My colleague Craig Lerner has a brilliant new piece exploring the implications of modern cognitive science for the idea of "hunches" by experienced police officers and the dilemma they present for the modern police system. As he puts it in the Abstract to his new article "Reasonable Suspicion and Mere Hunches":

In Terry v. Ohio, Earl Warren held that police officers could temporarily detain a suspect, provided that they could articulate the "reasonable inferences" for their suspicion, and not merely allude to a "hunch." Since Terry, the American legal system has discounted the "mere" hunches of police officers, requiring them to articulate "specific" and "objective" observations of fact to support their decision to conduct a stop and frisk. The officer's intuitions, gut feelings and sixth sense about a situation are all disallowed.

This dichotomy between facts and intuitions is built on sand. Emotions and intuitions can be reasonable, and reasons are often predicated on emotions. Even as courts have, over the past two generations, grown more dismissive of hunches, there has been a counter-revolution in the cognitive sciences. Emotions and intuitions are not obstacles to reason, but indispensable heuristic devices that allow people to process diffuse, complex information about their environment and make sense of the world. If the legal rules governing police conduct are premised on a mistaken assumption about human cognition, can one craft a doctrine of policing that credits the wisdom of hunches? Can the legal system defer to police officers' intuitions without undermining protections against law enforcement overreaching?

This article argues that, to some extent, judicial skepticism about police hunches can and should be abandoned. As a practical matter, the current legal regime substitutes palliative euphemisms for useful controls on police discretion. When an energetic police officer has a hunch that something is wrong and action is imperative, the officer will simply act. Months will pass before a suppression hearing, and by then it will be a simple matter to reverse-engineer the objective "reasons" for the stop - e.g., "I saw a bulge." The legal system in practice simply rewards those officers who are able and willing to spin their behavior in a way that satisfies judges, while it penalizes those officers who are less verbally facile or who are transparent about their motivations. It would be preferable if politically accountable authorities joined the courts in monitoring police practices. The focus should be less on what police say after the fact and more on what they do - that is, how successful police officers are in catching criminals and how respectful they are of all citizens.

The point will be familiar to readers of Hayek and Polanyi who draw on the notion of tacit knowledge, and in particular, the notion that much knowledge that derives from experience, tradition, and habit is unarticulated knowledge which is no less valuable or important for being unarticulated or unarticulable. As Lerner notes, the recognition that such knowledge exists places a Court in the position of either excluding this valuable knowledge from the adjudication, or forcing experienced police officers to falsely articulate an untrue explanation for what is fundamentally unarticulated knowledge.

Craig has identified a neat puzzle here.

Update:

Doh--I initially forgot to link to the paper. I've corrected it now.