"Probabilities, Perceptions, Consequences and 'Discrimination': One Puzzle About Controversial 'Stop and Frisk'"

The title of this post comes from this essay by Professor Kent Greenawalt, the abstract of which states:

The essay focuses on a particular issue about “stop and frisk” raised by the concern about concentration on young men in racial minorities. Although reference is made to the recent decision in Floyd v. City of New York, my focus is not on what has actually happened in New York and elsewhere, but whether, if there is substantial evidence that a higher percentage of members of a group commit certain kinds of crimes more than members of other groups, it is appropriate for police to stop partly on the basis of whether movements that are somewhat suspicious are made by members of that group, or whether that constitutes inappropriate discrimination.

Using an imagined private analogy and made up statistics about levels of criminal behavior and innocence among different groups, I contend that given the level of probability needed for a stop, i.e., “reasonable suspicion,” the likelihood that a group characteristic could figure importantly in the necessary probability is somewhat higher than when a more demanding standard such as “probable cause” is in play. In respect to a person’s gender and age, I suggest that use of that in applying a probability standard is really uncontroversial. But race is different. Both because of the risk of nonobjective appraisals, based partly on prejudices or implicit cultural assumptions, and because of the harmful negative message conveyed to members of minorities subject to stops, I conclude that police should not explicitly use race as a factor (unless they are looking for a person already identified by his race). I also believe this problem is serious enough so that such use should be viewed as a constitutional violation.