History
Starting in the late 1990s, law enforcement in the United States faced allegations that they were ”racially profiling.” The charge was that police were targeting drivers who were black or Hispanic for vehicle stops based on beliefs about the propensities of these groups to commit crime.
There are two ways to conceptualize the history of racial profiling—the short-term view and the long-term view. The short-term view extends back to the 1990s and focuses on police bias manifested in stops and searches of drivers. Two lawsuits were key to these allegations. In New Jersey v. Soto, a case brought in 1990, public defenders representing Pedro Soto and others sought to suppress evidence obtained during searches conducted by New Jersey troopers, alleging the searches were the result of racial profiling practices. These public defenders had detected an alarming number of cases involving black people who were stopped by troopers on the New Jersey Turnpike. The defendants obtained data on New Jersey State Police traffic stops in various areas from 1988 through 1991 and hired a social scientist to analyze the data. The court relied heavily on the results of the analysis of those data—which indicated an overrepresentation of blacks among people stopped—in deciding to suppress the evidence. Wilkins v. Maryland was brought in 1992 by a Harvard-educated public defender who was detained with his family by the side of an interstate highway by the Maryland State Police. After refusing the troopers’ request to search, the family was required to stand in the rain while a drug-sniffing dog checked the car, finding nothing. The American Civil Liberties Union filed a suit on behalf of Wilkins and his family, alleging that the police detention was the result of racial profiling. The case was eventually settled. The settlement included a mandate for the Maryland State Police to collect data on traffic stops. These cases sparked additional stories and lawsuits that police were engaging in racial profiling and the issue became one of the most critical facing law enforcement in the late 1990s.
In tracing the origin of racial profiling in the context of the short-term view, some law enforcement practitioners and other experts point to the use of drug courier profiles developed during the war on drugs and, in particular, training conducted by the Drug Enforcement Administration (DEA) during the 1980s. While DEA officials claim that race was never part of the training on profiles, other observers and some local/state law enforcement practitioners report that local and state police came to link blacks and Hispanics with criminal drug activity as a result of exposure to this training.
The long-term view goes far back into the history of this country and also conceptualizes racial profiling more broadly— that is, the broader view would encompass all law enforcement decisions by police, not just decisions to stop and search drivers. In this expanded, longer term view, racial profiling is just a new term for longstanding allegations that police (and other components of the criminal justice system) treat minorities differently because of their race and ethnicity. These charges were very much a part of the civil rights movement and, indeed, many of the riots occurring during the volatile years of this movement were precipitated by incidents in which officers were alleged to have misused force or other authority against minority subjects.
Terminology and Definitions
In addition to various histories, various terms and definitions are used to characterize this critical issue facing police and their communities (see Batton and Kadleck 2004). The most frequently used definition in the late 1990s included the word solely. Pursuant to this definition, police are prohibited from taking law enforcement actions (for example, stopping, questioning, arresting, searching) based solely on a person’s race or ethnicity.
The popularity of this definition has waned because stakeholders have recognized that it references a small (arguably very small) subgroup of incidents that might reflect bias on the part of police. Even a racially prejudiced officer likely uses more than the single factor of race when conducting biased law enforcement. For example, officers might make biased decisions based on the neighborhood and the race of the person, the age and the race of the person, or the gender and the race of the person. Activities based on these sample pairs of factors would fall outside this narrow definition of racial profiling. Moreover, one could interpret this common definition of racial profiling to exclude activities that are legally supportable in terms of reasonable suspicion or probable cause, but are nonetheless racially biased. Because the traditional definition only prohibits actions based solely on race, it does not encompass decisions based on reasonable suspicion or probable cause plus race. That is, this definition could be interpreted to exclude, for instance, officers’ pulling over black traffic violators and not white because of race, or citing Hispanic, but not white, youth for noise violations because of race/ethnicity. Such disparate treatment would not necessarily be encompassed by a definition that referred to actions based “solely” on race, because the officers would have acted on the basis of reasonable suspicion or probable cause, as well as race.
Over time definitions have expanded and new terms have been introduced to reflect these expanded definitions. For instance, the Police Executive Research Forum (PERF; Fridell et al. 2001) introduced the term racially biased policing and defined it to encompass all incidents in which race/ethnicity is used inappropriately by officers in deciding with whom and how to intervene in a law enforcement capacity. Terms such as bias-based policing and race-based policing were also introduced, and definitions of the various terms included the following, which maintain that the negative behavior in question occurs when police use ”race as a key factor in deciding whether to make a traffic stop” (Langan et al. 2001, 20). Another definition says ”. . . police routinely use race as a negative signal that, along with an accumulation of other signals, causes an officer to react with suspicion” (Kennedy 1999, 11).
Police rely ”on the race, ethnicity or national origin rather than the behavior of an individual or information that leads the police to a particular individual who has been identified as being, or having been, engaged in criminal activity” (Ramirez, McDevitt, and Farrell 2000, 3).
Agency Policies
When agencies formulate their policy on racial profiling, they adopt a definition so that they can tell officers when they can and cannot use race/ethnicity to make law enforcement decisions. There are significant differences of opinions regarding this issue and these views are reflected in the various policies that have been adopted around the nation to address racial profiling. While virtually no one believes officers should be allowed to use race/ethnicity as a sole basis to justify law enforcement action, beyond that consensus there are differences of opinion regarding the circumstances under which it is acceptable to use race and ethnicity as one factor among several to justify law enforcement action.
Many of the early policies (some still in effect) are based on the solely definition. These policies prohibit officers from using race/ethnicity as the sole basis for law enforcement actions (for example, make a stop, conduct a search, request for consent to search). For the reasons explained earlier, this policy does not place significant parameters on police use of race/ethnicity. The most restrictive policy model is commonly referred to as the ”suspect-specific policy.” Suspect-specific policies generally read as follows: Officers may not consider the race or ethnicity of a person in the course ofany law enforcement action unless the officer is seeking to detain, apprehend, or otherwise be on the lookout for a specific suspect sought in connection with a specific crime who has been identified or described in part by race or ethnicity.
The key to this model is that the set of identifiers—which includes reference to race/ethnicity—must be linked to a particular suspect who is being sought for a particular crime. Thus, if reliable witnesses describe a convenience store robber as five feet, eight inches tall, lean, long haired, and Asian, Asian can be considered along with the other demographics and with other evidence in developing reasonable suspicion to detain or probable cause to arrest.
The law enforcement field has received very limited guidance from the U.S. Supreme Court regarding when race can and cannot be used to make law enforcement decisions. Constitutional parameters could emanate from either the Fourth or Fourteenth Amendments. The Fourth Amendment requires that detentions, arrests, and searches be ”reasonable.” The Fourteenth Amendment guarantees equal protection of the law. The U.S. Supreme Court has decided some isolated cases that pertain to the use of race to make law enforcement decisions (for example, United States v. Brignoni-Ponce) as have lower courts (for example, U.S. v. Montero-Camargo), but case law has not fully identified the circumstances when race/ethnicity can (and cannot) be used to support law enforcement actions.
Causes of Racial Profiling
Why might an officer misuse race/ethnicity information in deciding with whom and how to intervene in a law enforcement capacity? We referenced earlier the possibility that some officers have been trained to use race/ethnicity in a manner that some would argue is unjustified. Another answer is that the officer may be racist. That is, the officer may harbor animus toward minority groups and believe that these groups are a danger to society, in need of harsh treatment, unworthy of respect, and so forth. Some stakeholders alleging widespread racial profiling believe that the cause is widespread racism among the law enforcement profession. This narrow way of characterizing the potential causes has led to defensive reactions on the part of chiefs who deny widespread racism in policing.
With the help of social psychologists, the discussion of the potential ”causes” of the misuse of race/ethnicity in law enforcement has broadened. Through lab experiments, social psychologists have determined that racial bias is still widespread in our society. Their research indicates that (1) people link minority status to criminal activity, and (2) this association affects perceptions and behavior (Eberhardt et al. 2004; Correll et al. 2002). This phenomenon has been referred to as unconscious bias because many subjects are not aware of the degree to which they link minorities to crime and many even test as ”nonprejudiced.” This science leads to another possible explanation for why an officer might misuse race/ ethnicity. Some police (like people in any other profession) may not be fully cognizant of the extent to which race/ethnicity affects their perceptions and enters into their decision making.
Such a view has important ramifications for the potential of training to reduce the incidence of racial profiling. Some progressive training programs help police recognize the existence and power of unconscious bias and give them tools to ensure that those biases do not affect decisions.
Community and Police Perceptions of Racial Profiling
A December 1999 Gallup Poll showed that a majority of both whites and blacks surveyed believed that racial profiling on the part of police was prevalent. Fifty-six percent of the whites and 77% of the blacks believed the following practice was widespread: ”police officers stopping motorists of certain racial and ethnic groups because the officers believe that these groups are more likely than others to commit certain types of crimes.” (Data were not provided for other races/ethnicities.) A 2001 Gallup Poll indicated that 44% of blacks surveyed believed that ”police have stopped them at some point in their life because of their race or ethnic background.” Only 7% of the white respondents and 29% of the Hispanic respondents felt this way.
In 2001, more than a thousand executives from a stratified random sample of local and state law enforcement agencies completed surveys about this issue (Fridell et al. 2001). Items on the survey included the following:
• To what extent do you think ”racial profiling/stereotyping” or racially biased policing is a problem in your jurisdiction?
• To what extent do the racial minority citizens in your jurisdiction think that ”racial profiling/stereotyping” or biased policing is a problem in your jurisdiction?
The results indicated that—at least at the time the survey was completed—law enforcement administrators for the most part did not believe that ” ‘racial profiling/ stereotyping’ or racially biased policing” was a serious problem in their jurisdictions, and they believed that the racial minority citizens in their areas would generally agree with them. While no subsequent survey has been conducted to empirically assess changes in agency executive views, there are indications that during the past few years, law enforcement leaders have increasingly perceived racial profiling and the perceptions of racial profiling to be a significant issue with which they must deal. This is indicated by the actions they have taken to address these issues as listed in the next section.
Police Response
Agencies have responded to the issue of racial profiling and the perception of its practice in the following ways:
• Adopting antibiased policing policy
• Training officers not to engage in racial profiling
• Modifying recruitment and hiring in an attempt to hire officers who can act in an unbiased fashion and officers who reflect the racial/ethnic makeup of the jurisdiction
• Enhancing supervision and accountability methods
• Reaching out to diverse communities to discuss racial bias and to improve police-minority relationships
• Collecting data on the stops made by police
Interestingly, data collection has been a particularly popular response as indicated by the number of state-level legislative enactments requiring data collection on the part of police and by the unknown number of agencies voluntarily collecting data. The agencies involved in data collection require officers to report information on each targeted stop. Most agencies are collecting data for just traffic-related stops or for all vehicle stops (the latter includes traffic-related stops and crime-related stops of vehicles). The information collected by officers includes the race/ethnicity of the driver and other information about the driver (for example, age, gender) and the stop (for example, reasons for the stop, disposition of the stop, whether a search was conducted, outcome of the search). A major debate centers on whether the data collected on police stops can indicate whether a police department is or is not engaged in racial profiling.
To draw definitive conclusions regarding police-citizen contact data that indicate disproportionate engagement of racial/ethnic minorities, a researcher needs to be able to identify and disentangle the impact of race from legitimate factors that might reasonably explain individual and aggregated decisions to stop, search, and otherwise engage people. In an attempt to rule out alternative factors, researchers strive to develop comparison groups against which to evaluate their police-citizen contact data. Agencies strive to develop comparison groups (“benchmark” groups) that most closely reflect the demographic makeup of groups at risk of being stopped by police assuming no bias. For example, a department collecting data on traffic stops would, ideally, want to compare the demographics of those stopped by police for a traffic violation with the demographics of those people legitimately at risk of a stop, taking into consideration numerous factors, including, but not limited to, driving quantity, driving quality, and driving location.
Many agencies that are collecting data are analyzing their data using relatively weak benchmarks. For instance, many agencies are conducting “census benchmarking” whereby the analyst compares the demographic profile of drivers stopped by police to the U.S. Census Bureau demographic profile of jurisdiction residents. This is a weak benchmark because the group of people who reside in a jurisdiction do not reflect the group of people who are at legitimate risk of being stopped by police. Not all people who are stopped by police live in the jurisdiction and not everyone who lives in the jurisdiction drives the same amount or violates traffic laws at the same rates.
Researchers and practitioners are struggling to identify methods for developing valid comparison groups. As yet, there is no consensus on what might be the most cost-effective and valid benchmark(s). Some jurisdiction teams are comparing the demographic profiles of persons stopped with the demographic profiles of licensed drivers, people involved in vehicle crashes, or persons observed to be driving in the jurisdiction. In jurisdictions with cameras that record the license plates of vehicles running red lights (a “color-blind” form of enforcement), the demographic profile of the owners (unfortunately, not the drivers, necessarily) of those vehicles can be compared with the profile of persons stopped by police. Similarly, radar stops can be a colorblind form of enforcement allowing agencies to compare the demographic profile of citizens stopped with radar to the demographic profile of citizens stopped without radar. Jurisdictions can also rely on internal comparisons—comparing the profiles of people stopped by individual officers or units to the corresponding data for other officers or units that are ”matched” in terms of their assignments (for example, beat, shift). Resident surveys can be used to collect data, not only on the extent to which residents are stopped by police but also on the nature and extent of their driving. Corresponding to the data that police collect on their own activities, the survey could solicit information from the respondents regarding the frequency and nature of their encounters with the police. Corresponding to benchmarking efforts for department-collected stop data, such a survey could ask the residents about the nature (for example, speed, passing behavior, driving violations), location (for example, interstate highways, around their neighborhood), and amount of their driving.
The Impact of September 11, 2001
The terrorist attacks on September 11,2001, had several impacts on how people in the United States conceived of racial profiling. Prior to the attacks the discussion about racial profiling focused primarily on the treatment of blacks and Hispanics. Following 9/11, the focus expanded to include people of Arab descent and Muslims. While many people in the United States recognized the potential to racially profile people of Arab descent and Muslims, they also reconsidered the extent to which racial profiling was bad and, accordingly, revised their views about when it was and was not appropriate to use race/ethnicity to make law enforcement decisions. As one manifestation of this changed view—a change made in the context of a national crisis and citizens’ associated fears—a Gallup Poll conducted soon after the terrorist attacks indicated that 71% of black respondents favored intense scrutiny of Arabs boarding airplanes.
