B. Scientific Proof
Criminologists seek a scientific justification for police actions, in lieu of normative reliance upon unverifiable “sixth sense” justifications. The contemporary scholarly focus on racial profiling rests upon a debate regarding whether a variation on game theory can provide such a foundation.
One thrust of the early racial profiling research took up the question of whether disparity alone constituted de facto discrimination, or could occur innocuously.Various studies identified different uses of the highway for minority groups based on employment opportunities and recreational pursuits (Meehan & Ponder, 2002a). Another documented disproportionate representation of young minority males in the population of high-speed law violators on the New Jersey Turnpike, inferring the legitimacy of at least one assumption of the original profile (Lange, Johnson, & Voas, 2005).
The contemporary debate centers on a proposition by scholars drawing from the field of econometrics, asserting that the question of disparity or discrimination can be resolved by the application of a variation of game theory (Engel, 2008; Persico & Todd, 2008). This solution hinges on an examination of the “hit rate” (rate of discovery of contraband) of police searches, independent of any other concerns (Persico & Todd, 2008), treating the populations of white and non-white drivers as “urns” that can be sampled. This view posits that rational choice underlies the options of drivers to carry contraband (in the present discussion, drugs), and the choice of police officers to conduct investigations for drugs. Rational police officers would focus their efforts on groups most likely to carry drugs.
In essence, this econometric argument takes the side of the police, tacitly accepting the argument that group characteristics do exist, and can be inferred at the individual level. Furthermore, the acceptance of group-specific criminality (drug use, in this instance) assumes facts not in evidence. It is a hypothesis in the classic methodological sense, to be proved or disproved by scientific testing that cannot be conducted, and thus remains hypothetical.
Methodologically, the hit rate proposal encounters several difficulties. Police searches are conducted under a variety of situations, including searches incidental to arrest, inventory searches when a vehicle is towed, and in circumstances when probable cause is developed at the scene independent of the original reason for the stop. In addition, searches are conducted for a variety of motives, not just drugs (Engel & Tillyer, 2008). As a result, analyses based solely upon the hit rate for drugs must be able to discern those stops and searches motivated by an officer’s desire to “maximize the probability of a ‘hit’” (Persico & Todd, 2008). A further complication arises in the need to distinguish personal-use hits from the bulk drug seizures arising from the proper application of the drug courier profile, although the modern debate fails to pursue that distinction.
The proposition of hit-rate analysis has the advantage of being theoretical, unconnected from the larger concerns of the criminal justice system. The individuals who comprise the “urns” that police officers supposedly sample are not research subjects, but citizens of a country that presumes their innocence. They are defended by rules and expectations of conduct by agents of the state that do not allow random selection for the purposes of testing. Basic civil rights preclude the use of the motoring public (or the ambulatory public) as a laboratory.