Should coercive interrogation be legal?

• Author: Posner, Eric A.

INTRODUCTION I. FIRST-ORDER CONSIDERATIONS: MORAL LIMITS ON COERCIVE INTERROGATION A. On Deontological Grounds, Coercive Interrogation Is Flatly Impermissible B. On Deontological Grounds, Coercive Interrogation Is Impermissible Except to Prevent "Catastrophic Harms". C. On Rule-Consequentialist Grounds, Coercive Interrogation Is Impermissible II. SECOND-ORDER CONSIDERATIONS: THE EMPIRICAL AND INSTITUTIONAL CONTEXT A. Rules and Standards B. Slippery Slopes C. Symbolism III. LEGAL REGULATION OF COERCIVE INTERROGATION A. Outlaw and Forgive 1. Popular Justice 2. The Necessity Defense B. The Torture Warrant C. A Proposal for Regulating Coercive Interrogation D. The Burden of Uncertainty IV. WHY IS COERCIVE INTERROGATION TABOO.'? A. Mistaken Generalization B. Moral Heuristics C. Judgment Falsification, Cascades and Herding CONCLUSION INTRODUCTION

Coercive interrogation is now a live subject, thanks to 9/11. At one time, coercive interrogation played a role only in philosophical disputes about consequentialism, in which scholars asserted or denied that the police could interrogate an individual in order to extract the location of a ticking nuclear bomb. None of the participants in those debates seriously considered the possibility that coercive interrogation could be justified except in extreme circumstances never likely to be met. Today, U.S. officials appear to engage in coercive interrogation or something very similar to it; so do other western governments; and the possibility that coercive interrogation may be justified in nonremote circumstances has entered mainstream debate. (1) The task for legal scholars at this point is to understand how this practice fits into legal norms and traditions, and how it ought to be regulated.

Let us define some terms, and delimit the topic. "Coercive interrogation," we will say, involves (1) the application of force, physical or mental (2) in order to extract information (3)necessary to save others. (2) Coercive interrogation can range from the mild to the severe. At some point of severity, coercive interrogation becomes a species of "torture," which is flatly prohibited by domestic and international law. (3) Coercive interrogation and torture are thus partially overlapping concepts; neither is a proper subset of the other. Mild coercive interrogation does not amount to legal "torture," which requires that a threshold of severity be met. And there are forms of torture that are not coercive interrogation--for example, when torture is used as a means of political intimidation or oppression, indeed for any purpose other than extracting information necessary to save third-party lives. Our interest is in the overlapping area of these two concepts: coercive interrogation that (by virtue of its severity) counts as torture. Henceforth, we will use "coercive interrogation" to denote this subset.

Given these stipulations, our inquiry is normative. We ask what legal regime should govern coercive interrogation. Should it ever be permissible? If so, what legal rules should be used to sort permissible from impermissible cases? Among legal academics, a near consensus has emerged: coercive interrogation must be kept "illegal," but nonetheless permitted in certain circumstances. (4) How is this trick accomplished? There are two popular suggestions. First, interrogators can use the necessity defense, which would permit government agents to argue in specific cases that violating the laws against coercive interrogation was necessary to discharge their duty to protect the public from an imminent terrorist threat. Second, interrogators can throw themselves at the mercy of the political process, and seek a pardon, or a favorable use of prosecutorial discretion, or some similar political immunization. The idea is to make coercive interrogation such an unattractive option for officials--they will be personally liable unless the strict conditions of necessity are met or the political process smiles on them--that they will use it only as a last resort. And this regulatory structure is meant to have an expressive dimension: maintaining the "illegality" of coercive interrogation expresses a moral commitment to human dignity mad autonomy, while the possibility of defenses and pardons allows its use where appropriate.

The whole idea is puzzling. Police are allowed to use deadly force in order to prevent dangerous suspects from harming other people. Killing a person is also a serious harm to dignity and autonomy; although we will see arguments holding that coercive interrogation is worse than killing in some respects, there are other respects in which killing is worse than coercive interrogation. To prevent officials from engaging in unjustified killings, governments take the conventional route of enacting laws that describe the conditions under which a police officer may use deadly force, making the police liable only if they violate these rules in bad faith. Why shouldn't the same system be used for coercive interrogation? Or, conversely, why not prohibit police killings on the theory that such a prohibition would ensure that police would kill only when they anticipate that, after they are charged for murder, they can successfully plead the necessity defense or obtain a pardon?

Or consider the use of force during war. The laws and usages of war permit soldiers to kill other soldiers, and civilians as well. Although the killing of civilians is generally regarded as a moral evil, it is justified and permitted when civilian deaths are not disproportionate given a legitimate military target. (5) If governments can authorize the killing of civilians in order to accomplish legitimate military objectives--which are all means to the end of national security--why can't government authorize coercive interrogation for the same purpose? Or, conversely, why not prohibit the killing of civilians and require soldiers to seek a pardon or some other form of political forgiveness, before or after they are tried for murder?

In short, the view that coercive interrogation should remain illegal assumes that coercive interrogation is special in a way that distinguishes it from police killings and other serious harms that officials are licensed to inflict; but what makes coercive interrogation special?

The answer, in our view, is that coercive interrogation is not special at all. If it is agreed that coercive interrogation is justified in certain circumstances, even narrow circumstances, there is no sense in treating it as "illegal" but subject to ex post political or legal defenses. It should be made legal, albeit subject to numerous legal protections--again, in this way like police shootings, wartime killings, preventive detentions, capital punishment, and other serious harms. The law should treat coercive interrogation the way it typically treats coercive governmental practices. Such practices are subject to a standard set of regulations defined ex ante: punishment of officials who use these instruments without a good justification, official immunity when they are used in good faith, various restrictions on the type of instrument that may be used, ex ante protections such as warrants, (6) and so forth. Our argument is that coercive interrogation should be treated in the same way.

Part I provides a brief and selective overview of the first-order philosophical issues. Our purpose here is to delimit the topic in two critical ways. First, we bracket and ignore the claim that coercive interrogation is deontologically impermissible per se, whatever the facts. With a very few exceptions, this is a view nobody holds; most mainstream philosophers--both consequentialists and deontologists--agree that coercive interrogation may be morally justified under certain conditions. Second, we outline the rule-consequentialist view that the harms of coercive interrogation are so great, the occasions for its justified use so infrequent, and the risks of decisionmaker error so high, that coercive interrogation should never be permissible. The rule-consequentialist view turns on empirical and institutional premises that we discuss in Parts II and III. The only philosophical point is that, for either deontologists or consequentialists who believe that coercive interrogation can sometimes be permissible, there is no philosophical justification for thinking that coercive interrogation should be considered special, and regulated differently from the other serious, coercive harms that government inflicts.

Part II addresses second-order empirical and institutional arguments for treating coercive interrogation as special in the legal system (that is, regulating coercive interrogation by a different legal regime than applies to other serious harms government may inflict). These arguments rely on various tropes of second-order argument--rules versus standards, slippery slopes, institutional failure, corruption, and so forth--that in this case turn out to rest on implausible empirical premises. Our more precise point, however, is that if these arguments were accepted for coercive interrogation, many other common practices would have to be prohibited as well--for example, the shooting of armed suspects.

In Part III, we argue that banning a practice and then asking officials to engage in it (when justified) and ask for public forgiveness is not a plausible strategy for giving officials the right incentives. All of the rule of law reasons for creating a set of ex ante regulations that govern official conduct--rather than regulating official conduct ex post--apply as much to coercive interrogations as to other forms of law enforcement. Moreover, a regime of ex ante illegality and ex post license is conceptually unsustainable. If officials and citizens know that ex post defenses and forgiveness are available, they will factor their knowledge into their understanding of what the law is, diluting the material and expressive effects of the "ban" on...