The mortification of Iraqi prisoners by American military personnel at the Abu Ghraib prison in Baghdad has been discomfiting far beyond the impact of the now-infamous images. Coupled with other reports about harsh post-9/11 tactics to garner information from captured terrorists, and with ongoing investigations into deaths alleged to have occurred in connection with interrogations, Abu Ghraib and the reaction to it have forced front and center a profound national evasion: the propriety of torture.

As one would expect, the scandal has produced no small amount of righteous indignation. The civil-libertarian lobby, operating in overdrive, has issued ringing declarations that torture is unacceptable under any circumstances; accused the Bush administration of giving a green light to the humiliation of captives; and demanded the jettisoning of established international norms in favor of protocols codifying new rights for mass murderers. The financier George Soros, who has thrown millions of his billions behind various left-wing causes, recently proclaimed that Abu Ghraib was the functional equivalent of the 9/11 attack, only committed this time by the United States.

On the other side, deep disapproval of the abuse has been joined to brave talk about how we must make allowances for a “new kind of war,” and to reminders that Abu Ghraib under American malefactors was a day at the beach compared with Abu Ghraib under Saddam and his ghouls and that our terrorist enemies, instead of stripping their captives naked and leashing them like dogs, tend to behead them instead. This is all true, as far as it goes, but it has been largely unaccompanied by any examination of the key question—namely, what are, and what are not, appropriate methods of interrogation? Appropriate, that is, according to American values and not the values of humanity's basest elements.

Finally, there are the centrists, who well understand that our enemies are covert operatives bent on killing us in sneak attacks, and that the only way to foil them is to get information about who they are and when and where they will strike next. Unfortunately, zealots inspired by Islamic militancy and willing to immolate themselves in suicide assaults are not likely to share their secrets under the comparatively mild duress of humane captivity. Thus, the centrists figure, there is probably some necessary torture afoot—which they think wrong, or at the very least unsavory, but from which they would prefer to avert their gaze.

This is all, as I say, to be expected. If the spectacle of ruthless mass murder à la 9/11 evokes blind vengefulness in some, in others it triggers deep-seated habits of denial or self-blame. In the meantime, the mere mention of torture is enough to engender disquieting thoughts of the dark brutality of which men (and women!) have historically shown themselves capable. Under the circumstances, rationality is not a good bet to rule the day Nevertheless, it is where any discussion of the terrorist threat and how to deal with it must begin and end.

Terrorism in general is not a new phenomenon, but today's global and systematic menace is plainly not the threat that was contemplated when international humanitarian law, in the form of the Hague Convention, the Geneva Conventions, and the United Nations Convention Against Torture, took root. Those agreements were designed to institutionalize the laws of war, to promote the humane treatment of captured combatants, and to reduce civilian casualties in times of armed conflict. They were written for wars among readily identified forces—nation-states and, to some extent, intra-state liberation movements and organized insurgencies—in which conflict-appropriate tactics were well established and victory (or defeat) was easily visualized. They did not anticipate militant Islam: a sub-national force of international scope, with access to weapons of unfathomed destructive power. They did not contemplate a core methodology—targeting civilians, randomly torturing and killing prisoners—that grossly and willfully violates the very premises of humanitarian law. They did not take account of a situation in which our highest priority would be to obtain not territory or treasure but intelligence, and in which victory would be exceedingly difficult to define.

Does this mean that the time has come to upend the entire structure of the law of armed conflict? Hardly. The impulse for making law is driven by a firm grasp of what kinds of conduct are condemnable. The fact that militant Islam has organized itself precisely within the parameters of the universally condemnable—that the terrorists, to adapt Daniel P. Moynihan's famous phrase, have defined deviancy right down to the bottom—is not a reason to reshape norms in order to license execrable behavior. Rather, it compels us to weigh the overriding imperative of our national security, without which our own civil liberties would be nonexistent, against the sanctity and dignity of human life, even when that life belongs to a captured terrorist. In the end, it compels us to ask rationally whether torture is ever permissible.


Before coming to that central issue, it would be helpful to clear away some underbrush surrounding the question of who may lawfully be held for questioning. This is necessary because humanitarian-law activists, many of whom seem more preoccupied with the treatment of terrorists than with the carnage wrought by terrorists, have been working feverishly to convince us that the government's denial of prisoner-of-war (POW) status to al Qaeda terrorists captured in Afghanistan and held in Guantanamo Bay, Cuba, somehow caused the abuse of actual POW's in Iraq.

As it happens, however, whether one is or is not a POW is a matter not of a moment's intuition but of long-settled principle—of, one might even say, merit. “POW” is an honorable legal status. In order to earn that status, and its protections, one is expected to conduct oneself honorably. Granting it to combatants who do not do so would only offer a further incentive to belligerents to act dishonorably. Granting it to terrorists, in particular, would gravely endanger the lives of countless civilians—which, in the greater scheme of things, would lead to far worse evil than the deplorable treatment of a relative handful of prisoners.

It may well be that the failure adequately to regulate and monitor interrogation practices led to a culture that made Abu Ghraib more likely. That, however, has little to do with the legal status of those being interrogated. And it is incongruous, to say the least, to assail the Bush administration for the alleged consequence of a perfectly appropriate decision concerning the status of terrorists while ignoring the foreseeable (in truth, inevitable) consequence of abolishing the very distinctions on which that status rests. The whole rationale for having POW rules in the first place is to encourage civilized warfare and to protect civilians.

Where, then, do those distinctions come from? The relevant principles, developed over centuries, have been passed down to us primarily as codified in the Hague Convention IV of October 18, 1907 and the Geneva Conventions of August 12, 1949, particularly the Third, which addresses the “Treatment of Prisoners of War.” It is crucial to remember that these are laws of war, and that the primary object in wartime is forcibly to defeat the enemy. In that sense, they constitute a practical effort to limit abuse and suffering while remaining ever cognizant of the martial context, which is to say of each side's drive to win. Moreover, these laws do not purport to prevent all discomfort to captives or all collateral damage to civilians and non-combatants. Rather, they seek to establish normative guidelines in which legitimate military objectives are pursued without causing disproportionate harm, fully understanding that harm is, to some extent, unavoidable.

The Hague Convention conveys the honorable notion of “lawful” or “privileged” combatant. During wartime, combatants are privileged to employ military force if they are members of a national army or of a militia that is part of such an army and that conducts itself accordingly meaning that its members are subject to a formal chain of command; wear uniforms (i.e., “have a fixed distinctive emblem recognizable at a distance”); carry their weapons openly; and conduct their operations in accordance with “the laws and customs of war.”

And what are those laws and customs? By the early 20th century, they comprised a well-founded balance of military necessity with humanitarian considerations based on the principles of “proportionality” and “distinction.” The first called for factoring the likely damage to civilian infrastructure into the calculus for identifying proper military objectives; the second, for limiting acceptable targets to those objectives. The legal scholars David B. Rivkin, Jr. and Lee A. Casey have usefully summarized the resultant set of rules, which were long ago accepted “by all civilized states”:

(1) only sovereign states have the right to make war; (2) civilians cannot be deliberately attacked; (3) combatants can be attacked either en masse or individually; (4) quarter is to be granted when sought; (5) lawful combatants, when taken prisoner or otherwise incapacitated by wounds, are to be accorded the respect and privileges due prisoners of war (POW's); and (6) while all forms of force can be deployed in combat, certain weapons designed to cause unnecessary suffering are proscribed.1

Two other laws are likewise important for our purposes here, and are logically antecedent to rule (5) above: captured combatants may be held until the conclusion of hostilities, and they may be interrogated. Lately there has been much caterwauling over these points, particularly as they relate to captured members of al Qaeda. Leaving aside for the moment whether such persons may be classified as lawful combatants, their detention is routinely decried as unacceptable because it is “indefinite”—as if there has ever been a war in which captives could be assured that hostilities would end on a date certain.

In fact, the theory behind these laws is simple and irrefutable. Captives may be held “indefinitely” because if they are released they are likely to rejoin the battle, thus prolonging the war and its attendant suffering. By contrast, imposing a maximum and arbitrarily chosen period of detention (say, three years) would defeat the logical purpose of the detention. Similarly, captives may be questioned because they are likely to have information which, if learned by the captor state, will protect its forces, make it easier to pinpoint legitimate military targets of strategic importance, and secure victory more promptly—all of which saves lives.


The Geneva Conventions, promulgated in the middle of the last century, offered additional humanitarian protections. What impelled their drafting were the widespread atrocities visited upon captured prisoners of war during World War I and the savagery of World War II, in which civilians were broadly targeted, exterminated, and subjected to sundry lesser horrors. These 1949 accords are generally applicable only between and among national powers that have entered into them.

The Third Geneva Convention affords specific protections to prisoners of war: that is, lawful or privileged combatants who have been captured while taking part in hostilities. POW status is generally limited to two categories: members of the armed forces of nations that are parties to the conflict, and members of militias and organized resistance movements that belong to a nation that is a party to the conflict, provided they fulfill the conditions mentioned above (being part of a formal chain of command, wearing uniforms, etc.).

Those who satisfy these criteria are entitled “in all circumstances to respect for their persons and their honor.” POW's, moreover, “must at all times be protected, particularly against acts of violence or intimidation and against insults and public curiosity.” Similarly prohibited are “measures of reprisal against” them. The detaining power, in addition, is required to provide all POW's with health care and maintenance and to show no preference based “on race, nationality, religious belief, or political opinions, or any other distinction founded on similar criteria.” Furthermore, while POW's may be questioned broadly, they are obligated to reveal—in the time-honored formulation-only their name, rank, and serial number. Finally, “prisoners of war who refuse to answer may not be threatened, insulted, or exposed to unpleasant or disadvantageous treatment of any kind.”

Militant Islamic terrorists like those belonging to al Qaeda manifestly do not qualify for POW status because they are not lawful combatants: they are not part of a nation state, they are not signatories of the Geneva Conventions, they do not wear uniforms, they do not as a rule carry their weapons openly, they hide among (and thus gravely imperil) civilian population and infrastructure, and they intentionally target civilians for indiscriminate mass homicide in order to extort concessions from governments they oppose. As a result, they do not enjoy the special privileges of POW's, such as entitlements during interrogation to limit their answers to “pedigree” information and to refuse to answer other questions.

But there is a complication. It is true that the 1949 protocols do not apply to al Qaeda. Nevertheless, even those who rightly maintain that the general protections of the Conventions are unavailing to terrorist organizations have found in them a high-minded proscription against torture that extends to everyone, including terrorists. As best I can tell, this interpretation is rooted in a pertinent passage of Article 3 (common to all four Conventions):

Persons taking no active part in the hostilities, including members of the armed forces who have laid down their arms and those placed hors de combat by . . . detention . . . shall in all circumstances be treated humanely[,] [and not subjected to] violence to life and person, in particular, murder of all kinds, mutilation, cruel treatment and torture[,] [nor to] outrages upon personal dignity, in particular humiliating and degrading treatment. [Emphasis added.]

This provision, it should be noted, does not literally compel the construction that has sometimes been placed on it. Indeed, it seems plausible that the phrase, “those placed hors de combat by . . . detention,” was intended only to modify “members of the armed forces”—i.e., lawful combatants belonging to the signatories' armies, some of whom may have been rendered hors de combat by detention. The fact that the provision is widely interpreted as extending anti-torture protection to nonsignatories almost certainly owes more to the general disrepute in which murder and torture are held by the civilized world than to any putative desire of the framers to create an entirely new category, otherwise left virtually unmentioned in the four Conventions.

But the bottom line is that, arguably, captured terrorists may not be lawfully subjected to torture. Is that the end of the matter?


To answer that question requires a look at more law: specifically, a convention developed more than a quarter-century after the 1949 Geneva Conventions took effect. This was a protocol to the Conventions, known as Protocol I Additional, dated June 8, 1977 and relating to the “Protection of Victims of International Armed Conflicts.”

The late 70's, one should recall, was the heyday of home-grown insurgencies and “national liberation” movements. These groups, in collusion with Western human-rights activists, pushed to extend Geneva protections to members of nonstate militias—often, guerrillas engaged in fighting colonial powers or other regimes backed by the West (especially the United States). Protocol I, as Rivkin and Casey have observed, was the fruit of that effort.

Not only does the accord grant POW status to such guerrillas, but it actually gives them a number of significant advantages over traditional nation-state armies. It allows them to maintain their privileged combatant status even if they have concealed their arms until very shortly before attacking, and it makes it unlawful to use force against them unless they are in the act of preparing an attack or attacking—thus permitting them to dictate the time and place of battle. Protocol I tips the scales even further by its rule of nonreciprocity, under which combatants in nonstate militias do not forfeit their privileged status even if they routinely violate international humanitarian law, which they are of course far more likely to do than are national armies.

It will come as no surprise that most of Europe—including England—signed on to Protocol I. The United States, however, did not, precisely because the protocol's loosening of traditional just-war strictures would have the effect of protecting, and thus encouraging, irregular forces like terrorist organizations that pose a lethal threat to civilian populations. This disagreement between the U.S. and its key allies on so rudimentary a matter as what rules should apply to the treatment of non-state combatants has caused no small number of problems in recent conflicts, including NATO's operations in the Balkans and coalition activities in Afghanistan and Iraq. The strains have induced Washington to fudge its opposition to Protocol I—formally opposing it but, for example, suggesting in such authoritative publications as the 2002 edition of the Army's Operational Law Handbook that much of its substance is “either legally binding as customary international law or acceptable practice though not legally binding.”

Unfortunately, this purposeful ambiguity has not only fostered uncertainty about what the law requires but has created a platform for domestic activists whose interventions have redounded to the benefit of terrorists at the very time we are most threatened. Thus, Kenneth Roth, the executive director of Human Rights Watch, warned in a December 2002 letter to President Bush that our treatment of al Qaeda detainees held at Guantanamo Bay could place the U.S. in violation of Protocol I. Skirting the inconvenient fact that the U.S. is not a signatory to the protocol, Roth deftly observed that the document “is recognized as restating customary international law”—which makes his position sound materially similar to the American Army's.

Still, ambiguity aside, Protocol I has never been ratified and as such is not binding on the United States. I hasten to add that this is immaterial as far as the Iraqis held at Abu Ghraib are concerned. Iraq is a party to the Geneva Conventions; its military personnel, apprehended in wartime, were POW's; and the U.S. conceded as much at the beginning of hostilities. Thus, even mild forms of abuse, much less the nightmarish indignities that occurred at Abu Ghraib, are violations of international law.

As for captured al Qaeda terrorists, however, none of this should be taken to suggest that it was, or is, improper to subject them to far more aggressive techniques of interrogation than can be applied to POW's. But torture? That is another matter.


Let me spell it out. It is illegal in the United States, under any circumstances, to torture anyone—even unlawful combatant terrorists who may have information about ongoing plots that, if revealed, could save thousands of lives. Period.2

There are two reasons for this, neither of them having to do with how one chooses to interpret the murky language of Article 3 of the 1949 Geneva Conventions (the one proscribing torture of some, or all, detainees). The first reason is that the United States is a signatory to an international treaty barring torture: this is the United Nations Convention Against Torture and Other Cruel, Inhuman, or Degrading Treatment or Punishment. The treaty is dated December 10, 1984 and was ratified in 1994, thus earning the force of binding law. The second reason is that federal law also categorically prohibits torture.

What constitutes torture under these laws? In Article 1 of the UN convention, torture is expressly and very broadly defined as any act, done at the direction or with the knowing acquiescence of a public official, by which

severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind.

What about exigent circumstances? What about a state of active combat in which causing severe “physical or mental” discomfort to a captive might elicit intelligence that will save the lives of troops? What about a war against a terrorist network in which nonlethal torture against a known mass murderer might induce him to reveal plans, say, to detonate a nuclear device in New York Harbor? The treaty turns a deaf ear. In Article 2, it states flatly: “No exceptional circumstances whatsoever, whether a state of war or a threat of war, internal political instability or any other public emergency, may be invoked as a justification of torture.” Not content with that, the convention dictates (in Article 16) that each signatory state must “undertake to prevent in any territory under its jurisdiction other acts of cruel, inhuman, or degrading treatment or punishment” even if they are not so severe as to “amount to torture as defined in Article 1.”

Pertinent questions arise here for Americans. What about the death penalty? What about when the police cause severe physical harm to a suspect during a disputed arrest? Well, such matters are indeed exempted from the treaty: its definition of torture excludes “pain or suffering arising only from, inherent in, or incidental to lawful sanctions.” More importantly for our purposes, the U.S. Senate, in ratifying the treaty, registered some stringent caveats, insisting on the preservation of capital punishment and limiting our acceptance of the proscription of “cruel, inhuman, or degrading treatment or punishment” to the relevant understandings enshrined in the “Fifth, Eighth, and/or Fourteenth Amendments to the Constitution.”

This was critical. As is well known, the Eighth Amendment bars “cruel and unusual punishments.” But, as Alan M. Dershowitz reminds us in his 2002 book, Why Terrorism Works, American courts, including the Supreme Court, have held that its protection extends only to those already convicted of crimes; that is, the “punishments” it regulates are those meted out after guilty verdicts. Similarly with the government's mandate under the Fifth and Fourteenth Amendments to provide “due process of law”: again, that term pertains almost exclusively to judicial proceedings and its substantive content is, to put it mildly, highly debatable.

Let us assume, for example, that the Abu Ghraib abuses were deemed to fall short of the “severe pain or suffering” needed to qualify as “torture.” Could a humiliated prisoner find recourse in the “cruel, inhuman, or degrading treatment” provision of the UN treaty? No, according to Dershowitz: his plight has nothing to do with judicial proceedings—he is being held by the military in wartime. Indeed, Dershowitz contends that the Senate's caveats limit the applicability not only of the treaty's “cruel, inhuman, or degrading” clause but even of its proscription against torture itself.3

Whatever qualifications may attach to the treaty, however, it cannot credibly be argued that torture is permissible—that is, beyond the reach of U.S. law—as long as it occurs outside the parameters of a judicial proceeding. That is because, fulfilling an aspiration expressed in the torture convention, the U. S. in 1994 enacted stringent anti-torture laws of its own.4 In some ways, these laws define torture even more broadly than the UN convention. They proscribe any act (other than those incident to lawful sanctions like the death penalty) that is “specifically intended to inflict severe physical or mental pain or suffering.” In other words, and contrary to the UN treaty's own slightly narrower definition, an act need not be motivated by a purpose to obtain information, to punish, or to intimidate for it to be considered torture in American law.

Nor is that all. Federal law also makes “severe mental pain or suffering” actionable with respect to a wide variety of menacing behavior. As long as mental harm is “prolonged,” it is grounds for torture charges if the harm results from the intentional or threatened infliction of severe physical pain or suffering; from the administration or threatened administration of “mind-altering substances or other procedures calculated to disrupt profoundly the senses or the personality”; from a “threat of imminent death”; or from threats that another person will imminently be subjected to such abuses. Not only those who engage in such behavior but those who conspire to engage in it, even if they are unsuccessful, are subject to prosecution.

If such behavior results in death, it may itself be punished by the death penalty; in other cases, it may be punished with up to twenty years' imprisonment. Nor does it matter where in the world the offense takes place; an American citizen may be prosecuted, and so may a foreign torturer who happens to be found on U.S. soil.5 As a matter of law, then, torture in this very expansive understanding is absolutely prohibited. The only question remaining is: should it be?


Just imagine, in this culture, having the temerity to say out loud: “I am in favor of torture.” One might as well declare oneself in favor of child molestation (or tobacco). In point of fact, however, many people—probably most people—who claim to be opposed to torture are not against it in all cases or in every form. Many, indeed, are no doubt secretly relieved that it goes on regardless of what the laws and the regnant pieties may dictate.

Seventy percent of Americans, all of whom presumably oppose killing, favor the death penalty. A comparably sizable number who oppose abortion favor its availability in cases of rape, incest, or where the life of the mother is at risk. All sensible people oppose the slaughter of innocent civilians, but an overwhelming number favor war if the evil it seeks to defeat is worth fighting against, even if war will ineluctably lead to the slaughter of innocent civilians.

Torture is not meaningfully different. Considered in a vacuum, it is a palpable moral evil. Moral evils, however, do not exist in a vacuum; they exist in collision with other evils, and sometimes we are forced to choose. Ask the average person if he opposes torture and the answer will surely be yes. But present him with a real-world scenario and the answer may well change.

Let us posit a terrorist, credibly believed to have murdered thousands of people. Suppose this terrorist is aware that a radiological bomb will be detonated momentarily in the heart of a major metropolis, but is refusing to impart the details to interrogators. Now, suddenly, black and white becomes gray: perhaps there are worse evils than some forms of torture. That does not mean our average person “favors” torture, but he may well be amenable to keeping it on the table as an option, and henceforth not so disposed to declare confidently that he opposes it in any form under all circumstances.

The obverse goes for the proponent of torture. As Dershowitz notes, arguments for torture are often crudely cast in terms of raw numbers: e.g., the torture of one guilty person would surely be justified to prevent the torture or death of a hundred innocent persons. But would it, without qualification? Are there lengths to which proponents might be unprepared to go? Of course there are. The vast majority of people who would favor torture in this case would oppose it if the form of abuse were inhumanly grisly; most would probably oppose a method that might lead to death (although that too might change if the situation were dire enough and especially if the person to be tortured were already facing execution). Moreover, most if not all people open to the rare application of torture in order to prevent terrorist acts would be immovably opposed if the torture involved abusing moral innocents—like harming a terrorist's children to induce him to talk—or in the absence of persuasive grounds for believing a terrorist attack was imminent.

All of which is to suggest that a significant gap exists between our wishes and reality. This is hardly unusual, either in life or in law. Take the canons of ethics that govern the behavior of attorneys. They are divided into two sections: ethical considerations and disciplinary rules. The former are the high principles that lawyers are enjoined to emulate; the latter are threshold commands of which they must not fall short on pain of sanction. So it is with torture. Even as we sincerely aspire never to resort to it, we are required to acknowledge that there are some instances in which it might be employed; therefore, it is incumbent upon us to regulate how and under what circumstances that could permissibly be done, and to prosecute aggressively those who step outside the parameters we undertake to define.


This conclusion, however reluctantly arrived at, is informed in part by the unpersuasiveness of the tactical and moral arguments parroted ubiquitously by the opponents of torture in the wake of Abu Ghraib. On the tactical level, it has been repeatedly asserted that torture simply does not work. The victim, it is said, has such a powerful incentive to tell the torturer what the torturer wants to hear that anything he says is by definition unreliable.

This claim might be made about countless other human endeavors in which inducements to act for ignoble reasons are similarly rampant, from politics to business to police work. It is thoroughly specious. A witness tempted to cooperate by the promise of a lenient sentence has a powerful motive to testify falsely to a version of events that he thinks might improve the prosecutor's case. A defendant who testifies in his own behalf has a different but no less strong incentive—namely, avoiding jail or perhaps death—to give a false exculpatory account. In neither situation is it necessarily the case that the witness will lie, or that, if he does, the lie will escape notice. Information from witnesses beset by intense pressure is inherently suspect but far from intrinsically unreliable.

Torture is no different. The victim may have been given an incentive to say what he thinks his torturer wants to hear, but that does not mean his compelled words are untrue. Nor does the victim necessarily even know what his torturer wants to hear. Nor can he know that he will not be tortured further if he intentionally misleads his tormentor. In fact, it could just as plausibly be asserted that torture is an ironclad guarantee of honesty as of misinformation. Neither statement holds water as an absolute, and, on the level of tactics, neither offers a sound reason either to permit or prohibit torture. As Dershowitz documents in detail, torture has been known to be a very effective method to get at truth; that it is not foolproof is hardly a reason to prohibit its selective use.

Then there is the moral argument: torture is an abomination so profound that permitting it, even if limited to rare and dire emergencies, constitutes an indelible blight on a society and its laws. So stated, the proposition has undeniable appeal. But “torture” is a loaded word. No one, it is fair to say, favors a policy of complete laissez-faire. What is envisioned instead is the administration of pressure that is capable of causing extreme pain—Dershowitz gives the example of sterile needles forced under the fingernails—but is nonlethal.

To be sure, even reading or thinking about such practices may make the teeth clench and the stomach churn. But consider: when sufficiently provoked, we already permit far worse. Capital punishment may be more humane than it used to be, thanks to lethal injection, but unlike torture it is forever. If we were to offer a choice—severe pain or execution—to convicts on death row, can there be any doubt which most of them would elect? Our weapons of war are “smarter” than they used to be, with precision targeting and the rest, but they still kill, mutilate, and maim with much less discrimination than we comfort ourselves to imagine. Is torture, with just cause and creating far less devastation, morally worse just because it is inflicted in a room looking the victim in the eye rather than from thousands of feet in the air where victims are unseen?

Equally significant from the moral point of view, one should think, are the consequences of the current system, in which we mouth our opposition to all torture while knowing full well that forms of it are occurring, with greater frequency than should be acceptable to anyone, inside many officially civilized countries that are signatories to the UN convention. It is here that Dershowitz is at his most trenchant: true civil libertarians are required to concern themselves with real-world outcomes rather than with proclaimed intentions, whether cynical or pure-hearted. By imposing an absolute ban on something we know is occurring, we promote disrespect for the rule of law in general and abdicate our duty to enact tailored and meaningful regulations. Both of these failings have the juggernaut effect of increasing the total amount of unjustifiable and otherwise preventable torture.


The task, then, is to create controlled, highly regulated, and responsibly accountable conditions. Toward this end, Dershowitz has proposed the notion of torture warrants. Under such a system, the government would have to apply to a federal court for permission to administer a predetermined form of non-lethal torture. The warrant would be issued only on a showing of reasonable grounds for believing that a catastrophe was impending, that the person to be subjected to torture had information about this event, that he had been given immunity (meaning, his statements could not be used against him in court and therefore he could not invoke the Fifth Amendment privilege against self-incrimination as a basis for refusing to answer), and that he had nevertheless remained silent.

Other conditions might be added. There could be limitations on who would be eligible for such treatment: for example, convicted terrorists or those who, even if not previously convicted, could be demonstrated to be terrorists according to some rigorous standard of proof. Application would have to be made with the approval of a high-ranking government official—the decision could not be entrusted to a twenty-three-year-old reservist assigned as custodian of a brig for mass murderers. Since torture would now be permitted, under stricture and with scrupulous judicial monitoring, no excuse would exist for engaging in torture outside the process, and those shown to have done so would be vigorously prosecuted.

In my estimation, this is a worthy proposal, far superior to the current hypocrisy that turns a blind eye to that which it purports to forbid. But my own approach, although based on the same underlying aims—minimizing the instances of actual torture and making its application more forthright, more transparent, and more accountable—is somewhat different. The struggle against militant Islamic terrorism, I believe, calls for an across-the-board rethinking of our current system as it relates not only to interrogations but also to the detention of unlawful combatants and to the trials of members of international terrorist networks, which our judicial system and its due-process standards are not designed to accommodate.

Elsewhere I have proposed the establishment of a national-security court, structured along the lines of the current Foreign Intelligence Surveillance Court.6 This would be a tribunal, drawn from the national pool of federal judges, that would have jurisdiction over, and develop an expertise in, matters of national security. It would monitor the detention of terrorist captives (deferring to the executive branch's decision to detain in the first place), and it would conduct trials of terrorists whom the government elected to charge under special rules that would apply only in national-security cases. This would replace the current paradigm in which the enlightened procedures of our criminal-justice system, designed to protect Americans accused of crimes and presumed innocent until proved otherwise, has been warped as we strain to apply them to terrorists in whose hands those same procedures imperil public safety.

Refining our consideration of torture would be of a piece with this proposed overhaul. With torture strictly limited to national-security matters, and unavoidably involving highly classified information about terrorist networks, the new national-security court would be the place to consider, and monitor the execution of, torture warrants. These could be sought, in turn, only by the Attorney General or by a high-ranking Justice Department official designated for this purpose by the Attorney General.

Centralizing this sensitive matter in a single court would ensure that the standards developed for warrants win rigorous adherence rather than (as in Der-showitz's proposal) being subject to tinkering by hundreds of federal judges in scores of districts throughout the country. It would also ensure speedy determinations—a critical point since torture would be permitted only under circumstances of imminent peril. Warrants would undoubtedly be rarely sought and rarely granted, but a judge who had previously dealt with even one would be in a far better position to decide quickly than a judge for whom this was uncharted territory. To bolster the new system's integrity and effectiveness, periodic reports would be made to the appropriate committees of Congress.

It goes without saying that amending our laws to permit limited, regulated torture would unleash torrents of obloquy throughout the world, not least in the Islamic countries and in Europe—two places, it must be observed, with shameful legacies of ruthless prisoner abuse. Nevertheless, our cause would be just: to demonstrate our seriousness about dealing with torture, to reduce its incidence, to make its practitioners accountable, and to ensure that in this new kind of war that is almost entirely about intelligence, we are not unduly deprived of information that may save thousands of lives.

Abu Ghraib presents an opportunity to deal with this wrenching issue in a practical, responsible, and honorable way. The wrong would lie in failing the summons.


1 “Unleashing the Dogs of War,” the National Interest, Fall 2003.

2 The very clarity of the proscription is what prompted controversy earlier this month when reports emerged that government lawyers had struggled unconvincingly in 2003 to craft ways around these laws.

3 Why Terrorism Works, p. 136. I respectfully disagree: the Senate's caveats apply only to the “cruel, inhuman, or degrading” clause, not to torture. See, for example, Senate's Advice & Consent (; and see also Douglass Cassel, “The United States and the Torture Convention: A Useful Dialogue,” Center for Human Rights Commentaries (May 24, 2000).

4 Codified under Title 18, U.S. Code, Sections 2340, 2340A & 2340B.

5 The 1991 Torture Victim Protection Act, codified under Title 28, U.S. Code, Section 1350, note.

6 “Abu Ghraib and Enemy Combatants,” National Review Online, May 11, 2004.


+ A A -
You may also like
Share via
Copy link