VIII. The Flagship Cases in Racial Profiling
Three cases exposed the police application of the “drug courier profile” concept to scrutiny. In April 1998, two New Jersey State Troopers initiated a vehicle stop of a van on the New Jersey Turnpike that resulted in the shooting and wounding of three of four young minority men. The “profile” was cited as one reason for the stop, even though the vehicle was southbound. No drugs or other contraband were found in the van, and there were significant discrepancies between the official police report of the incident and other evidence.
The Turnpike shooting case revived scrutiny of New Jersey State Police practices that had languished since the 1996 case of State of New Jersey v. Pedro Soto et al. In that case, a New Jersey court overturned 17 drug possession cases brought by a state police drug interdiction team working another section of the New Jersey Turnpike. The defendants produced evidence that the team stopped and searched minorities, particularly African Americans, almost 5 times more frequently than they did white drivers.
Though black drivers were only 15% of the “violator” group (people driving at excessive speed, or committing other observable moving violations), vehicles driven by blacks comprised almost half of the stops made by the interdiction unit. By comparison, a state police unit doing speed enforcement in the same area of the New Jersey Turnpike stopped drivers at rates much closer to the observed proportion of highway use. The difference strongly suggested that the drug interdiction team equated racial identity with a greater likelihood of participation in illicit distribution of drugs.
A civil case, Wilkins v. Maryland State Police (2003), used the same analytical technique employed in the Soto case. Though whites comprised almost three quarters of drivers and speed violators, only 20% of the cars searched were driven by whites. Blacks made up 17% of the driving population, and slightly more of the violators, but were almost 73% of all persons searched. A total of 4 of every 5 drivers searched by the Maryland State Police interdiction team were of minority status.
If minority drivers were indeed more likely to have drugs than their white counterparts, a stronger argument might be made that the observed disparities were justified. However, the Maryland case revealed that the proportion of black and white drivers found to be carrying drugs was practically equal: 28.4% of blacks, and 28.8% of whites (other sources cite a slight difference: 34% of blacks and 32% of whites; the difference remains marginal). The New Jersey proportions were even lower: 13%of blacks, 10.5%of whites, and 8% of Latinos carried drugs. The New Jersey proportions are closer to the findings in other early studies; the Maryland hit rates are the highest in that first round of profiling inquiries.
More telling, none of the cases studied suggested that the drugs found in profile-based searches were pre-market bulk quantity (the quantum envisioned by the drug courier profile). New Jersey’s post-Soto review explicitly noted that almost all seizures were of small amounts, consistent with post-market personal use (Verniero & Zoubeck, 1999); the conclusion is implicit in the absence of any discussion of the quantum of seizures in other studies. The inherent difference in the danger posed by market-quantity shipments, in contrast to personal-use quantities, is a potential limit on the latitude to be allowed government agents, and a fundamental problem with the game theory school of proof (discussed below).
Further inquiry into the New Jersey State Police practices in the wake of the April 1998 Turnpike shooting revealed that the state’s training regimen included material that essentially equated minority citizens with greater criminality, effectively directing troopers to focus on minority motorists (Verniero & Zoubeck, 1999).
Both Maryland and New Jersey entered into consent decrees with the U.S. Department of Justice to amend their practices, monitor trooper performance, and revise training that had perpetuated the racial stereotype. Those findings established racial profiling as a “fact,” but only in the two jurisdictions. In the public sphere, an ironic reversal of positions occurred. Where the police had tacitly assumed the actions of a small group of drug couriers were typical of the minority communities, the public now assumed that the practices of two agencies represented police practices everywhere. The distinction remains central to the debate over racial profiling in any guise: The generalizability of localized findings (or enterprise-specific profiles) to larger groups sharing only superficial aspects of the offending groups is limited.
The police insist that profiling is a legitimate tool of inquiry, well grounded in the experience of practice. The public points to the Maryland and New Jersey statistics, standing firm in their belief that it is merely a modern continuation of racial prejudice, now dressed up with pseudoscientific prose. The scientific questions that have arisen in the wake of the Soto and Wilkins cases have examined a wider range of police activities and actions in numerous other venues. They embody an attempt to discern whether and where the “fact” of racial profiling by police exists beyond the specific milieu of the interstate highways in New Jersey and Maryland.
Overall, four issues are salient. The first is whether criminal propensity is more likely in one group than another, with a collateral question of what purported proofs should be considered valid. The second area involves the continuing refinement of methodology, and the factors to be considered when examining aggregate police data in vastly different locations for evidence of racial bias. A contemporary assertion, promulgated in different fashion by the police and by the econometric school of criminology, is that the “criminal propensity” point can be proven by a variation of game theory, based upon the results of consent searches without regard for any other considerations (including those of civil rights). The third is a legalistic consideration of whether elevated government intrusion justified against large-scale harm (proliferation of drug markets, or another 9/11-type attack) can be applied equally to smaller gains (seizure of personal-use drug amounts). A fourth operational issue rests upon the degree of precision with which the profile is applied by police officers and other agents of the state.