Samuel Duncan (Tidewater Community College), "Why Police Shouldn't Be Allowed to Lie to Suspects"

Forthcoming, Journal of the American Philosophical Association

By Samuel Duncan

If you’ve watched a cop drama you know that even good cops lie. As countless interrogation scenes show us, good cops are supposed to lie. Lying is part of the art of interrogation and absolutely necessary to get confessions and evidence from criminals who have no reason to give up either. And of course we all know that manipulating a suspect into confessing or surrendering evidence is a  far different thing than beating it out of them as the stereotypical dirty cop like The Shield’s Vic Mackey or L.A. Confidential’s Bud White.

For all their exaggeration and simplification this part of police dramas reflects actual theory and practice in U.S. law enforcement. Before the 1929 Wickersham commission report, violence was the standard operating procedure of American police interrogations. After it uncovered just how widespread and extreme such violence was, law enforcement had to find a new way. The Reid method stepped in to fill that role and occupies it still. The Reid method strictly forbids physical violence and relies instead on manipulating and psychologically coercing them. While it boasts a few more subtle tricks and tactics, its main approach is to simply lie to them (and if necessary to suspects’ friends and family and even witnesses as well). The adoption of the Reid method has long been celebrated by American law enforcement as clear and unambiguous moral progress. That beating or otherwise abusing suspects is wrong and has no place in the ordinary day to day practices of law enforcement is almost universally accepted by both those who theorize about the ethics of policing and police officers themselves.  On the other hand, lying to suspects is supposed to be a far different thing, and replacing physical violence with deception marks clear progress in bringing the practice of policing in line with the ideals of respect for persons so central to liberal democracy.

I think this is bunk. In my “Why Police Shouldn’t Be Allowed to Lie to Suspects” I argue that lying to suspects is wrong in fundamentally the same way as physically abusing them is. The best account of why beating a confession of a suspect is in principle wrong is a broadly Kantian one that holds that doing so is wrong because it fails to respect them as a rational being. Such Kantian reasoning also provides the best account for another fixed point in the ethics of policing: It is wrong for officers to perjure themselves on the witness stand. However, a Kantian account also forbids lying to suspects. It is simply inconsistent to hold that lying to suspects for the sake of expediency is acceptable while beating them or lying to judges and juries when doing so is expedient is not. In fact, many police officers have themselves noted that absolutely forbidding lying on the witness stand while recommending it for its usefulness in the interrogation room is contradictory, though the implication they draw from it– that perjury is perfectly fine when useful to ends of law enforcement– is not the one we should draw from it.

I know that many readers will object that all this is part and parcel of Kant’s own ridiculously strict absolute ban on lying, which forbids lying to a would-be murderer to save his victim. In response, I show that on the best accounts of why a Kantian moral theory should in fact allow lying in cases like the murderer at the door it would still forbid practically all lying to suspects since police officers cannot claim that doing so is necessary to defend others. In the same vein I challenge the claim that the widespread practice of lying to suspects is even justified from a purely consequentialist perspective. Lying to suspects may seem to make investigations easier, but it has steep hidden costs. When members of the public simply become aware of the fact that lying is so fundamental to the practice of interrogation it corrodes the trust between police officers and the communities they serve necessary for effective law enforcement. Would you cooperate if the police wanted to question you after reading this post? If you were a juror, how much would you trust a detective who you know lies to suspects, and even witnesses whenever it is expedient? Even more damaging in this regard is the fact that the Reid method has produced false confessions on numerous occasions. This leads to miscarriages of justice like the convictions of the Central Park Five that do profound damage to trust and faith in not just the police but the entire American criminal justice system. I conclude that lying to suspects should be strictly forbidden and the Reid method consigned to the same fate as the more openly immoral methods it replaced.