- viduals to quickly contact the police and, in turn, for law enforcement agen- cies to quickly dispatch officers in response. Most important among these was the widespread adoption of a three-digit emergency number--911. People also initiate encounters by walking into their local police station, by flagging down patrol officers in the streets, and by requesting assistance via the telephone. Widespread use of cell phones has made contacting police departments or even individual officers even easier. On the police-to-resi- dent side of the coin, the most important advancement has been the use of computer-aided dispatch (CAD) systems that allow dispatchers to track the status and whereabouts of patrol officers and direct them to the person who is requesting assistance in an efficient fashion. By 1999, virtually all police agencies serving cities of 50,000 residents or more relied on computer-aided dispatching systems (Brown and Langan, 2001). About half of all calls to the police result in the dispatch of a police officer. Perhaps contrary to expectation, most of these calls do not involve either serious crimes or pressing emergencies. Officers are dispatched in a variety of situations. Research indicates that between 70 and 80 percent of dispatches are based on requests for order maintenance and service, rather

THE NATURE OF POLICING IN THE UNITED STATES 59 than criminal activities (Scott, 1981:28-30). Moreover, when officers ap- pear on the scene, they often find a somewhat different situation from what they expected on the basis of the initial call; many calls that initially appear to involve criminal activity in fact involve no actionable offense (Klinger and Bridges, 1995). The ability to easily telephone the police has affected the allocation of police personnel and the nature of patrol work itself (Reiss, 1971; Sparrow, Moore, and Kennedy, 1990). In the absence of a formal program to actively screen calls and make decisions on whether or not to dispatch a car in response (McEwen, Connors, and Owen, 1986), what the general public defines as worthy of police attention dominates police patrol work. On the positive side, this ensures that, to no small degree, the police will be highly responsive to what local residents think they need from their local police department. On the negative side, the "you call, we come" system reduces the capacity of both police chiefs and the elected officials who represent the collectively defined goals and purposes of the police to achieve those goals. Too many of their resources may be committed to responding to individu- als' insistent calls. Recent thinking on policing, which has been driven by community po- licing and increased interest in crime prevention, has placed a high priority on increasing proactive activities initiated by uniformed patrol officers. Con- sequently, attention has turned to the larger management process through which police agencies engage in strategic planning, set clear priorities, and mobilize their resources. Proactive or officer-initiated police work involves a number of different activities: field interrogations of pedestrians, traffic stops, checks of buildings or other areas for possible criminal activity, and informal contacts with the public. There is substantial variation in the level of officer-initiated activity in different departments (Mastrofski et al., 1998), although there is no research on the factors that shape these differences. While these variations appear to be a consequence of department priorities (as communicated through management styles) and local organizational cultures, direct research of this issue would be particularly valuable, since community policing and problem-oriented policing programs have placed much emphasis on increasing proactive, officer-initiated activity. Even though reacting to calls for service constitutes the bulk of police officer encounters with the public, patrol officers retain a certain degree of autonomous control over portions of their working hours. Officers spend considerable amounts of time handling routine paperwork and other ad- ministrative duties (Kelling et al., 1974; Parks et al., 1999). Moreover, they have some limited capacity to manipulate their availability for calls for service. Patrol officers also spend a significant part of their unassigned time conducting what is known as "preventive patrol" (Kelling et al., 1974).

60 FAIRNESS AND EFFECTIVENESS IN POLICING Since the advent of modern police forces in the 19th century, preventive patrol has been the core police activity, designed to prevent crime by deter- ring potential offenders through a visible police presence, to create feelings of public safety, and to make officers available for service in local commu- nities with uniformed patrol forces. One important idea animating interest in problem-solving and commu- nity policing (which are discussed in detail below) is that both call for less reactive and more proactive uses of the uniformed patrol force. Of course, the idea that the police patrol operations ought to be proactive is hardly a new idea. One goal of patrol was always to spot and investigate suspicious activity as well as respond to calls for service. Such activities were justified by a strong interest in preventing crime as well as responding after crime had occurred. Indeed, in the late 1960s, some police departments were rely- ing on a police tactic described as "aggressive preventive patrol" (which bears a close resemblance to what is often described as "zero tolerance policing" today). The aim was to make the police a credible presence on the street by using their powers to stop people and ask questions. The hope was that, by doing so, the police could discourage offenders from committing offenses and interrupt situations that were leading to crimes. To create time for such activities to occur, police administrators had to find some way to protect some of their patrol resources from nonemergency demands of the public. To accomplish this goal, they developed "differen- tial response systems." This included efforts to distinguish among the calls for which an immediate response was very valuable (e.g., a crime in progress), from those for which an immediate response was not needed (e.g., a call about a burglary that had happened several days before, or a request for a kind of assistance that was relatively low priority compared with either responding to an urgent crime call or remaining available to respond to such a call). It also included efforts to sequester certain parts of the patrol force from the calls for service so that they could concentrate on proactive efforts. These "special operations units" often engaged in "di- rected patrol" tactics. Some of these directed patrol operations were de- scribed as "location-oriented patrol" (which focused on locations that were thought particularly likely to be the site of crimes). Others were "person- oriented patrols," which focused on individuals who were thought to be particularly likely to commit offenses. Research on these proactive patrol methods tended to produce ambigu- ous results in terms of their crime control effectiveness. On one hand, Wil- son and Boland (1980) found that one of the most powerful variables in explaining which police departments seemed to be more effective at reduc- ing crime was the frequency of police traffic stops. They interpreted the traffic stops as an indicator of how proactive police departments were in searching for crime and trying to prevent it rather than simply waiting for it

THE NATURE OF POLICING IN THE UNITED STATES 61 to occur. On the other hand, studies of location-oriented and perpetrator- oriented enforcement strategies carried out by the Police Foundation found little evidence of crime control effectiveness (Martin and Sherman, 1986). These issues are examined in detail in Chapter 6. Regardless of their crime control effectiveness, these proactive police tactics generated a great deal of resentment among those who found them- selves the objects of this special police attention. The methods also drew the ire of political figures representing the interests of the minorities and the poor who were overpoliced by these methods. The Kerner Commission, established to understand the causes of civil disorder in U.S. society, also found the abrasive encounters between police and citizens fostered by these tactics to be one of the important causes of the urban disorder of the late 1960s. An important question is whether the new forms of proactive policing are any different from the old forms, especially in what is done about a wide variety of calls that are "in the middle" in terms of urgency. Today, police agencies have reasons to focus on both less serious offenses and on noncrime calls. The theory of order maintenance or "broken windows po- licing" has made minor offenses an important focus of police patrol opera- tions. Such activities have been shown both to reduce serious crime and to enhance feelings of security. The theory of community policing also makes a virtue of providing service that is responsive to public priorities. Such efforts can be considered valuable in themselves ("service to the customers of policing"). But they can also be viewed as important ways of building both the legitimacy of the police and engaging the public in efforts to help the police accomplish their law enforcement objectives. It also remains true that some police departments, under the rubric of community and problem-oriented policing, have embraced strategies that look a great deal like some of the old proactive strategies of policing. Ag- gressive preventive patrol has been resurrected as zero tolerance policing. Directed patrol operations, in which the police simply dispatch police offic- ers to places and times where crimes are likely to occur, have reemerged as a limited form of problem solving policing. Location oriented patrol has reemerged as "hot spot policing" in which the police identify particular locations in which crimes are likely to occur, and dispatch patrol officers to watch over the hot spot. Yet what seems most importantly new about the proactive forms of policing recommended under the rubric of community policing has less to do with proactivity than with the creation of the "warrant" for the proac- tive police activity. A crucial difference between reactive and proactive po- licing is that individuals calling the police for help provide legitimacy for the police in reactive policing, while in proactive policing legitimacy comes from the professionalism of the police themselves, as they identify crime

62 FAIRNESS AND EFFECTIVENESS IN POLICING and other community problems. One key concept of community policing is that communities should have a role in shaping police priorities with re- spect to the policing of their neighborhoods. Citizen guidance emanates not just from city hall and not just from individuals calling the police. Instead, formal and informal consultations with the community give the police some guidance about priorities, allowing them to be responsive not only to indi- vidual demands or city wide mandates, but also to local neighborhood con- cerns. This consultation, in turn, could be expected to warrant police ac- tions through a political agreement as well as their legal authority and professional commitments. What might be new about the way that commu- nity policing approaches the idea of proactivity, then, is not only to agree that it is important, but also to take some pains to limit and justify the form that proactive policing takes. Community policing urges community con- sultation to create a political warrant for proactive policing. Interacting with the Public Police patrol work involves face-to-face contacts between the public and police officers. The nature and outcomes of these encounters are occa- sionally the subject of public attention and controversy. Understanding these common but complex and potentially volatile interactions has therefore been a major focus of police research. Yet studying this subject is a daunt- ing research task, since police-public contacts take place episodically, in the field, 24 hours a day (Mastrofski et al., 1998). Official police data (e.g., 911 calls) do not provide meaningful information on police-public encounters, either in terms of the nature of the incident or the behaviors of civilians and police officers. Consequently, the bulk of knowledge comes from researcher- initiated, observational studies. One approach to investigating the frequency and type of police-public interactions is through surveys of the public, asking them about their recent experiences, if any, with the police. A national survey conducted by Langan et al. (2001) found that about 21 percent of people age 16 and older (or almost 44 million people) had face-to-face contact with a police officer during 1999. One-third of them reported having more than one encounter. More than half of all contacts were motor vehicle stops, highlighting the importance of how police conduct these encounters. Only about 3 percent reported that they were stopped because police suspected them of involve- ment in any sort of nontraffic crime. More--about 8 percent--were in- volved in a traffic accident. About one-third of those interviewed recalled coming into contact with police because they reported or witnessed a crime or a traffic accident, and 12 percent because they asked police for some sort of assistance.

THE NATURE OF POLICING IN THE UNITED STATES 63 While such surveys provide useful context, they do not fully divulge the nature of police-public contact. An alternative method of assessing encoun- ters is direct observation, usually by riding with officers on patrol or walk- ing with them on foot (Teplin, 1986). This is a very powerful way of ob- taining accurate information about the character and the dynamics of police encounters with the public, but it is also an extremely difficult and expen- sive type of research (Mastrofski et al., 1998). The first systematic observa- tion of police patrols generating quantifiable data occurred in 1966, when a team of researchers accompanied officers from the Boston, Chicago, and Washington, DC, police departments on patrol, systematically recording what transpired during each shift (Reiss, 1971; Black, 1980). Reiss (1971) found that patrol work is organized around discrete interactions with the public, the vast majority of which are initiated by the publics' requests for police assistance. He labeled these public-initiated interactions "reactive" encounters to distinguish them from interactions that were initiated by the officers themselves, which he called "proactive" encounters. Subsequent studies have explored the various contours of police-public encounters (Bayley, 1986; Teplin, 1986; Bayley and Garofalo, 1989; Mas- trofski, Snipes, and Supina, 1996; Mastrofski et al., 1998). Earlier studies, focusing on police actions, characterized those with whom they came into contact using demographic categories. More recently, researchers have at- tended to the behavior of individuals who encounter the police, attempting to analyze how the actions of one party affect the actions of the other (Mastrofski, Snipes, and Supina, 1996). Exercise of Authority The core element of the police role in society is the use of their author- ity (Bittner, 1970). This authority consists of the capacity, at one extreme, to deprive people of their liberty (e.g., arrest), use physical force, or even to take human life. But lesser uses of force and authority include observations of conduct in public spaces, interrupting ordinary activities to ask ques- tions, and the development of records that help shape individual reputa- tions. A police officer's authority is implicit in the role and is understood by members of the public even when it is not overtly exercised in a particular incident. In a classic essay, Bittner (1970) argues that the capacity to use force lends "thematic unity" to all the police contacts with the public. Even when officers do not use force, the knowledge that they have the authority to use it shapes the behavior of both the public and police officers. In this context, use of authority involves different dimensions and levels of coer- cion. Because police authority includes degrees of coercion, police depart-

64 FAIRNESS AND EFFECTIVENESS IN POLICING ments increasingly train officers to use the least amount of coercion or force appropriate to resolve any given situation (Garner and Maxwell, 1999). The exercise of authority represents only one of several kinds of re- sponses available to a police officer. Black (1976:5; 1980:130-131) concep- tualized the range of responses in terms of four styles: penal (e.g., arrest), conciliatory (e.g., mediation), compensatory, and therapeutic (e.g., counsel- ing, referral). His research found that police utilized the penal and concilia- tory styles far more often than others, and that they frequently employed more than one style in the course of an encounter. He also found that, in 4 percent of cases, police declined to take any action at all. Other studies of how they handle disputes confirmed that police occasionally refuse to get involved. Klinger (1996) developed a measure that rank-orders officers' ac- tions according to the amount of authority they exerted to resolve encoun- ters. He reported that officers typically used low levels of authority, taking such actions as offering advice and resolving points of contention via dis- cussion in 60 percent of the cases they handled. Worden (1989) developed a 16-category measure that captured the combinations of actions that officers might undertake to resolve disputes. He found that officers used tactics that involved some form of counseling, coercion (exclusive of arrest), or both in about two-thirds of the disputes they handled. Arrest as an Exercise of Authority The arrest of an individual represents a major exercise of coercive au- thority, a deprivation of liberty. The exercise of discretion in arrest has been a major focus of research, with particular attention to the question of whether there are patterns of discrimination against racial and ethnic mi- nority groups (LaFave, 1965; Reiss, 1971; Black, 1980). The first study of the arrest decision confirmed that officers exercised broad discretion in making arrests, but they often made arrests for purposes other than crimi- nal prosecution and in some cases used their powers illegally (LaFave, 1965). The first large-scale quantitative study of arrest discretion found that offic- ers systematically underenforce the law, making arrests in only about half of all situations in which there was legal grounds for an arrest (Black, 1980:85-108). This study concluded that the arrest decision was largely shaped by situational factors, including primarily the seriousness of the al- leged offense, the strength of the evidence, the preference of the victim or complainant, the relationship between the victim and offender, and the de- meanor of the suspect. This study did not find a clear pattern of race dis- crimination in arrest. Because of public concern about domestic violence in the past 30 years, police handling of domestic disputes has been the subject of considerable research and experimentation. One highly publicized reform effort in the

THE NATURE OF POLICING IN THE UNITED STATES 65 early 1970s sought to train police in nonarrest dispositions of domestic disputes (Bard, 1970). As a result of lawsuits and concern that battered women were placed in jeopardy because of police failure to arrest male batterers in domestic violence situations, public policy shifted to an empha- sis on mandating arrest in such situations (Loving, 1980). The first observa- tional study of domestic violence incidents (Worden and Pollitz, 1984) found that officer decisions were affected by the same situational factors that affect arrests generally, concluding that police do not treat domestic and other disputes differently. This conclusion was supported by Klinger's (1995) finding that the odds of arrest were no lower in spousal assault than in other assault cases. Observational studies of police responses to intimate partner (or domestic) disturbances or violence also found that encounters involve far more than a simple arrest/no arrest decision. However, more recently, some scholars have raised the possibility that police do treat inti- mate partner violence differently (Fyfe et al., 1997; Avakame et al., 1999; Avakame and Fyfe, 2001). Confirming this differential would once again raise important questions regarding police discretion and the factors that most influence officers in situations that require relatively quick and impor- tant decisions. Another line of research has investigated the deterrent effect of arrest, including the assertion that failure to arrest in domestic violence situations contributes to violence against women. The first study of this subject--one of the most highly publicized studies in all police research--found that ar- rest deterred future domestic violence more effectively than either counsel- ing or separating the disputing parties (Sherman 1992). Replications of the original Minneapolis Domestic Violence Experiment have not supported its findings, however (Maxwell et al., 2001). Some evidence suggests that, in recent years, the police have generally been responsive to concerns about the need to strike a sound balance be- tween the individual citizen's rights and the need for social order. This re- sponsiveness (like much in policing today) grew out of social conflicts that came to a head during the 1960s. Social protest against segregation, America's involvement in the Vietnam War, and other concerns led to nu- merous mass demonstrations (which sometimes led to rioting) in many major cities in the 1960s and into the 1970s. The way that the police typi- cally responded to the mass demonstrations during that era was to use force (e.g., nightsticks, water from fire hoses, and so on) to disperse the protest- ors. This approach, dubbed "escalated force" by scholars who study the policing of protest (e.g., McPhail, Schweingruber, and McCarthy, 1998), was often ineffective in dealing with mass demonstrations and led to a con- cern in many quarters that the police were using excessive force. In the late 1980s, these concerns led many police departments to move away from the escalated force model to one that relies instead on meeting with protest

66 FAIRNESS AND EFFECTIVENESS IN POLICING leaders before demonstrations take place in order to set the parameters for demonstrator behavior and police action. Some large police departments, for example New York City, had been communicating with demonstration leaders since the late 1960s, pioneering this model. With this new negoti- ated management model of dealing with protest (e.g., McPhail, Schwein- gruber, and McCarthy, 1998), the police agree to protect the First Amend- ment rights of protestors so long as they do not blatantly violate the law (and sometimes even negotiate where and when arrests will be made when protest leaders indicate that the planned demonstration will include civil disobedience). Use of Lethal Force Reflecting intense public concern over fatal shootings, a substantial body of research has accumulated over the past 30 years on police use of deadly force. Following some initial research in the 1960s (Robin, 1963), the most influential work began to appear in the late 1970s. Milton et al. (1977) examined shooting incidents in seven large U.S. cities during 1973-1974 and found that shootings were relatively infrequent compared with other police actions (e.g., arrest). They recorded 320 incidents across the seven cities during the two study years and noted that over two-thirds of civilians shot by the police survived their wounds and that blacks were shot at higher rate than whites, but at a rate consistent with arrests of blacks for serious crimes. Finally, the researchers found that the seven cities had mark- edly different rates of police shootings, ranging from 1.6 to 8.5 shootings per 100,000 population. Such variability in the use of lethal force has been a consistent finding. Fyfe (1978) reported on shootings by police in New York City during 1971-1975, expanding the scope of research from situations in which indi- viduals were struck by police gunfire to all situations in which officers fired shots. He found that, in fact, the vast majority of shots fired by police did not hit anyone. Fyfe also reported that the rate at which shootings occurred varied substantially across the city; that shootings were more likely to occur in areas with higher crime rates; and that area of assignment was associated with individual officer shooting rates. The race of the officer did not ex- plain variations in shooting rates. In Fyfe's study, black officers fired their weapons more often than white officers, and this was explained largely by their low seniority, which resulted in more assignments to high-crime areas and also because these officers were more likely to live in such neigh- borhoods. In addition, it should be noted that New York's highest crime neighborhoods at the time were largely nonwhite, and the assignment of nonwhite officers to these neighborhoods was largely a result of the depart- ment's attempts to make officers on the street representative of the neigh-

THE NATURE OF POLICING IN THE UNITED STATES 67 borhoods they policed. Fyfe also found that the number of shootings de- creased substantially following the promulgation of a new, more restrictive department policy regarding the use of firearms by officers. Subsequent studies have confirmed such matters as variability in shoot- ing rates both between and within police agencies, and they have offered fresh insights into the use of deadly force by police officers (Skolnick and Fyfe, 1993). One line of research has focused on the effects of a U.S. Su- preme Court decision (Tennessee v. Garner, 1985) limiting the use of deadly force to the situations in which it is necessary for the defense of life or to effect the arrest of fleeing suspects who committed violent crimes. The use of deadly force by police officers decreased markedly following the Garner ruling, falling from about 400 in 1983 to approximately 300 in 1987 (Geller and Scott, 1992). Brown and Langan (2001) noted that the immediate post- Garner decrease was a continuation of a trend in fatal shootings that began in the late 1970s. However, the number of fatal police shootings increased again during the late 1980s, then began to decline again during the second half of the 1990s. Use of Physical Force Public attention has also focused on police use of physical force (Langan et al., 2001; Alpert and Dunham, 1999; Geller and Toch, 1995). Studies have consistently found that police officer use of force is statistically rare. Observational studies have found force to be used in about 1 percent of all police-public encounters, with the rate rising to about 4 percent for encoun- ters involving criminal suspects. The most comprehensive national survey of police use of force is the victimization-style survey conducted by Langan et al. (2001). As reported earlier, this national study found that an esti- mated 21 percent of all adults in America had any contact with the police during the previous year, that more than half of those contacts were the result of a traffic stop, and that officers used force in slightly less than 1 percent of all encounters. Studying police use of force raises a number of extremely difficult meth- odological issues. There is no universal definition of what constitutes force. Policies requiring officers to report use of force incidents vary from depart- ment to department. The extent of officer compliance with reporting re- quirements is not known and probably varies from department to depart- ment. Finally, judgments as to what constitutes excessive force are difficult, whether they are made by department administrators or researchers (Adams, 1995; Alpert and Dunham, 1999). The BJS national survey, examining police use of force during encoun- ters of all kinds, found that most respondents who were subject to force felt police had acted improperly. Subtracting the 20 percent who were sub-

68 FAIRNESS AND EFFECTIVENESS IN POLICING jected only to a threat to use force unless they complied, about 0.8 percent of Americans age 16 and older (about 340,000 individuals) were pushed, grabbed, kicked, hit, or sprayed by police or bitten by police dogs during 1999. More than half reported that they had themselves been drinking or were obstreperous at the time, and half had charges filed against them. However, three-quarters of those interviewed characterized police use of force as excessive, and more than 90 percent felt police had acted improp- erly during the incident. About 15 percent reported they were injured in some way as a result of these police actions (Langan et al., 2001). The BJS estimate that police use force in approximately 1 percent of all encounters with the general public is consistent with previous estimates using different methodologies (Adams, 1995). Beginning in the 1980s, a smaller body of literature has developed on police use of nonlethal force--actions ranging from mild physical restraint, punches and kicks, and baton blows. Officers use this type of force far more frequently than firearms. Friedrich (1980) found police used some form of nonlethal physical force in about 5 percent of their encounters with sus- pects. Nine years later, Bayley and Garofalo (1989) reported that the New York City officers they observed used some form of nonlethal force against civilians in 8 percent of the encounters that the researchers classified as "potentially violent." In 1996 Worden reported a nonlethal force rate of 4 percent among the police-public encounters observed in many neighbor- hoods in three metropolitan areas during the 1970s. In sum, observational research conducted during the 1960s, 1970s, and 1980s indicates that while officers use nonlethal force far more often than they use deadly force, they use it in a small minority of their interactions with the public. Research on nonlethal force during the 1990s considered the techniques that officers use when applying force. McLaughlin (1992) examined use of force reports filed by the Savannah, Georgia, police department. He devel- oped a use of force continuum, based on police training that directed offic- ers to use force proportionate to the degree of resistance offered by their protagonists. This and similar taxonomies rank-order specific actions that officers might take, from high physical force (e.g., shootings, baton strikes) to low force (e.g., grabbing someone) (Alpert and Dunham, 1999). Klinger (1995) utilized this continuum to assess the amount of force officers in the Miami area used when handling disputes between civilians. He reported that more than half of the time officers used physical force they merely grabbed them; that officers rarely struck members of the public with their batons; and that when officers used force more substantial than simply grabbing someone, they used more than one type of force (e.g., punching and grabbing, baton strikes and kicks). Klinger's force continuum included voice commands, and he found that verbal directives were used more often than physical force.

THE NATURE OF POLICING IN THE UNITED STATES 69 Research using a variety of methods has confirmed the key findings of Garner et al. (1996), Klinger (1995), McLaughlin (1992), and others that officers use physical force infrequently or that, when they do, it typically consists of minor forms of force (also see Garner and Maxfield, 1999; Alpert and Dunham, 1999; Terrill, 2001). It is important to note that changes in police technology have had some effect on patterns of use of force. In part because of public controversy over police use of deadly force, nonlethal technologies have been developed and adopted by police departments. These technologies include chemical agents, electronic stun guns, and various other tools designed to subdue the combative without killing or seriously injuring them. During the 1990s, more police agencies began to deploy "impact munitions," that is, firearm-delivered projectiles, such as beanbags and plas- tic batons that are designed to minimize serious bodily injury or death when they are deployed. Klinger and Hubbs (2000) examined 373 cases in which officers fired impact munitions that struck an individual and found that injuries were limited to bruises, abrasions, and lacerations in most of the cases, but that several targets of impact munitions suffered more serious wounds, and 8 died. While impact munitions often work as intended, the police have yet to find a technological fix for the consequences of their elemental task, the application of lawful force to protect society. Other Means of Exercising Authority Relatively little attention has focused on the other less serious forms of the use of authority, although academic research and a number of police department policies recognize a continuum ranging from the least to the most serious forms. Officers frequently gain compliance through their mere physical presence. Officers can increase their level of control over a situa- tion through different forms of body language and where and how they place themselves. Verbal statements can also be authoritative, ranging from asking questions, making suggestions, and giving orders (Bayley and Garof- alo, 1987). An officer's tone of voice can radically affect the level of coer- civeness of a statement or a question. Misunderstandings related to verbal expressions and nonverbal behavior, including use of force, are often the source of conflict between police and racial and ethnic minority groups. Maintaining Order Recent thinking on policing has placed a high priority on increasing proactive or police-initiated activities. A major part of the proactive work- load of police has historically been what is called order maintenance or peacekeeping policing. The idea that policing consisted primarily of enforc- ing laws dominated scholarly thinking about the police until the late 1950s,

70 FAIRNESS AND EFFECTIVENESS IN POLICING when the results of the American Bar Foundation study (1957) of policing emerged and, as noted earlier, demonstrated that police officers exercised considerable discretion and frequently did not arrest individuals whose ac- tions clearly violated criminal statutes (Goldstein, 1960; LaFave, 1962, 1965:61-143). A second key finding was that officers often arrested people with no intention of preferring criminal charges against them; for example, in order to protect inebriates whose drunken state renders them vulnerable to criminal attack (see especially LaFave, 1965:437-489). Taken together, these findings indicate that the police are not automatons who merely en- force legal codes, but rather, they exercise great discretion in enforcing the law selectively to accomplish a variety of purposes. Research conducted in the 1960s disclosed additional insights into the nature of police work vis-à-vis the criminal law. Bittner (1970), LaFave (1965), Reiss (1971), and Wilson (1968), among others, found that officers often invoke their powers to take individuals into custody for the purpose of maintaining control over the immediate situation or controlling social problems for purposes of maintaining public order, as opposed to arresting them for the purpose of punishment. Officers, for example, often jail indi- viduals as a means to prevent potentially volatile situations from spinning out of control and to restore order when it had already been breeched. The research also disclosed that police officers exercise their authority through a variety of means other than arrest to control situations and to maintain order. This includes using their physical presence (body language), verbal expressions (questioning, suggesting, ordering), and the use of physical force (LaFave, 1965; Mastrofski, Snipes, and Supina, 1996; Terrill, 2001). The early research on police activity (President's Commission on Law Enforcement and the Administration of Justice, 1967b:13-41) led to a new consensus that the police were generally peacekeepers rather than law en- forcement agents, and that the central goal of the police was to maintain social order with criminal law enforcement as simply one of a number of means of achieving this goal. Goldstein (1977:33) characterized the police as "an agency of municipal government housing a variety of functions." As Wilson (1968:31) put it, officers deal with many incidents "not in terms of enforcing the law, but in terms of `handling the situation.'" Many order maintenance activities of the police are relatively mun- dane. Traffic control, for instance, can be seen as a form of order mainte- nance to the extent that it facilitates people going about the routines of their daily lives without interruption. Special events, such as sporting events or parades, also call on the police to maintain order. The policing of overtly political events is far more complicated, however. As noted earlier, in events in which there is an element of political protest, the police must balance their responsibility to maintain order with the equally important responsi- bility to protect individual civil and political liberties, such as freedom of

THE NATURE OF POLICING IN THE UNITED STATES 71 speech and assembly. The changing face of disorder, however, militates against simplified and universal strategies in dealing with protests or distur- bances. For instance, during recent World Trade Organization meetings, a minority of protesters deliberately attempted to create disturbances; inter- ference reached such an extent during one meeting held in Seattle that it was dubbed the "battle for Seattle." Needless to say, the police represent the front line in these "battles" and often confront difficult situations, and these new strategies of the minority jeopardize any accord reached between the majority of demonstrators and the police. It must also be noted that some disturbances--such as rioting after athletic events--feature the im- paired judgment of a number of inebriated participants; this can also create difficulty for police trying to maintain order without relying on broad use of law enforcement powers. Providing Service Another core feature of police activity involves the provision of miscel- laneous services that are divorced from the law enforcement and order main- tenance or peacekeeping functions (Goldstein, 1977). Research indicates that service activities constitute a major portion of what patrol officers do day in and day out, making up between one-third (Cumming et al., 1965) and one-half (Wilson, 1968) of all calls for service to police departments. These services include but are not limited to giving directions, answering questions from the public, monitoring crowds at public events, finding lost children, assisting motorists who have locked themselves out of their ve- hicles, escorting merchants to late night depositories, and ensuring that a drunk person makes it safely home. Much police research over the past three decades has been directed toward enhancing understanding of how police officers ration their time among these three groupings--law enforcement, order maintenance, ser- vice provision--on a daily basis. Studies of patrol work focusing on con- tacts with the public, for example, indicate that patrol officers devote be- tween 15 and 30 percent of their time to criminal law enforcement, with the remaining time devoted to order maintenance and service responsibilities (Reiss, 1971:96; Scott, 1981:28-30). Kelling et al. (1974:40-43) examined officers' use of time through full 8-hour shifts and found that about 14 percent was allocated to mobile police-related patrol, with an almost equal amount related to nonenforcement activities. These studies focus exclusively on patrol officers, however, and there are no estimates of the distribution of department-wide resources. Studies of criminal investigation, for instance, have focused only on the factors that correlated with the successful appre- hension of criminal suspects (Eck, 1983a) and not on the allocation of in- vestigators' time.

72 FAIRNESS AND EFFECTIVENESS IN POLICING Controlling Traffic In most large agencies, traffic enforcement is conducted primarily by officers assigned to a specialized traffic unit (although general patrol offic- ers also make traffic stops). Traffic stops comprise the most common form of contact between police officers and the public (Langan et al., 2001). Recently, traffic enforcement has found itself at the center of great interest because of the controversy over racial profiling (Ramirez et al., 2000a; Fridell et al., 2001; Harris, 2002). During 1999 police stopped about 10 percent of all licensed drivers and gave more than half of those a traffic ticket. They searched about 7 percent of those they stopped (for a total of about 1.3 million searches) and hand- cuffed and arrested about half of those they searched (Langan et al., 2001). It is hard to establish the context behind these numbers, since no detailed study of traffic enforcement has been undertaken since 1969 (Gardiner, 1969). The current national controversy over racial profiling has provoked changes in legislation, administrative policies, and record-keeping practices (Ramirez et al., 2000b; Harris, 2002). A large number of data collection efforts at both the local and state levels are under way; some, which came as the result of lawsuits against the police, are already complete or have been ongoing for some time (e.g., in New Jersey, Maryland). Because of this massive new investment in data collection, one might expect that knowl- edge about traffic enforcement patterns will increase dramatically in the near future. Yet one major unresolved issue behind understanding these numbers involves the proper baseline to be used to determine whether or not enforcement patterns represent impermissible patterns of racial and eth- nic discrimination (General Accounting Office, 2000a, b; Walker, 2001). Racial profiling is discussed in more detail in Chapter 8. Preventing Crime Developments in policing over the past 20 years have blurred the dis- tinction between the traditional categories of law enforcement, service pro- vision, and order maintenance. There has also been a new interest in ex- panding the traditional notions of the police role in crime prevention, creating a fourth general category of police activity. While many crime pre- vention programs include activities that have been defined along with more traditional categories, emergent ideas of crime prevention also include op- portunity reduction strategies, such as target hardening, situational crime prevention, and reduction of repeat victimization. In addition, expanded crime prevention strategies often involve police mobilization of third par- ties to exert informal social control (Clarke, 1997).

THE NATURE OF POLICING IN THE UNITED STATES 73 Crime prevention embraces many of the police activities that have tra- ditionally been conceptualized as order maintenance and law enforcement, but attention to it has recently been bolstered by community policing. Crime prevention relies on both punitive and positive models of deterrence: The arrest of criminal suspects, for example, is a law enforcement activity, but insofar as it deters future criminal behavior, it also has the effect of prevent- ing crime. By the same token, the successful deescalation of a domestic disturbance by a patrol officer has traditionally been thought of as order maintenance but has the effect of preventing crime. (On the traditional false dichotomy between crime prevention and law enforcement, see Sherman et al., 1997). The old (pre-1980s) crime prevention activities primarily involved pub- lic education efforts. Typically, officers assigned to a special unit would seek to educate people about how they could prevent residential burglaries ("target hardening") and avoid behaviors that would increase their risk of violent victimization. The new crime prevention model tends to involve a wider range of officers, to be more closely integrated into basic police op- erations and to be more proactive. Many, if not most, of these activities are closely related to community policing and problem-oriented policing activi- ties and involve neighborhood residents and partner agencies in the com- munity (Rosenbaum, 1989). Investigating Crimes The investigation of alleged criminal incidents is a basic police func- tion, but probably one more significant in the popular mind than in the context of everyday policing. While investigation has been emphasized in entertainment media, in practice, criminal investigation plays a relatively small role in the day-to-day activities of police departments. In large depart- ments, investigations are generally conducted by specialized units, such as homicide, robbery, sexual assault, property crimes, and so on. In small agen- cies, there is little specialization of function of any sort. Criminal investiga- tion units typically involve only about 10 percent of all sworn officers in an agency. Criminal investigative activities fall into two broad categories: reactive and proactive. Reactive activities involve attempting to identify and arrest the perpetrator(s) of reported crimes. Among the most important functions in this regard are the collection and official cataloging of the criminal evi- dence required to successfully prosecute individual offenders. Proactive ac- tivities involve attempting to expose criminal events or behavior and only then to identify and arrest the perpetrators. Among the targets of proactive efforts are so-called victimless crimes, including narcotics, gambling, and prostitution (Greenwood et al., 1977; Eck, 1983b).

74 FAIRNESS AND EFFECTIVENESS IN POLICING The portrayal of detectives in the entertainment media is not a reliable guide to actual criminal investigative work. In 2000, the police "cleared by arrest"--meaning officers arrested, charged, and turned over to the courts for prosecution a person or group of people--only about 20.4 percent of all violent and property crime (Federal Bureau of Investigation, 2001). In the majority of these cases, an arrest is made by the first officers responding to the reported crime. Among crimes that are solved, success in making an arrest is largely owed to a victim or a witness who names the suspect even before detectives arrive on the scene (Eck, 1983a). In these cases, criminal investigators "begin with an identification, then collect evidence; they rarely collect evidence and then make an identification" (Bayley, 1994:27). Much detective work is therefore done retroactively and begins with an arrestee in hand or readily accessible. Detectives have the job of gathering the evidence to convict the offender, most notably a confession, and do the bulk of their work after the suspect is in custody (Ericson, 1981; Brandl, 1993). Cases in which no suspect is initially identified typically go unsolved; the RAND Corporation estimated that only 3 percent of arrests were the result of the activities of detectives (Greenwood et al., 1977). Low clearance is particularly true for property crime (Bynum et al., 1982), but Wellford and Cronin (2000) also attribute declining clearance rates for homicide to the increasing proportion that are perpetrated by strangers, who leave fewer clues to their identity. Research on the factors that are linked to "solvabil- ity" have been used by most police agencies to establish guidelines for de- termining which cases will actually be investigated, with the remainder be- ing administratively closed (Eck, 1983b). Factors included in assessments of the ability to "clear" a crime are related to the nature of the crimes them- selves, in particular whether or not the officers initially responding to the crime were able to obtain good information about a specific suspect. The quality of the investigations and the witness interviews conducted by uni- formed officers who are first on the scene seems to be the primary determi- nant of their eventual solution (Greenberg and Hantz, 1975; Eck, 1983b; Skogan, 1985; Wellford and Cronin, 2000). As a result of this research, some have questioned the need for elaborate detective units (Bayley, 1998); similarly, Greenwood et al. (1977:141) concluded that most detective work could be performed by clerical personnel. During the 1970s there was some interest in delegating responsibility for conducting follow-up investigations to responding officers on the scene, in light of findings about the impor- tance of their role in actually clearing cases. Others are more sanguine about the role of detectives, especially when their work is disciplined by priority- setting and careful case selection and management (Eck, 1983b; Brandl and Frank, 1994). Proactive police investigations raise troubling questions about police behavior. First, gambling and narcotics units have historically been the cen-

THE NATURE OF POLICING IN THE UNITED STATES 75 ter of the worst forms of police corruption (Knapp, 1973; Sherman, 1978a, b). Second, proactive police work often involves the use of wiretaps or other intrusive measures that raise civil liberties concerns (Marx, 1988). Finally, undercover work necessarily involves the use of deception by police offic- ers, a practice that potentially compromises their moral integrity (Marx, 1988). In response to the terrorist attacks on the United States on September 11, 2001, there is increased interest in the role of state and local law en- forcement agencies in both the prevention of terrorist crimes and the re- sponse to any such crimes as occur. At present, however, little is known about these capacities. It is generally believed that most law enforcement agencies have some formal plans for responding to large-scale disasters, such as civil disorders, earthquakes, tornados, and the accidental release of hazardous materials. There are neither survey data on the prevalence and nature of such plans, nor social science research evaluating the effectiveness of these plans or special units dedicated to these purposes. The periodic reports from the Law Enforcement Management and Administrative Survey (LEMAS), for example, report data on various special units but include no data on disaster planning (Reaves and Hart, 2000). Processing Information In a broad sense, one of the most important roles of the police is to process information (Erickson and Haggerty, 1997). That is to say, law enforcement agencies collect information, process it, act on it, and in some cases provide it to other agencies including the public; some of the informa- tion collected by the police is unique to their organization--for example, police reports--and plays an important role in the larger systems--for ex- ample, processing insurance claims. Police data have important organizational and political consequences, and improving the systems by which they are collected and processed re- mains a top priority for many agencies. The 1994 federal crime law pro- vided support for efforts to advance the information-processing capacities of local police, a capability that promises to become even more important in a post-September 11 world. Calls for service represent the most common form of data "input" to police agencies, and these calls are the source of most reported crimes. Pa- trol officer encounters with the public also involve an important element of information processing. In addition, the investigation of crimes also involves information processing (Eck, 1983b). Typically, investigators take state- ments from crime victims and witnesses. Officers routinely make judgments about the credibility of these individuals and process the information ac- cordingly. These judgments have been found to be particularly problematic

76 FAIRNESS AND EFFECTIVENESS IN POLICING with regard to the processing of alleged sexual assault cases (LaFree et al., 1983). Similarly, in the handling of complaints about police service, investi- gators receive and process information from complainants and witnesses, making judgments about the credibility of those individuals (Walker, 2001). Proactive investigations of such crimes as drug trafficking or other victim- less crimes often involve the use of informants, who are themselves fre- quently involved in criminal activity (Marx, 1988). Bolstered by the 1994 federal funds, important innovations of recent years involve police department efforts to engage in systematic data collec- tion and analysis, for purposes of both crime fighting and accountability. First among them is the technical ability to engage in sophisticated crime mapping, that is, the collection and analysis of official crime data for the purpose of identifying patterns in the distribution of crime and in particular for identifying hot spots with high rates of criminal activity that need spe- cial police attention (Sherman, Gartin, and Buerger, 1989; Weisburd, 2002). Modern management accountability systems are also built on the timely analysis of accurate small-area data about both crime and police opera- tions. The best known of these is New York City's CompStat process: a layered system of information and oversight that focuses patrol activity on clearly defined and strategically important problems, rigorously monitors data concerning the implementation of crime reduction tactics by local com- manders, and evaluates their impact on crime (Silverman, 1999; McDonald, 2001; Karmen, 2000). Although CompStat was originally developed by the New York City Police Department, it has been adopted by a number of other departments (Weisburd et al., 2003). Finally, police early warning systems involve the systematic collection and analysis of performance data on individual officers, particularly with regard to indicators of problems in dealing effectively with the public. Typi- cally, data are collected on use of force incidents, complaints, resisting ar- rest charges filed by officers, involvement in civil suits, use of sick leave, and so on. Early warning systems were originally developed for the purpose of identifying potential problem officers who would then be subject to some form of intervention (counseling or retraining) designed to correct their performance problems (Walker, Alpert, and Kenney, 2001). All of these developments illustrate the increasing importance of police collection of systematic data, the use of advanced computer technology, and the analysis of data for the purpose of identifying problems that need departmental attention. At the national level, the FBI and the Bureau of Alcohol, Tobacco and Firearms (BATF) maintain data bases that federal, state and local agencies can access for information about individuals. Per- haps the most important of these is the National Crime Information Center (NCIC), which was launched in 1967 (Manning, 1992). The NCIC main- tains data on criminal histories, fingerprints, mug shots, and the names of

THE NATURE OF POLICING IN THE UNITED STATES 77 people under correctional supervision. It includes a separate data base of missing and unidentified persons. Also, following enactment of the 1993 Brady Act, the FBI has maintained the National Instant Criminal Back- ground Check System (NICS), designed to identify persons who are ineli- gible to purchase a handgun. Investigations by the U.S. General Accounting Office (GAO) and con- gressional hearings (General Accounting Office, 1996) have brought sev- eral problems in these systems to light. Many local law enforcement agen- cies do not fully cooperate with the NCIC system, and certain data are missing, incomplete, inaccurate, or not entered in a timely fashion. The BATF system for recording serial numbers of guns seized in conjunction with police investigations is also extremely incomplete. Nevertheless, data collection by law enforcement is increasing in its scope and sophistication. Specialized Services Since the Progressive Era at the turn of the 20th century, professional police administration has emphasized the development of specialized units to carry out particular police responsibilities (Walker, 1977). The most com- mon specialized units are those related to criminal investigation, traffic en- forcement (Gardiner, 1969), juvenile delinquency (IACP, 1973), police com- munity relations (Law Enforcement Assistance Administration, 1973), crimes of vice, domestic violence (Sherman et al., 1992), gangs (Katz, 2001), and hate crimes (Garofalo and Martin, 1992). Specialized bomb and arson squads and SWAT units, or specially designated and trained officers are found in most law enforcement agencies. Some community policing pro- grams have been carried out by specialized units, while others are depart- ment-wide efforts (Rosenbaum 1994). The rationale for the formation of these units is both strategic and managerial. Specialization enables officers to focus their efforts on prob- lems important to their organization and to gain special knowledge and expertise. In this respect, specialized unit officers stand in contrast to the generalist function of patrol officers. The existence of a specialized unit also signals to the wider community that police are taking a particular problem seriously (Crank and Langworthy, 1992). SWAT teams are a good example. Prior to the late 1960s, police de- partments responded to special threat situations--such as those involving barricaded gunmen and hostages--on an ad hoc basis, simply handling the situation in whatever way seemed appropriate at the time. When several notorious incidents during the 1960s--one example is Charles Whitman's 1966 murderous sniping rampage in Austin, Texas--showed that the ad hoc approach was wanting in many regards, many police departments de- veloped cadres of specially trained and equipped officers who could handle

78 FAIRNESS AND EFFECTIVENESS IN POLICING crisis situations in a systematic fashion. Over the years, SWAT teams have handled a myriad of assignments besides the aforementioned hostage and barricaded subject situations. These tasks include dignitary protection, re- sponding to civil disturbances, stakeouts, and the service of search and ar- rest warrants that pose a greater than normal risk of injury to police officers and the public. Thus SWAT teams give police managers the ability to deal with a variety of problems that require specialized knowledge and equip- ment without having to go to the expense of training and equipping all of their officers (e.g., Mijares, McCarthy, and Perkins, 2000). At the same time, specialized units have often been beset by serious management problems. Overspecialization can result in inefficiencies by di- verting too many officers from other core functions (especially patrol) and the development of overly complex bureaucratic structures (Sparrow et al., 1990). Specialized units also create potential problems of control and ac- countability. Vice units, for example, have historically been the loci of some of the worst forms of corruption (Knapp, 1973; Rampart Independent Re- view Panel, 2000). Intelligence units engaged in illegal spying on legitimate political activities (Donner, 1990). Police-community relations units cre- ated in the 1960s were often detached from the mainstream of police opera- tions and deemed largely irrelevant to the problem they were created to solve (Law Enforcement Assistance Administration, 1973). Some current gang units have become detached from the official channels of supervision, thereby raising serious issues related to effectiveness or accountability (Katz, 2001). The dilemmas surrounding specialized units affect the implementation of community policing (Rosenbaum, 1994). As noted above, some com- munity policing programs involve specialized units, while others are de- partment-wide efforts. Supporters of the former note that assigning com- munity-oriented work to units that often are composed of volunteers and run outside the management structure of the department enables them to get started quickly and forestall resistance from mid-level managers dubi- ous about the concept and officers who do not want to participate. Others emphasize that making community policing a "specialized" function can marginalize it within the department, and it may signal that community policing is not "real policing" as it is defined by police culture (Skogan and Hartnett, 1997). The new concerns about terrorism and the appropriate response of state and local law enforcement agencies to that problem are likely to raise anew the issues surrounding specialized units. Police admin- istrators will face the dilemma of responding appropriately to public concerns without diverting excessive resources from other core police functions.

THE NATURE OF POLICING IN THE UNITED STATES 79 STAFFING THE POLICE The changing face of American policing is another important and fairly recent innovation. As of the mid-1960s, police officers were overwhelm- ingly white males from blue-collar backgrounds with little more than a high school education (President's Commission, 1967). Historically, policing was an all-male and virtually all-white occupation. Women were not employed until early in the 20th century, and even then in a severely limited role (Schulz, 1995). Racial and ethnic minorities were employed after the Civil War in some police departments, but until the 1960s only in a limited and largely token fashion (Dulaney, 1996). The dominant male and white com- position of police forces had a significant impact on the norms of the police subculture and on the perception of the police by racial and ethnic minority communities. Today, the rank and file includes a growing number of racial and ethnic minority group members and women. Levels of educational achievement have also risen substantially, with the typical recruit having about two years of college (Carter, Sapp, and Stephens, 1983). Employment of Women Through the early 20th century, American police departments employed women only as matrons who handled only female arrestees. Female officers were assigned almost exclusively to juvenile operations and, in fact, the advent of women officers often accompanied the creation of the first juve- nile units. This, in turn, was part of the specialization and development of complex police bureaucracies spurred by the professionalization movement in the early years of the 20th century. Female officers were not assigned to routine patrol but generally kept watch in areas where juveniles congre- gated. They did not wear traditional uniforms and did not carry weapons. The status of women in policing began to change in the early 1960s as a part of the larger women's rights movement, and barriers to the full em- ployment of women in policing finally began to fall in the late 1960s. The major turning point was the assignment of two female officers to routine patrol duty by the Indianapolis police department in 1968 (Schulz, 1995:131). This is generally believed to be the first such assignment in the history of American policing. Within a few years, virtually all departments eliminated formal barriers to employment and assignment, although infor- mal barriers remain. The advent of women as patrol officers provoked a national policy debate over their suitability for such duty. This debate prompted several studies of the effectiveness of women as patrol officers. The first and most influential was a comparative study of new male and female recruits in the Washington, DC, police department by the Police Foundation. This study

80 FAIRNESS AND EFFECTIVENESS IN POLICING and others that followed found that female patrol officers performed as effectively as male officers, with only slight differences in activity levels, for example in arrest practices (Bloch and Anderson, 1974). The research not only failed to confirm antifemale stereotypes that women were too physi- cally weak to perform effectively, but also did not confirm the profemale stereotypes that women would be more effective than male officers in deal- ing with the public and in particular defusing conflict (but see discussion below). There is considerable evidence of continuing resistance to women in police work, including a "glass ceiling" with respect to promotion. In 1998 only 7 percent of law enforcement agencies reported that more than 20 percent of their sworn officers were female, with no departments reporting as many as 30 percent female (National Center on Women and Policing, 1998). In contrast, an estimated 46 percent of the total U.S. workforce is female (U.S. Census Bureau, 1999). Affirmative action plans, including court-approved quota systems for recruitment, while they have increased the number of female officers in many departments (Martin, 1990), have not raised the level of female employment in any single agency to a level equal to the general labor force participation for women. In the nation's largest police agencies, which serve cities with populations of at least 250,000, women constitute 16.3 percent of the full-time sworn personnel. It is not known whether this continued underemployment is the result of a lack of recruiting effort on the part of police agencies or a lingering percep- tion on the part of potential female applicants that policing is a male occu- pation and that police departments are likely to be hostile workplaces. The advent of women had a significant impact on the norms of the police officer subculture. Academic research and popular accounts indicate that the very first female patrol officers encountered significant resistance from male officers. This resistance took the form of overt verbal hostility, attempts to embarrass or humiliate female officers, sexual harassment, and exclusion from choice assignments (Martin, 1980). Susan Martin's (1980: 102-108) research, however, found considerable variation in the attitudes of male officers. Those she characterized as "moderns" had little difficulty accepting female officers as equals (unlike the "traditionals" and to a lesser extent the "moderates"). This important finding suggested that traditional characterizations of the police subculture have been overly broad and do not take into account variations among even white male officers. With the advent of more female, black, Hispanic, Asian, and gay officers by the early 21st century, the officer rank and file has become steadily more heterogeneous.

THE NATURE OF POLICING IN THE UNITED STATES 81 Employment of Racial and Ethnic Minorities Historically, blacks have been significantly underrepresented in police departments in the United States. In the 19th century, departments outside the Southeast often hired a token number of black officers but confined them to patrolling black neighborhoods. Discriminatory patrol assignment patterns in such cities as Chicago and Los Angeles were not ended until the early 1960s. During the era of segregation, most police departments in the old South did not employ any black officers. Some cities hired a few black officers but assigned them to black neighborhoods only and did not permit them to arrest whites (Dulaney, 1996). The underemployment of black officers has been a major civil rights issue since the 1960s. The urban riots of that decade, for example, were explained in part by the lack of minority officers. The President's Crime Commission (1967) and the Kerner Commission (1968) found serious underrepresentation in big-city police departments. The civil rights move- ment and the 1964 Civil Rights Act ended de jure and other blatant but informal forms of employment discrimination in police departments in the Southeast. Insofar as covert employment discrimination persisted into the 1970s, departments in the South were essentially no different from depart- ments in other parts of the country in this regard. By the 1980s, some of the most racially and ethnically integrated police departments were found in the Southeast, notably in Atlanta, New Orleans, and Miami (Walker and Turner, 1992). One of the reasons that Southeastern departments were so integrated by the early 1980s was that the Sunbelt was booming (and able to hire officers to deal with growing populations) while much of the Rust Belt was broke and had not engaged in significant police hiring since before the important Equal Employment Opportunity legislation and litigation of the 1970s (Fyfe, 1983). Employment patterns outside the South followed an inconsistent pat- tern from the 1970s to the present. The generally accepted standard is that the racial composition of a police department should reflect the composi- tion of the local population (CALEA, 1994:31-2), defined as the agency's official jurisdiction and not, for example, the larger metropolitan area. Only a few police departments met this standard by the 1990s. While big-city police departments are more representative today than they were even 10 years ago, minorities represent 38 percent of the police force, a percentage that does not equal the minority presence in the communities that these police agencies serve (Brown and Langan, 2001). Underrepresentation con- tinues in most departments, and a few departments made little progress in minority employment. Civil rights leaders have demanded increased employment of black of- ficers in the belief that such officers will be better able to relate to black

82 FAIRNESS AND EFFECTIVENESS IN POLICING citizens and, consequently, that they will be less likely to use unjustified deadly force or excessive physical force and will be less likely to engage in discriminatory arrests (Kerner Commission, 1968). The social science evi- dence on police behavior does not support this view, however. Studies of police work have failed to find significant or consistent differences in the behavior of white and black officers. The use of deadly force, for example, is primarily a product of assignment, with officers assigned to high-crime areas more likely to fire their weapons (Fyfe, 1981). Official data on com- plaints by the public indicate that officers of different races and ethnicities receive complaints at rates equal to their presence in a police department (Walker, 2001). In contrast to black police officers, there is very little research on His- panic police officers (Carter, 1986). Yet this group made the largest gains in representation on big-city police forces throughout the 1990s; only 9.2 per- cent of large police departments were Hispanic in 1990, but by 2000, His- panics constituted 14.1 percent of the full-time sworn personnel in these agencies. The lack of research in this area is a function of the general lack of research on Hispanics in the criminal justice system compared with research on blacks (Walker, Spohn, and Delone, 2000). This is true even as employ- ment data indicate that Hispanics are generally underrepresented as police officers, in a pattern that parallels the underemployment of blacks. Excep- tions to this rule include a few cities where the local population is predomi- nantly Hispanic. Employment of Gay and Lesbian Officers A number of departments currently employ gay and lesbian officers without official prejudice. Some departments, particularly in cities with large homosexual communities, actively recruit gay and lesbian officers. In im- portant respects, this change in police employment practices represents an even greater transformation of the traditional police subculture than the employment of female officers as equals in the 1970s (Leinen, 1993). None of these innovations has been complete or comprehensive, however, in part because police organizations change slowly even under the best of circum- stances. The fragmented world of policing therefore prompts closer exami- nation of the process of innovation itself. RECENT INNOVATIONS IN POLICING The past three decades have witnessed a remarkable degree of innova- tion in policing. The innovation has been occurring at several different lev- els (Moore, Sparrow, and Spelman, 1996). At the operational level, police have experimented (both formally and informally) with new programs and

THE NATURE OF POLICING IN THE UNITED STATES 83 tactics for dealing with recurrent operational tasks. The police have, for example, experimented with the use of mandatory arrest policies to reduce domestic violence, with the use of decoys to reduce armed robbery, with the use of various tactics for disrupting street level drug markets, and with enforcement tactics focusing special attention on those adults and kids who seem to commit both frequent and serious criminal offenses, among others. They have also experimented with a variety of methods for managing calls for service to help ensure that their patrol force was being most effectively used both to reduce crime and to provide callers with something they could recognize as high-quality service. These operational innovations use exist- ing resources better to accomplish the important substantive purposes of the police. At an administrative level, the police have experimented with new ideas about how to structure and manage their organizations. They have sought to decentralize operations to give mid-level managers in the police depart- ment real responsibility--and to hold them accountable--for controlling crime in their areas, as well as to recognize and encourage the initiative and innovation of officers. As noted above, they have made special efforts to recruit members of the community who were not traditionally represented in police forces. They have also experimented with new methods of select- ing and training police officers and with new methods of identifying those officers who seem to be particularly prone to abuse. They have also experi- mented with different ways of making themselves accountable to the public for their use of force and authority and to more effectively control corrup- tion. And they have sought to find new ways of developing working part- nerships with other law enforcement agencies, with other civilian agencies in government, and with groups of various kinds. These administrative in- novations structure and lead departments differently while engaging in a wide variety of tasks. At a still different level, the police have experimented with new uses of technology. For example, in the domain of information technology and processing, they have made extensive investments in improved computer- aided dispatching and record-keeping systems. In the domain of offender identification, they have invested in automated fingerprint identification systems and developed capacities for DNA testing. In the domain of train- ing, they have developed computer-supported "shoot don't shoot" training programs to help officers practice making the split second decisions in- volved in the use of deadly force. In the domain of officer safety, they have developed lightweight, comfortable, bullet-proof vests. In the domain of offender safety, they have developed and deployed nonlethal weapons of various kinds. It is worth noting that these technological innovations can be further classified into those that support police programs and operations (e.g., DNA

84 FAIRNESS AND EFFECTIVENESS IN POLICING testing, nonlethal weapons, records systems that can support the unbiased targeting of frequent offenders) and those that support the administrative operations of the police department (e.g., the training of officers, the im- proved ability to report on the aggregate performance of the police to the public). These are likely to evaluated on entirely different sets of criteria. At another level, innovations can be described as strategic. A strategic innovation seeks to redefine and reprioritize the purposes that the police are trying to achieve, their most important methods for achieving the desired results, and the key working and reporting relationships that exist within the police on one hand, and between the police and other agencies of gov- ernment and the community on the other. Strategic innovations in policing may entail or require a great many other innovations of the programmatic, administrative, and technical type in order to be implemented successfully. But what distinguishes them as strategic innovations is the radical challenge they pose to traditional definitions of police purposes and operations. Two contemporary strategic innovations are community policing and problem- oriented policing. Community and Problem-Oriented Policing as Strategic Innovations The ideas of community policing and problem-oriented policing have emerged as among the most important innovations in policing over the last few decades. There has been uncertainty about whether these ideas repre- sent programmatic innovations, administrative innovations, or genuinely strategic innovations that seek to redefine, or at least prioritize, the ends of policing, to call for dramatically new means of accomplishing those ends, and to propose importantly new working relationships both within the po- lice department and between the police department and other government agencies and the communities in which they operate. To many, the idea of community policing, for example, seems to consist of certain kinds of pro- grammatic ideas (e.g., that the police should shift from patrolling in cars to patrolling on foot or on bicycles), or administrative ideas (e.g., that the police should create administrative positions with special responsibilities for liaisons with community groups) rather than a new strategic view of policing as a whole. Similarly, to many, the idea of problem-oriented polic- ing represents nothing more than a shift to a more proactive approach to crime prevention that draws much of its inspiration from the ideas of loca- tion-directed patrol and hot spot policing. To create room for the analysis that guides proactive policing, departments must introduce administrative changes that will either protect time for problem solving or create a special unit to do it. But this is an administrative strategy that is very familiar to the police--the creation of another specialized unit. While it is possible to view both community and problem-solving polic-

THE NATURE OF POLICING IN THE UNITED STATES 85 ing as small programmatic and administrative innovations that claimed a small portion of police resources and personnel to operate in a slightly new way, the supporters of community and problem-solving policing see these innovations in much larger and grander terms--as a set of ideas that can transform the structure, activities, operations, and even the culture of police departments. Those changes reflect a different mix of operational activities, a different set of administrative relationships, and (most importantly) a dif- ferent set of results measured in terms of the police to reduce crime, en- hance security, provide service, and enjoy legitimacy in the communities in which they operate. Community Policing Community policing, arguably the most important development in po- licing in the past quarter century, is described and presented by its advo- cates not as a programmatic innovation but as a strategic innovation. (Moore et al., 1992) It is characterized as something that transforms the "professional" model of policing, dominant since the end of World War II, to one that is "more focused, proactive, and community sensitive" and "por- tends significant changes in the social and formal organization of policing" (Greene, 2000:301). Community policing has been described as both a phi- losophy of policing and an organizational strategy. Greene (2000:301) de- scribes it as a "plastic" concept, because the range and complexity of pro- grams associated with it are large and continually evolving. When asked if they practice community policing, many agencies can point to a long list of projects as evidence that they are doing so. These activities range from bike and foot patrols to storefront offices and advisory committees. At root, however, community policing is not defined by a list of particular activities but rather as an organizational adaptation to a changing environment. Po- lice departments embracing the philosophy of practicing community polic- ing tend to adopt at least four new, interrelated organizational stances. Police Functions First, community policing departments tend to embrace a larger vision of the police function. Skogan and Hartnett (1997) describe this as "an expansion of the police mandate." Controlling serious crime remains the first priority of policing, and enforcing the criminal law remains the pri- mary and distinctive method of the police in accomplishing that important objective. But instead of seeing the police exclusively in these terms and viewing any activities that depart from direct efforts to control serious crime by threatening and making arrests of offenders as a distraction from the fundamental mission of the police and a waste of police resources, those

86 FAIRNESS AND EFFECTIVENESS IN POLICING who embrace community policing recognize that the police have other ad- ditional important functions to perform and other ways than making ar- rests of controlling crime and enhancing security. Indeed, they argue that the overall crime control and security enhancing effectiveness of the police might actually be strengthened by recognizing the importance of different police functions rather than focusing solely on reducing crime by threaten- ing and making arrests. An example of this is fear reduction. One of the important reasons the police should be interested in reducing crime is not only to reduce injuries and economic losses to victims, but also more generally to reduce fear. Indeed, we could easily cast reducing fear or enhancing safety and security as ideas that would be high on the list of the ultimate goals of policing. One means of accomplishing this, of course, is reducing criminal victimization. However, research reveals that the relationship between criminal victimiza- tion and fear is far from perfect and that there are other things that can be done to reduce fear. For example, one reason that advocates of community policing encouraged the police to take minor disorder offenses more seri- ously than they tended to do under the dominant professional model of policing was because these offenses were generating fear. They were con- cerned that fear was discouraging people from acting as though they owned the streets, thus undermining many forms of informal social control that could help to keep communities both safe and feeling safe. These observa- tions made the reduction of fear an important and somewhat different goal than the reduction of crime. New thinking about the ends of policing also includes the idea that police should provide quality service to the taxpayers who employ them. Traditionally, quality service meant fast, courteous response to calls for service. Now, for example, it has grown to encompass the role of police in providing information, practical advice, and counseling and referring call- ers to public and private agencies that are able to assist them further with their problems. Research summarized in Chapter 8 documents the wide range of positive benefits for the police themselves that this generates among victims and even suspects. Goals such as reducing fear and providing quality service can be seen as important in their own right, regardless of the effect of such activities on the crime reduction mission of the police. But these goals are also seen as im- portant ways for the police to enhance their ability to reduce crime, by promoting the reclamation of public spaces and the exercise of more effec- tive informal social control, as well as by fostering public cooperation, crime reporting, and willingness to step forward as witnesses. In short, the idea of community policing called for a wider engagement with the community in the interests of more effective crime control. Community policing also involves an expansion of the means of polic-

THE NATURE OF POLICING IN THE UNITED STATES 87 ing. The professional model of policing implied that police use of their en- forcement powers to threaten or make arrests of those who committed crimes ensured both a just result (offenders would be called to account for their crimes) and an effective result (offenders would either be discouraged from committing crimes or prevented from committing them via arrest, prosecution, and imprisonment). To threaten and produce arrests, the po- lice organized themselves to patrol the streets looking for crimes (random and directed patrol), to be able to hear and respond to emergency crime calls from witnesses and victims (rapid response to calls for service), and to find guilty offenders (criminal investigation). The first two activities are carried out by patrol units; the third is carried out primarily by detective units. To this, community policing introduces the idea that the police can enlarge and widen their efforts to prevent as well as respond to crimes. The ideas of prevention focused on at least two different concepts. One was intervening earlier with individuals who seemed on a trajectory to sus- tained, serious, criminal offending. Of course, much of the work of prevent- ing the development of sustained offenders lay with agencies outside the police, including family courts, child protection agencies, and schools. But the police often found themselves involved with young people who were headed in the wrong direction and sought to use their resources and their commitment to controlling crime to prevent this from happening. In the past, this idea of preventing crime by preventing the development of offend- ers justified such activities as Police Athletic Leagues and specialized juve- nile bureaus. More recently, preventive efforts focused on emergent offend- ers have included the DARE program, which sought to discourage drug use among children; special efforts to reduce violence in families as a crime problem in itself and as something that was likely to lead to the emergence of offenders in the future; the Serious Habitual Offender/Drug Involved (SHO/DI) Program which sought to focus police attention on serious youth- ful offenders; and various antigang initiatives that focused particular atten- tion on the recruitment of kids into gangs. Another idea focused less on offenders and more on the circumstances or situations that seemed to encourage or allow offenses to occur. Several scholars noted that crimes emerged from "routine activities" that could be understood in advance and disrupted. Others argued for a "situational ap- proach to crime prevention," which sought to alter conditions in ways that reduced the likelihood of an offense being committed. These situational approaches to crime prevention are often very similar to ideas generated by problem-solving policing. The reason is that problem solving urges a close analysis of the circumstances leading to offenses or other community prob- lems, as well as a search for an economical intervention that would solve the problem, not by arresting the offender, but by altering the social condi- tions so that there is less reason or opportunity to commit an offense.

88 FAIRNESS AND EFFECTIVENESS IN POLICING These two ideas about crime prevention shift the focus of the police from people who are already committing offenses to people who are not yet committing serious offenses, and to situations that tend to produce or allow criminal offenses to occur. Once a police department begins thinking about crime prevention of this type, they quickly discover that their ability to interrupt offender trajectories or alter situations often depends on agencies and actors outside the boundary of the police department. To deal with youth trajectories, the police have to deal with social work agencies, schools, parents, community groups, and even gangs. To alter criminogenic circum- stances, the police often have to work with community groups, with land- lords, and with merchants. In effect, the operational capacity they need to pursue these preventive ideas often lies outside the boundaries of the police organizations. Thus, an additional idea that comes with the idea of preven- tion is the idea that the police need partners and coproducers to fully ex- ploit the potential for crime prevention activities. This, in turn, means alter- ing the operational style of the police from less autonomous action to more cooperative action with other government agencies and groups of various kinds. It is these things that begin to transform the means of policing and the key working relationships within the departments and between the depart- ments, the other agencies of government, and the communities they try to police. And it is these ideas about which goals are important to achieve and how they might best be accomplished that lead to the three other principles of community policing. Decentralization Decentralization is an organizational strategy that is closely linked to the implementation of community policing in law enforcement agencies. Decentralization happens at two levels. First, more responsibility for identi- fying and responding to crime conditions and other community problems is delegated to mid-level precinct and district commanders in police depart- ments. Second, more responsibility for identifying and responding thought- fully to community problems is delegated to individual patrol officers, who are encouraged to take the initiative in finding ways to deal with the prob- lems of the communities in which they are operating. The consequence of this kind of decentralization is not only that the organization can become more proactive and more preventive, but also that it can respond to differ- ent sized problems. In the professional model of policing, the police are set up to respond to two different kinds of demands: calls from the public (which typically identify individual problems) and city-wide initiatives and programs that represent the initiative of the mayor or the city council. They cannot hear from residents formed in groups that are larger than individual

THE NATURE OF POLICING IN THE UNITED STATES 89 citizens, but smaller than the kind of groups that can command a city-wide agenda. The decentralization of initiative in police departments, paired with a commitment to consultation and engagement with local communities and conditions, allows the police to respond to problems that are important to specific communities. Community policing often leads departments to assign officers to fixed geographical areas and to keep them there during the course of their day. Usually they attempt to devolve authority and responsibility further down their agency's organizational hierarchy, to facilitate decision making that responds rapidly and effectively to local conditions, thus assisting the devel- opment of local solutions to neighborhood problems. It is also intended to encourage communication between officers and neighborhood residents and enables them to engage in community-oriented projects. Often there are moves to flatten the structure of the organization by compressing the rank structure, to shed layers of bureaucracy and speed communication and de- cision making. Community Engagement Another key feature of community policing is community engagement. Community policing encourages agencies to develop partnerships with com- munity groups, designed to help the police to better listen to the commu- nity, enhance constructive information sharing, and discuss with residents their concerns and priorities. To this end, departments hold community meetings and form advisory committees, establish store front offices, survey the public, and create information web sites. In some places, police share information with residents through educational programs or by enrolling them in citizen-police academies that give them in-depth knowledge of law enforcement. Civic engagement usually extends to involving the public in some way in efforts to enhance community safety. Residents are asked to assist the police by reporting crimes promptly when they occur and cooper- ating as witnesses. Community policing often promises to strengthen the capacity of communities to fight and prevent crime on their own. Residents sometimes get involved in the coordinated or collaborative projects, such as when they participate in crime prevention projects or walk in officially sanc- tioned neighborhood patrol groups. Even where these ideas are well estab- lished, moving them to center stage as part of a larger strategic plan show- cases the commitment of police departments to resident involvement. An important set of partnerships involves relationships with other gov- ernment organizations that have some direct responsibility for neighbor- hood quality of life. These agencies include those responsible for sanitation, housing inspection, street repair, and so forth. Following Wilson and Kelling's broken windows theory (1982), these partnerships are designed to

90 FAIRNESS AND EFFECTIVENESS IN POLICING complement and enhance traditional police activities (Eck and Spelman, 1987; Green, 1995; Braga et al., 1999). Problem Orientation Finally, as a predominant means of doing their work, community po- lice organizations typically try to shift away from traditional reactive patrol and investigative activities designed to threaten and arrest offenders toward the embrace of the kind of problem-solving methods of policing that are advocated by the concept of problem-oriented policing. As noted above, problem-oriented policing is an analytic method for developing crime re- duction strategies. The key difference between problem solving and com- munity policing is that the latter stresses civic engagement in identifying and prioritizing a broad range of neighborhood problems, while the former frequently focuses on patterns of traditionally defined crimes that are iden- tified on the basis of police information systems. As Eck (2003) notes, in problem-oriented policing, working with communities is a means to ad- dress problems, whereas in community policing, working with communi- ties is an end in itself insofar as it enhances police legitimacy and public support, improves other police services (e.g., investigating crimes), and im- proves community functioning (Eck and Spelman, 1987; Goldstein, 1990; Eck and Rosenbaum, 1994). Problem-oriented policing, by contrast, can be assigned to specialized teams that have little intersection with the public and range widely, rather than focusing on particular service areas (Eck and Spelman, 1987). Since community policing frequently involves a broad defi- nition of neighborhood problems, including many items that lie outside the traditional competence of the police, it can involve an expansion of the police role. In turn, this requires that police form partnerships with other public and private agencies that can join them in responding effectively to residents' priorities (Eck and Spelman, 1987; Green, 1995; Braga et al., 1999). It also requires extensive training, for traditional policing rarely equips officers for analyzing problems or crafting innovative solutions to them. Problem-Oriented Policing Problem-oriented policing is another important innovation gaining cur- rency in the United States. As indicated above, there is significant overlap between it and community policing, particularly when one is examining the ways in which these ideas challenge the prevailing means of policing. Both make a virtue of proactive as opposed to reactive responses to crime, disor- der, and other community problems. Both community and problem-ori- ented policing focus on preventing as well as reacting to crime. Both call for

THE NATURE OF POLICING IN THE UNITED STATES 91 solutions tailored to the particular problems and circumstances at hand rather than rely on more generalized techniques of patrol and investigation. Both community and problem-oriented policing view deterring and inca- pacitating offenders as only one of many different kinds of interventions that might be mounted by the police. Both community and problem-ori- ented policing contemplate the police organizing problem interventions that depend on cooperation with agencies and other actors beyond the bound- aries of the police organization. What community and problem-oriented policing have in common is an idea about how the police should respond to problems they have taken as their responsibility. The heart of problem-oriented policing is that this con- cept calls on police to analyze problems, which can include learning more about victims as well as offenders, and to consider carefully why they came together where they did. The interconnectedness of person, place, and seem- ingly unrelated events needs to be examined and documented. Then police are to craft responses that may go beyond traditional police practices. Solu- tions to problems might, for example, require the help of other city service agencies, or using the civil courts or the health department, or turning to residents to help them "take back the night." Today, well-organized de- partments identify and promote "best practices" that draw from previous experience in solving problems. Finally, problem-oriented policing calls for police to assess how well they are doing. Did it work? What worked, ex- actly? Did the project fail because they had the wrong idea, or did they have a good idea but fail to implement it properly? Together, these steps define what is known as "the SARA process." SARA is an acronym, developed by the Newport News Police Department and the Police Executive Research Forum, which stands for Scanning, Analy- sis, Response, and Assessment (Eck and Spelman, 1987). It is a policing philosophy that enjoys widespread popularity and has adherents through- out North America and Great Britain. Sherman and Eck (2002) reviewed problem-oriented policing evaluation findings and methods, including the recent rigorous evaluations of Braga et al. (1999) and Mazerolle et al. (2000). The evidence to date suggests that problem-oriented policing con- sistently shows effectiveness at preventing crime relative to common alter- natives. In the same volume, Eck (2002) points to a body of evaluations of place-based prevention schemes implemented by police agencies that also provide evidence for the effectiveness of problem-oriented policing. Thus, there is a growing body of empirical evidence supporting the assertion that problem-oriented policing can deliver what it promises. If community and problem-solving policing differ in any important way, it may have to do with how each regards the ends of policing and the key working relationships that must be established to ensure that the police can succeed in achieving their goals. Recall that community policing explicitly

92 FAIRNESS AND EFFECTIVENESS IN POLICING acknowledges that the important ends of policing include not only reducing crime, but also reducing fear, resolving disorder, and providing high-qual- ity services to callers. Recall also that community policing views the com- munity as a particularly important partner not only in nominating prob- lems that they consider important for the police to solve, but also in taking the actions that are necessary to resolve the problem. Problem-solving po- licing is less explicit about the important ends of policing and less focused on the importance of relationships with the community. As described by Eck (2003), problem-oriented policing is about putting problems first in all police decision making. Problems are defined as chronic conditions or clusters of events that have become the responsibility of the police, either because they have been reported to them, or they have been discovered by proactive police investigation, or because the problems have been found in an investigation of police records. In a seminal article, Herman Goldstein (1979) proposed this as an alternative to the "one inci- dent at a time" approach that characterized policing of the day. Goldstein raised the point that clusters of calls, often from the same address, might have a common cause. He reasoned that if police came to understand these clusters of calls--he dubbed them "problems"--they could reduce the vol- ume of future calls by resolving their common cause. In this model, policing would become problem-oriented rather than response-oriented. Subsequent research on 911 calls indicated that they were indeed heavily clustered. For example, a study in Minneapolis found that more than 50 percent of calls came from just 3.3 percent of the city's addresses (Sherman et al., 1989). It is important that the idea of problems that become the responsibility of the police include more than crimes. Of course, many of the problems for which the police become responsible are, in fact, crime problems; for ex- ample, a rash of armed robberies of convenience stores, the sudden appear- ance of street level drug dealers in a residential neighborhood, an upsurge in drunken assaults in bars. But Goldstein understood that because the re- sponsibilities of the police went well beyond the effective control of crime, the problems they would be expected to handle could also go beyond the issue of crime. They could be expected to find ways to reduce fear, control disorder, and provide certain kinds of service and assistance that would allow the public to use public space more securely and more easily than they otherwise could. So while the focus on problems seems to admit a wider set of purposes than the idea of crime control, problem-solving polic- ing does not usually take an explicit stance on whether the police could or should take responsibility for problems in the community that are not, strictly speaking, crime problems, but to which the police can make impor- tant contributions. Similarly, with respect to key working relationships, problem-solving policing acknowledges the potentially important role of the community in

THE NATURE OF POLICING IN THE UNITED STATES 93 nominating problems for police attention, but it does not emphasize this as the best or only basis to rely on to nominate or prioritize problems the police should try to solve. In problem-solving policing, the police are often encouraged to use their own sources of information and professional judg- ment to decide what is an important problem to solve. They are urged to examine their own records of calls to find the hot spots. Whether it is im- portant for the police to have their ideas about important community prob- lems vetted and discussed by community, so that the views of the commu- nity about what is important can be added to the views of the police themselves, is not much discussed or emphasized as a desirable characteris- tic of problem-solving policing. Similarly, while community and neighbor- hood groups are viewed by problem-solving policing as potentially valuable in dealing with the problems the police have identified as important (with or without community guidance), they tend to be viewed as less important and less reliable than other law enforcement agencies or other government agencies. In all these respects, problem-solving policing, to the extent that it differs from community policing, relies a bit more on police autonomy and professionalism and gives less influence over police operations to commu- nity groups than does community policing. In these respects, problem-ori- ented policing represents a more incremental step away from the profes- sional model of policing than does community policing. THE PROCESS OF ADAPTING TO CHANGE There is a powerful element of geographic consistency and temporal stability in policing. To a remarkable degree, the overall philosophy and strategy of the police (understood as both the important goals of the police as well as the primary theory of how they will operate to achieve their goal) has remained constant. While the initial mission of the public police (circa 1880) was to deal with any issue in society that would concern the public, for the past 50 years or so, the overriding goal of the police has narrowed to a sharp focus on reducing serious crime. The primary theory of how to accomplish that result has been to make (or threaten) an arrest. Arrests (both the threat and the reality) are deemed important not only because they produce a certain kind of justice (offenders can be called to account for their crimes), but also, more practically, because arrests can be expected to reduce crime through the mechanisms of general and specific deterrence, incapacitation, and (ideally) the rehabilitation of the offender while under state control. The organizational structure of police has also been remarkably consis- tent. Most police departments think of themselves as centralized in a para- military structure. They also have divided themselves functionally by divid- ing the patrol force from the investigative units; both of these operational

94 FAIRNESS AND EFFECTIVENESS IN POLICING units are separated from the administrative units that produce the infra- structure that prepares the police to act effectively operational, and to be accountable to the broader public for the way that police resources are used. Finally, the core operational procedures of the police have not much changed since the advent of the patrol car and the modern communications system: individuals call for service, a patrol car is dispatched, and a police officer handles the situation with a high degree of discretion (Mastrofski, 1990). Crimes that are not solved at the scene by the arrest of a suspect are referred to detectives for subsequent investigation. This basic process still accounts for the bulk of police contacts with the public and the majority of police department activities. Recognizing Innovation Despite this basic consistency, Bayley (1994) argues that the last two decades of the 20th century have witnessed the greatest period of change in all of police history. It is important to understand that this is an empirical claim that needs to be verified. In order to verify such a claim, we would have to have some way of recognizing important changes in the ways that police have thought about themselves, organized themselves, and (most importantly) acted. To a degree, we might use the discussion of the ways in which the police seem to have remained constant as a way of investigating the extent to which they have innovated. Viewed from this perspective, important innovations in policing would show up as some important changes in any of these relatively constant features: the conception of the overall mission and strategy of police (or at least, the relative importance of various elements of the police mission); the organizational structure of po- lice departments; or the core operational procedures. Of particular impor- tance would be large changes at the operational level, that is, the extent to which the police have expanded their range of activities beyond this core process (i.e., in the extent to which they have sequestered resources to en- gage in different activities), or in the particular ways in which the police are now performing these standard functions of patrolling, responding to calls for service, or investigating completed offenses. As noted above, the concepts of community and problem-oriented po- licing could be seen as ideas that call for innovation at the strategic level of policing. They seek to widen the focus of police attention to include minor offenses, disorder, fear, and other community problems beyond serious crime. They seek to change the structure of police departments by encour- aging operational initiative and creativity at lower levels of the organization and by encouraging the police to abandon their habitual autonomy and seek close working relationships with other government agencies and com- munity groups. They seek to create room for the police to step away from

THE NATURE OF POLICING IN THE UNITED STATES 95 the insistent demands of the 911 system so they have time to think about the important problems they were asked to solve and to find the means for solving the problems. These all mark a significant change in policing and justify Bayley's claim. But one of the most important things to understand about these ideas of community and problem-oriented policing is that they are not designed to make only a one-time change in the way police departments operate. The fundamental idea of both these ideas is that police departments should not only engage in this one important strategic innovation, but also that the point of each of these strategic innovations is to create police organizations that are continually innovative. Often when we think about the process of innovation, we imagine that some new policy or program has been tested and found to be effective. Consequently, that new policy or program ought to be incorporated into the operational routines of all police departments. We are satisfied when the innovation has diffused to all police departments. But the ideas of community and problem-oriented policing are not sim- ply programmatic ideas that are implemented and, once implemented, re- main constant. The ideas of community and problem-solving policing are that effective policing requires continuing innovation, not the consistent use of a single operational method. Continuing innovation is needed in part to deal with the highly varied and heterogeneous tasks that the police confront. Street-level drug dealing might be usefully controlled by regular "sweeps" of known drug markets. But sweeps are a bit hard to visualize and are not likely to work in dealing with either domestic violence or shoplifting. As Herman Goldstein reminded us, the police confront varied problems, and the problems require different solutions. The police have to be innovative to adapt successfully to the wide variety of problems they face. Continuing innovation may also be necessary to search for more effec- tive policies, programs, and tactics to deal with problems that loom large in the world of policing and are relatively homogeneous. For example, it is important for the police to find better ways of dealing with the big, con- stant problems they face, such as dealing with cold homicide investigations, or stopping strong-armed muggings, or taming street-level drug markets, or weakening and controlling criminally active youth gangs. So the police need innovation not only to deal with heterogeneous, unique problems, but also to learn how to deal with larger and more constant problems that continue to frustrate them. Indeed, one of the important findings of this committee has been that tailored approaches seem to be much more effective than more general en- forcement approaches in achieving crime control objectives. This means that if the police wish to be effective in controlling crime, they must find

96 FAIRNESS AND EFFECTIVENESS IN POLICING ways to make themselves into continuously innovative organizations, not just organizations that can adopt an important strategic, programmatic, or administrative innovation every now and then. There is not much police-specific research on any of the following sub- jects that would be important to understand: (1) what causes or allows organizations to embrace a strategic change, (2) what has caused or al- lowed police departments in particular to embrace a strategy of community or problem-oriented policing, or (3) what causes or allows the police to become the kind of continuously innovative organization that seems neces- sary for them to achieve their full potential. What we offer here, then, are some ideas about innovation in policing drawn partly from more general studies of innovation and partly from some specific examinations of inno- vation in policing. We also seek to identify points at which additional re- search would be particularly valuable. Serious limits exist in the diffusion of innovation literature on policing. First, there have been surprisingly few studies of the diffusion process among law enforcement agencies. This is an important research area for two rea- sons. First, economic analysis of private business has demonstrated that large firms adopt innovations more quickly but diffuse the innovation inter- nally at a much slower rate (Meisel and Levin, 1985). The extent of success- ful implementation is therefore presumably open to question even after an organization reports that it has "adopted" the innovation. This seems par- ticularly pertinent, since many diffusion studies use reported answers in the Law Enforcement Management and Administrative Survey (LEMAS) as their measure, yet Walker (bias crime) has shown that agencies occasionally misreport their status on adopting a certain policy or practice. Second, and perhaps most important, there is no systematic analysis, either in the policing or other literatures, of discontinuation of an innova- tion. Some researchers have speculated that both nonadoption and discon- tinuation are remarkably "unconscious" processes: that is, an organization does not engage in conscious review or decision making in either case. How- ever at least one researcher in the field of health care has speculated that discontinuation can be correlated to, among other things, the degree of agency involvement in the adoption of the innovation itself and the degree to which the innovation itself can be successfully integrated into established policies and procedures (Scheirer, 1990). While at least some attention has been paid to the institutional and sociological factors that inhibit or ob- struct innovation among law enforcement in general (Sandler and Mintz, 1974), more analysis of the actual process of discontinuation would iden- tify and highlight key levers in the adoption and implementation process.

THE NATURE OF POLICING IN THE UNITED STATES 97 Studying Innovation In studying innovation, one can focus on three somewhat different phe- nomena. One is the process that produces an idea that is wholly new to the world--the original innovation. This can happen inside an operating orga- nization as it tries to deal with a problem that has been frustrating it for a long time, or as it discovers a new and unexpected use of its capabilities. Or it can happen as the result of a planned research and development effort that is focused on coming up with either new products and activities for the organization or new processes for producing new products or delivering new services. A second process is the one that causes an innovation to diffuse across an organizational field; in effect, the process that allows an innovation cre- ated in a lab, or in one organization, to jump organizational boundaries and be embraced by other organizations. This can happen as a result of professional learning as each organization in the field, committed to achiev- ing its mission, discovers the superiority of some new method for accom- plishing its goals. It can happen as a result of a process that Powell and DiMaggio (1991) labeled "institutional mimesis" as each organization in a particular field tries to bolster its reputation and legitimacy by looking like other organizations or quickly adopting innovations that are thought to be at the cutting edge. It can happen as a result of legal action that makes organizations vulnerable to civil suits or government regulatory enforce- ment if they fail to embrace the new innovation. It can happen as a result of spending by the federal government. It can happen as a result of a political movement (often spearheaded by an advocacy organization of some kind) that can operate nationally (and even internationally) to press a new idea on a field. There are two important distinctions between these distinct processes of original organizational innovation on one hand, and the diffusion of that innovation across organizations on the other. The first is the difference between developing an idea that is new to the world versus copying that idea in other organizations. The second is the difference between the char- acteristics of a process that produces a new idea within a lab or within an organization, and the process that helps diffuse that idea across the organi- zational field. It is worth noting, however, that while a social analyst look- ing at the development and diffusion of an innovation might be interested in where the idea first came from and the mechanisms that spread the idea, those who are operating organizations and facing decisions about whether and how to adopt (or adapt) a new innovation look at the world very differ- ently. From their point of view, they may think that they are inventing a new idea, or at least adapting an idea for their particular purposes. Thus, they do not think of themselves as merely embracing a standard idea; they

98 FAIRNESS AND EFFECTIVENESS IN POLICING think of themselves as developing a new idea for themselves. They also face the problem of gaining acceptance for the idea in their own organization. This means that they have to innovate inside their own organizations in order to embrace the new idea, and that process may not be all that differ- ent from developing the idea from scratch in the first organization that developed the concept. One implication of these observations is that organizations might be considered more or less innovative based on three somewhat different ideas. First, how likely they are to come up with the wholly original new idea. Second, how likely they are to embrace ideas that are not "globally new" but are "locally new" (in the sense that their organization has not ever done this before). Third, how likely in embracing the idea are they to make sig- nificant changes in the idea in ways that improve the idea for future use in their own and other organizations. An organizational field could be consid- ered more or less innovative depending on the representation of these dif- ferent kinds of organizations in the field. The more creative originators, the more rapid adopters, and the more imaginative adapters in the organiza- tional field, the more innovative the organizational field is likely to be. Factors Shaping Innovativeness The broad decentralization of American policing creates both opportu- nities and problems for the development and dissemination of innovations. On one hand, precisely because the system is very decentralized, and be- cause, as a consequence, police organizations face somewhat different task environments and respond to different political aspirations, one might ex- pect there to be lots of variability in what police departments do, and that that variability would be a reflection of a high degree of innovation and variety in the field as a whole. On the other hand, precisely because the field is decentralized, it is difficult to find a central point of leverage to move the field. There is not enough coherent authority over the field, nor enough centralized control of funding to be able to push a specific model of policing through the field as a whole. Given these observations, it is surprising, then, to see how much consis- tency exists in the field of policing, how stable those patterns have been over time, and how, when strategies, programs, or administrative systems change in these organizations, they seem to change very rapidly and all at once. Apparently, there are some forces acting on thousands of local police departments that push them toward conformity with some kind of profes- sional norm. Furthermore, these pressures produce a high degree of inertia most of the time. But finally, when these norms change, they seem to change widely and fairly quickly--say within a decade.

THE NATURE OF POLICING IN THE UNITED STATES 99 What these forces might be that produce consistency and stability within the field in some periods, and rapid change in the embrace of new strategic, programmatic, technological, or technological innovations on the other are hard to discern for the field of policing. Recently, police researchers have differed in their theories for explaining the factors that lead to innovation within individual police departments and across the field: Zhao (1996) ar- gues that environmental factors stimulate innovations. Mullen (1996), sees internal, agency-specific decisions and factors as key. Given that there is as much consistency as there is in the field, and that the field seems to change all at once, one would be tempted to conclude that it must be environmen- tal factors operating across the nation that cause the changes to occur. Yet one could save the idea that internal, agency-specific decisions and factors were the key if one considered as an important environmental factor the number of other police departments that seemed to be embracing or resist- ing a new innovation. If the individual choices of departments to embrace or reject an innovation are based in part on what the other organizations around them or known to them have done, then one could conclude that the distinction between the environmental hypothesis on one hand and the organizational decision-making process on the other were not so different. Indeed, one could say that the way the environmental factors work on po- licing is only through the choices made by organizational leaders and their skill in leading their organizations, and what makes it seem as if organiza- tional factors are producing the results is that organizational leaders are all acting similarly in response to the changed environment. As King (1998) points out, many researchers have formally or infor- mally studied the adoption of a single innovation, from community policing (Zhao, 1996), to police technology (Seaskate, Inc., 1998), to even police uniforms (Monkonnen, 1981). However, only a smaller body of literature is devoted to the process of diffusion itself (Weiss, 1998; King, 1998; Weis- burd, 2002). The lack of convergence among these studies (King, 1998) reflects the instability of the larger diffusion literature (Damanpour, 1991) and limits the ability of the committee to draw comprehensive conclusions that are sure to be useful to policy makers. Drivers of Innovation in Policing There needs to be more research on the sources of innovation in polic- ing. Of particular importance would be findings about particular institu- tions and processes that give leverage on shaping the field of policing. We summarize this literature in terms of particular institutions and processes that have been investigated as important drivers of change in policing. 1. Innovation by Edict: Supreme Court Decisions and Civil Suits. Courts are obviously important drivers of change in the field of policing. Court

100 FAIRNESS AND EFFECTIVENESS IN POLICING decisions can invalidate existing practices and define new minimum legal standards (e.g., the Miranda and Garner decisions). In many instances, court decisions defining minimum constitutional standards compel police depart- ments to undertake various collateral changes (e.g., changes in recruitment standards to comply with equal employment opportunity requirements). More recently, consent decrees arising from Department of Justice suits over a pattern or practice of abuse have imposed major organizational re- forms on some police departments. These consent decrees have, in effect, ratified and promoted a short list of best practices related to accountability and integrity (Klockars et al., 2000). 2. Innovation Through Research. As noted, academic research has been another important source of innovation. The creation of the Police Founda- tion in 1970, with funding from the Ford Foundation, introduced a new era of research and demonstration projects in policing. Published research on innovations has questioned many traditional assumptions about police work and validated and popularized certain innovations. The Kansas City Pre- ventive Patrol Experiment (Kelling et al., 1974), for example, played a ma- jor role in undermining the traditional assumptions about preventive patrol and forced the search for alternative modes of service that culminated in the development of community policing and problem-oriented policing. The Minneapolis Domestic Violence Experiment (Sherman, 1992) and the con- trols over police use of deadly force (Milton et al., 1977; Fyfe, 1979) were particularly influential in the development of new police policies. There is considerable concern in the field, however, about basing re- forms on only one or perhaps a handful of scientific studies (Lempert, 1984, 1989; Sherman, 1992). Compared with other areas in which science and public policy intersect--such as the health impact of tobacco or global warming--and there are literally thousands of studies, the scientific litera- ture on major police issues is scant. 3. Innovation as a Confluence of Problems and Solutions. One of the reasons that innovations of the programmatic and technological means tend to take off quickly in policing before adequate research has been completed is that one of the important drivers of innovation in policing are those occasions on which research findings, police practices, and technology sud- denly come together in a powerful way. A detailed case study by Weisburd illustrates the point. After years of results that demonstrated the ineffective- ness of core policing practices, such as generalized patrol and decreased response time, research began to focus on the spatial concentration of crime and police. Some who are aware of this research, and some who are not, initiated hot-spot policing, a strategy designed to target specific crime areas. Crime-mapping software, a highly refined version of older police pin maps, demonstrated its usefulness in geographically tracking crime trends at the

THE NATURE OF POLICING IN THE UNITED STATES 101 very time when hot-spot policing seemed to promise the best results (Weis- burd, 2002). Thus the coincidence of strategy with technology led to the diffusion of crime-mapping. It seems likely that the cues to innovate differ based on the type of innovation, whether technological, as was the case in crime-map- ping, administrative, or as a matter of policy. Finally, the perceived advan- tage of the innovation over and above the practice or item that it seeks to replace--what economists call the relative advantage--plays an important role behind the selection of any given innovation. 4. Innovation Stimulated by Local Governance. It is important to note a distinctive difference between private-sector innovation and that in the policing domain: the relative openness or permeability of police agencies to change is a function of the democratic structure of the governance of polic- ing. In private organizations, profit or competitive advantage motivates most innovations. In policing, the external social and political environment penetrates police organizations and leads to change. The professionalization movement at the turn of the 20th century, for example, was part of the larger Progressive Era reform ethos that sought to make government agen- cies more efficient and at the same time promote social justice. The civil rights movement of the 1960s stimulated a wide range of changes designed to achieve greater racial equality in policing, through greater employment of minority officers, the creation of specialized police-community relations units and civilian oversight agencies, and the development of policies de- signed to curb police misconduct. As discussed earlier, the changing compo- sition of police forces has occurred in response to larger trends related to the employment of women and racial and ethnic minorities. The commu- nity policing movement was driven by increased public demands for both more effective crime control and more positive relations between police and the public (Walker, 1977, 1998). 5. Innovation Through Professional, Accrediting, and Auditing Organi- zations. A third potentially important driver of change in policing are the influences that are brought to bear by professional organizations such as the International Association of Chiefs of Police (IACP), the Police Execu- tive Research Forum (PERF), the National Sheriffs Association (NSA), and others. These organizations are important influences on innovation through two mechanisms. On one hand, they convene police professionals and en- courage open-ended discussion through formal and informal means. This creates a sort of free market in ideas and a chance for the widely decentral- ized police field to come together to discuss their work. On the other, these organizations sometimes work to develop and disseminate new ideas and issue recommended policies on various issues in a manner similar to profes- sional associations in other fields.

102 FAIRNESS AND EFFECTIVENESS IN POLICING Accreditation is another source of change, if not innovation. The Com- mission on Accreditation for Law Enforcement Agencies (CALEA) was established in 1979 and published its first set of standards in 1983. The process of becoming accredited often forces departments to undertake sig- nificant organizational change in order to comply with required standards. The CALEA standards tend to codify innovations that have gained recogni- tion as important minimum requirements. However, accreditation is en- tirely voluntary, and only a few hundred agencies have been accredited to date. Similarly, auditing agencies, often called in by mayors or city councils to review the efficiency and effectiveness of police organizations, have a powerful effect on innovations in the field. To be able to audit an organiza- tion on some objective basis, there has to some agreement about the stan- dards to be used. Those standards, in turn, tend to be set by past traditions and understandings rather than emergent ideas. As a result, many innova- tive ideas in policing are viewed by conventional auditing agencies as prob- lems rather than advantages for police departments, and it takes a while for new ideas about what is valuable, efficient, and effective to work their way into existing audit standards. 6. Innovation as a Process of Social Learning. Closely related to the idea that innovation might be spread by professional associations is the idea that innovation is spread by a more generalized, informal process of social learning that animated, led, and legitimated by the action of particular or- ganizations that are known to be particularly innovative. The importance of these particular organizations is suggested by research that shows that the process by which an innovation is spread is far from random across organizations. Indeed, organizational research has found that if one were to track the rate of any successfully adopted innovation, it would follow an S- curve pattern. The early adopters grouped at the lower end initiate the pat- tern and seem to set the stage for the adoption by others. The process of diffusion continues more rapidly in a distinctive spike in the rate of adop- tion and then settles into a plateau, indicating that adoption has spread to, and through, a community of potential adopters and is now at equilibrium. Importantly, the early, adopters are more "cosmopolitan" than their peers (Rogers, 1983)--that is, they participate in forums that mediate the ex- change of ideas. Studies have confirmed this in the policing field (Moore, Spelman, and Young, 1992; Weiss, 2001; Weisburd, 2002). One finding that has emerged from research is that police agencies, left to their own devices, follow a pattern of social learning (Kapur, 1995) in deciding whether or when to adopt an innovation (Weiss, 2001, 1998). Social learning describes an unstructured but effective process whereby an organization, in this case a police agency, judges whether to adopt an inno-

THE NATURE OF POLICING IN THE UNITED STATES 103 vation based on the experience of a similarly situated police department. In most medium or large-size agencies, police planners function as the con- duits of social learning: they identify, contact, and communicate the experi- ences of other departments to their own (Klockars and Harver, 1993). Weiss (1997) conducted a survey of 400 state and local police planners and found three key factors that determine which agencies police planners contact: the agency is in the same state or region; the agency faces the same problems and issues; and, finally, the agency has a good reputation. The first two criteria might be thought of as different kinds of peer agencies--regional or situational--whereas the last judgment seems to be based on subjective po- lice professional knowledge. This "good reputation" category may also in part be determined, in the minds of practitioners, by selecting policing agen- cies that hosted well-publicized research projects germane to the innovation in question (Weiss, 2001). Since early adoption among cosmopolitan agencies sets off the diffu- sion process, identification of these agencies becomes a priority for those who wish to spread desirable practices or technology. In drawing these distinctions among police agencies, Weiss's finding that networks of com- munication vary based on the type of innovation must be considered. For instance, Weiss found that on administrative topics, police planners con- tacted their "regional dyad or clique," yet on substantive strategies or is- sues, such as domestic violence or community policing, police agencies con- tacted the cosmopolitan "opinion leaders" (Weiss, 2001:12). Again, it is significant that publicized research projects played a role in the identifica- tion of this latter group. 7. Innovations Stimulated by the Federal Government. Of particular current interest is the role played by the federal government in encouraging innovation in policing. Much less is known of the role of federal executive branch agencies in stimulating innovation in policing. In an earlier period, special blue ribbon commissions, such as the President's Crime Commis- sion in the 1960s (President's Commission, 1967), promulgated new ideas and minimum standards (Walker, 1985). The federal government continues to sponsor research and demonstration projects through the National Insti- tute of Justice and the Bureau of Justice Assistance, but only a few follow- up studies have been made of their impact on police practices generally. A rare exception to this rule was a study carried out by the Urban Institute that evaluated the impact of the federal Community Oriented Policing Ser- vices (COPS) program on the spread of community and problem-solving policing across the field of policing. This study is sufficiently important in its own right as a source of information about how much the field of polic- ing has changed, and the ways in which the federal government can make change, that it is worth exploring in some detail.

104 FAIRNESS AND EFFECTIVENESS IN POLICING Adoption of Community Policing: A Case Study The adoption of community policing over the past 20 years represents an important case study of the change process in policing. Hickman and Reaves (2001) report that the development of strategic plans incorporating community policing is common in most departments, especially among those serving communities over 50,000. By 1999, local departments em- ploying about 80 percent of the nation's police officers had adopted some form of geographical responsibility for patrol personnel, and about half worked for agencies that were giving detectives turf-based assignments. In addition, by the same time, half of all officers worked for agencies engaging in problem-oriented policing in a systematic way, and one-third of these incorporated problem-oriented policing successes in performance evalua- tions. Most people lived in places where police had formalized problem- oriented policing partnerships with other agencies and groups, virtually all (96 percent) lived in places where police reported meeting regularly with residents, and over 90 percent of Americans live in cities providing residents with routine access to crime statistics. Other estimates are available of the shape of American policing in the wake of the Violent Crime Control and Law Enforcement Act of 1994. The Law Enforcement Management and Administrative Statistics program con- ducted by the Bureau of Justice Statistics found that, by 1999, almost two- thirds of local police departments reported they had officers serving in full-time community policing roles. More than 91,000 officers reportedly were serving in that capacity, or 21 percent of all sworn personnel (Hickman and Reaves, 2001). In terms of agencies, the Urban Institute's evaluation found that about 55 percent of the 19,175 law enforcement agencies that were eligible for federal grants (which includes many agencies that are not municipal police departments) requested and received at least one by the end of 1999 (Roth et al., 2000). Hickman and Reaves (2001) found that 34 percent of local police departments reported having officers with full-time assignments to community policing in 1997, a figure that rose to 64 percent by 1999. Full- time assignments to community policing were more common with increas- ing city size in 1999, ranging from 54 percent in towns under 2,500 to 100 percent in cities of 1 million or more. However, between 1997 and 1999, there were increases in the proportion of cities reporting assigning officers to community-oriented jobs in every size category. The largest percentage of new adoptions of community policing could be found in cities below 50,000 in population, and smaller towns were among those committing the largest fraction of their officers to these assignments. What do agencies practice under community policing that is new? Zhao, Lovrich, and Thurman (1999) examined its adoption in cities of 25,000

THE NATURE OF POLICING IN THE UNITED STATES 105 and larger by surveying police agencies. They found that by 1996, 90 per- cent or more reported adopting activities implementing the kinds of core organizational strategies reviewed earlier: decentralization, community en- gagement, and problem solving. Roth et al. (2000) report substantial in- creases in partnership building by police between the mid-1990s and 1998. This included holding regular community meetings, surveying the public, and forming advisory committees. Over this same period, agencies were more likely to adopt new mission statements as well. Federal Role in Community Policing The federal government has played a large role in promoting commu- nity policing. Federal agencies funded demonstration programs and research projects that explored its possibilities, and they promoted it in a series of national conferences and through their publication channels. The Violent Crime Control and Law Enforcement Act of 1994 funded new positions for officers with community assignments, supported community policing train- ing academies in every region of the country, and facilitated the purchase and utilization of new technologies that promised to support neighborhood- oriented policing and return sworn officers from desk jobs to the street. These responsibilities were assigned to a new agency in the Justice Depart- ment, the Office of Community Oriented Policing Services. Federal funding did increase the number of officers with community- oriented assignments, although the magnitude of its effect is uncertain. An evaluation of the 1994 act by the Urban Institute concluded that, by the end of 1998, it had led to a net increase of 36,000 to 37,000 police officers. By the end of 1999 that number had risen to almost 61,000, and the federal government had awarded a total of about $4.3 billion for officer hiring. Many agencies also applied for technology support grants from the COPS Office. Their stated goal was to release sworn officers from office jobs by acquiring efficiency-enhancing equipment and hiring civilian technical staff. By the end of 1999, just over $1 billion had been allocated through this part of the program, resulting in claims of a projected yield of about 40,000 officer full-time equivalents for community policing (Roth et al., 2000). Community policing requires more than hiring and assigning officers; police officers must also acquire new skills and perspectives. Major new innovations such as community policing highlight the importance of train- ing for officers. The 1994 crime law also funded the creation of regional community policing centers around the country. By 1999, 88 percent of all new recruits and 85 percent of serving officers worked in departments that were providing some community policing training. Smaller agencies with fewer officers were less likely to provide specialized training, especially in cities of less than 50,000. Because those agencies are numerous, in 1999

106 FAIRNESS AND EFFECTIVENESS IN POLICING only about 41 percent of all departments reported providing community policing training for all of their new recruits, and only 28 percent provided it for all of their serving officers (Hickman and Reaves, 2001). There are a number of unanswered questions about the adoption and staffing of community policing programs around the country. Whether po- lice departments will continue to expand and staff these projects in the absence of federal support remains an open question, as is whether local governments will continue to invest in new technologies for policing. Even where there is such a commitment, implementing an effective program also can prove difficult, for adopting community policing can involve significant structural changes in departments and a reorientation of their basic mis- sion, as well as asking officers and their supervisors to think and act in new and unaccustomed ways. Roth et al. (1999) and others have observed that many agencies continue to define community policing only by the list of projects or programs that they have under way. In an earlier study, Wycoff and Skogan (1993) found that only about one-third of departments thought that structural changes were necessary to do community policing. Assessments of the adoption of policing innovations of all kinds are limited by the inadequate measurement of the structure and functioning of American departments. All of the studies cited above made extensive use of data from questionnaires distributed to police departments by mail or con- ducted by telephone. As such, a single informant who completed the survey served to represent events and processes affecting an entire organization. Studies of the reliability and validity of such studies indicate that the results can be highly contingent on who is given responsibility for representing the agency and where they are positioned in the organization. For example, multiple informants answering questions about the same police department have been demonstrated to disagree significantly on such questions as whether and when their agency adopted a policy on domestic violence and about their stock of equipment (Weiss, 1997). Finally, many are uncertain whether community policing can live up to its promises. Like many new programs, its adoption in many instances pre- ceded careful evaluation of its consequences. The effectiveness of commu- nity policing has been the subject of some research, ranging from its impact on crime to how openly it is embraced by the officers charged with carrying it out. The findings of this research are evaluated along with other policing strategies in Chapter 6. RESEARCH RECOMMENDATIONS Policing is a highly complex enterprise in terms of the structure and governance of police agencies, the roles and responsibilities of the police, and the factors that influence policing. Because of this complexity, policy

THE NATURE OF POLICING IN THE UNITED STATES 107 makers should not expect dramatic changes in policing from any single policy or innovation. Given the decentralized structure of policing, whereby control is exercised by local units of government, there is no central con- trolling authority capable of implementing any single policy change. Nor would a policy change in one area of police responsibilities necessarily af- fect performance in other areas. In a time of rapid innovation in the fragmented and complex world of the policing industry, it is invaluable to monitor the pulse of policing in the United States. The committee recommends that the Bureau of Justice Statis- tics continue to conduct an enhanced, yearly version of its current Law Enforcement Management and Administrative Statistics (LEMAS) survey. Ongoing support for the survey, which began in 1987, has recently been supplemented by the Office of Community Oriented Policing Services, and the committee recommends that this extensive and now yearly survey be continued. The research utility of the survey would be enhanced by ensur- ing that a panel of consistently surveyed agencies be maintained within the framework of the survey sample. The committee notes that other re- search projects have come to rely on the results of the BJS Agency Directory Survey of the Bureau of Justice Statistics, which forms the basis for the LEMAS sample, and recommends that they improve and update its cover- age on a regular basis. After reviewing methodological research on agency surveys like LEMAS, the committee recommends that the Bureau of Justice Statistics conduct a special study of the validity of responses to the survey and experiment with methods to ensure accurate reporting of agency characteristics. Although several federal agencies are charged with encouraging inno- vation in law enforcement, the committee finds that little is known about the innovation process or how it can be facilitated. The committee recom- mends a special study of innovation processes in policing, one that is designed to include factors that can be influenced by federal and state governments. There appears to have been tremendous growth in the private security sector in America, but the committee was unable to identify any reliable indicators of its size or operating characteristics. The committee recom- mends a special study of the dimensions of the private security industry, and that the Current Population Survey be used to secure an estimate of the size and characteristics of the labor force in this sector. The committee recommends the launching of a periodic national sur- vey to gauge the extent and nature of police-public contacts and public assessments of the quality of police service in their community. Method- ological testing should be conducted to maximize the accuracy with which public encounters with the police can be recalled and profiles of those en- counters can be drawn. The survey should be large enough to account for

108 FAIRNESS AND EFFECTIVENESS IN POLICING both traffic and other kinds of encounters, including pedestrian stops. The survey should include measures of subjective assessments of the quality of those encounters. In addition, the survey should be the vehicle for gathering attitudinal data from a general sample of adults (including young adults) concerning the performance of police serving their immediate communities.

Next: 4 Explaining Police Behavior: People and Situations »
Fairness and Effectiveness in Policing: The Evidence Get This Book
×
Buy Paperback | $54.95 Buy Ebook | $43.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Because police are the most visible face of government power for most citizens, they are expected to deal effectively with crime and disorder and to be impartial. Producing justice through the fair, and restrained use of their authority. The standards by which the public judges police success have become more exacting and challenging.

Fairness and Effectiveness in Policing explores police work in the new century. It replaces myths with research findings and provides recommendations for updated policy and practices to guide it. The book provides answers to the most basic questions: What do police do? It reviews how police work is organized, explores the expanding responsibilities of police, examines the increasing diversity among police employees, and discusses the complex interactions between officers and citizens. It also addresses such topics as community policing, use of force, racial profiling, and evaluates the success of common police techniques, such as focusing on crime “hot spots.” It goes on to look at the issue of legitimacy—how the public gets information about police work, and how police are viewed by different groups, and how police can gain community trust.

Fairness and Effectiveness in Policing will be important to anyone concerned about police work: policy makers, administrators, educators, police supervisors and officers, journalists, and interested citizens.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!