Drug Enforcement Agency

In defining racial profiling, it is important to distinguish it from two law enforcement practices that bear some resemblance to it, but which are legal and, if carried out correctly, nondiscriminatory. The first is “criminal profiling,” which can take two forms. One is the development of a constellation of characteristics that are predictive of a perpetrator of a particular crime. This can be thought of as profiling broadly defined. It is generally a more formal process than racial profiling and can result from either formal agency guidelines (as in the case of early Drug Enforcement Agency drug courier profiles starting in the 1970s) or informal stereotypes of criminals held by individual officers. This practice is also referred to as “behavioral profiling,” although the emphasis in behavioral profiling protocols is, as the name indicates, typically not on trait characteristics (e.g., age, gender), but rather on behaviors such as loitering, avoiding eye contact, and furtiveness (in the case of drug crimes), or purchasing one-way tickets with cash and then traveling without luggage (in the case of terrorism). T he second form of criminal profiling is an older variety that is qualitatively different. Whereas racial profiling involves searching for suspects of crimes (e.g., drugs or weapons possession, terrorism) that are not yet known to have been committed, classical criminal profiling involves narrowing the pool of suspects for a known (already perpetrated or, based on valid intelligence, imminent) crime. For example, investigators seeking to apprehend serial killers have utilized criminal profilers (sometimes forensic psychologists or psychiatrists) who develop a demographic (and possibly psychiatric) profile of the likely perpetrator. There are some remarkable success stories, but on the whole this kind of criminal profiling has had limited success, at best (Silke, 2001; Winerman, 2004). Another important distinction is between racial profiling and the use of race in suspect descriptions. Although some legal scholars (e.g., Banks, 2004) argue that use of race in suspect descriptions can, under some circumstances, be a discriminatory practice akin to racial profiling, the two procedures must be kept distinct from one another. Use of race in suspect descriptions necessarily involves a known crime with specific evidence (e.g., a witness report) of a suspect’s race, while racial profiling involves as yet unknown crimes with no direct basis for inference about a possible criminal’s race. Racial profiling must be distinguished from these other practices because they are often conflated, with the effect of confusing the issue and, more troublingly, justifying racial profiling. By arguing that use of race to identify suspects in any form is racial profiling, some seek to raise concerns about throwing out a valuable law enforcement tool. However, using race or ethnicity to describe a known suspect of a known crime is a perfectly legitimate and, no doubt, effective investigatory practice. Similarly, criminal profiling and offender profiling are legitimate tools to narrow the pool of suspects for a known crime, although there is only anecdotal support for this approach and no empirical evidence that it is a reliable practice. Behavioral profiling, which by definition focuses on behaviors and not traits like race, has been found to be effective, at least in the case of counterterrorism (Silke, 2011). A precise definition of racial profiling that avoids conflation with other legitimate law enforcement practices affords a meaningful consideration of the subject on its merits. O ne other source of definitional imprecision that can undermine a productive consideration of the subject is the tendency for some commentators