It is difficult to think about torture honestly. In a recent article on the interrogation techniques employed by the US, the writer Mark Bowden observed that few "moral imperatives make such sense on a large scale, but break down so dramatically in the particular." The moral imperative—do not torture, any time, anywhere, in any circumstances—is mandated by the UN convention against torture and other cruel, inhuman or degrading treatment or punishment. "No exceptional circumstances whatsoever, whether a state of war or a threat of war, internal political instability or any other public emergency," says the convention, can "be invoked as a justification of torture." That terrorists themselves torture does not change these imperatives. Our compliance does not depend on reciprocity.
As long as we stay on this high ground of unconditional prohibition, we seem to know where we are. Problems begin when we descend into the particular, when we ask what exactly counts as torture.
Since no state wants to be seen as torturing suspects but all states want to be able to extract information to protect their citizens, the key question is whether states can use methods of "coercive interrogation" that do not qualify as torture.When the torture convention was ratified by the US Senate in 1994, maintaining a meaningful distinction between coercive but lawful interrogation and outright torture was a central concern. The Senate ratified the convention on the understanding that torture should be reserved for "severe physical or mental pain or suffering" resulting in "prolonged mental harm." Once the war on terror began, the parsing of the convention went still further. In the now notorious memos submitted by the office of legal counsel to the White House in 2002, these definitions were stretched to the point that the threshold for torture "must be equivalent in intensity to the pain accompanying serious physical injury, such as organ failure, impairment of bodily function or even death." Any physical abuse below that standard counted as "coercive interrogation." Some forms of coercive interrogation, the lawyers admitted, might not be torture, but they would still be defined as "inhuman and degrading treatment."
There is thus a conceptual and practical distinction between torture and coercive interrogation. There is a further distinction—at least in theory—between methods of coercive interrogation that are lawful and permissible and those that may be inhuman and degrading. While this distinction exists in theory, most human rights activists would deny that such a distinction can be observed in practice.
Human rights activists accept that reliable information is essential for combating terrorists and that interrogation is a central feature of any counterterrorist strategy. Kenneth Roth, of Human Rights Watch, argues that "respect for the Geneva conventions does not preclude vigorously interrogating detainees about a limitless range of topics." What work is the word "vigorously" doing in this sentence? It is intended to make it clear that a human rights defender takes seriously the necessity of getting from detainees real information that may prevent future terrorist attacks. But what, in specific terms, might "vigorous" interrogation actually entail? Clearly, Roth and anyone else who cares about human rights wants to exclude any form of abuse. But what exactly counts as abuse in a "vigorous" interrogation?
In order to prevent vigorous interrogation from slipping down any slope, human rights activists want to collapse the distinction between "coercive interrogation" and "torture," and to ban any physical or psychological coercion. But there is a significant distinction between the two. As legal theorist and federal judge Richard Posner has argued, "almost all official interrogation is coercive, yet not all coercive interrogation would be called 'torture' by any competent user of the English language." As the political philosopher Jean Bethke Elshtain writes, "when human rights groups label 'unpleasant or disadvantageous treatment of any kind' torture… they fail to discriminate between cases," for example, between "sleep deprivation and amputation or burning or some other horror."
Clear thinking about torture is not served by collapsing the distinction between coercive interrogation and torture. Both may be repugnant, but repugnance does not make them into the same thing. If coercion and torture are on a moral continuum, at what point on the continuum, to use Posner's words, does queasiness turn to revulsion? Vigorous interrogation might mean lengthy, exhausting, harassing exchanges with interrogators. Provided that there was no physical contact between interrogator and subject, no deprivation of food or water harmful to health, this might qualify as lawful interrogation. But at every ratchet of coercion, moral problems arise. Sleep deprivation will not leave physical or permanent psychological scars, but as Menachem Begin, who was interrogated in Soviet Russia, remembered, "anyone who has experienced this desire [for sleep] knows that not even hunger or thirst are comparable with it."
It might be lawful to deceive a subject under interrogation, by stating that all of his associates are already in detention when they are still at large. But other forms of deception can inflict excruciating psychological anguish. Threatening a subject with the imminent death or torture of those dearest to him may not leave any physical marks, but it rightly can constitute torture, not just coercion, in even the US Senate's definition. Both Elshtain and Posner have argued against the moral perfectionism that elides the distinction between coercion and torture, and have stressed the cruel, if regrettable, necessity of using coercive methods on a small category of terrorists who may have information vital to saving the lives of innocent people. Posner justifies coercive interrogation on utilitarian grounds: saving the lives of many counts more, in moral terms, than abusing the body and dignity of a single individual. Elshtain justifies coercive interrogation using a complex moral calculus of "dirty hands": good consequences cannot justify bad acts, but bad acts are sometimes tragically necessary. The acts remain bad, and the person must accept the moral opprobrium and not seek to excuse the inexcusable with the justifications of necessity.
My own work on "lesser evils" brings me close to the Elshtain position. I agree with her that necessity may require the commission of bad acts, which necessity, nevertheless, cannot absolve of their morally problematic character—but I still have a problem. If one enumerates the forms of coercive interrogation that have been judged to be inhuman and degrading by the Israeli and the European courts—hooding, holding subjects in painful positions, exposing them to cold or heat or ear-splitting noise—these techniques also seem unacceptable, though at a lower threshold of awfulness, than torture. Like Elshtain, I am willing to get my hands dirty, but unlike her, I have practical difficulty enumerating a list of coercive techniques that I would be willing to have a democratic society inflict in my name. I accept, for example, that a slap is not the same thing as a beating, but I still don't want interrogators to slap detainees because I cannot see how to prevent the occasional slap deteriorating into a regular practice of beating. The issue is not, as Elshtain implies, that I care overmuch about my own moral purity but rather that I cannot see any clear way to manage coercive interrogation institutionally so that it does not degenerate into torture.
On the issue of regulation, there are those—Alan Dershowitz, for example—who believe that banning torture and coercion outright is unrealistic. Instead, the practice should be regulated by court warrants. But judicialisation of torture, and of coercive interrogation techniques involving stress and duress, physical abuse, sleep deprivation and so on, could lead to torture and coercion becoming routine rather than an exception. A position in favour of outright prohibition of both torture and coercive interrogation has gained strength from the abuses at Abu Ghraib, and from the memos of the office of legal counsel and the White House parsing the torture convention into permission for coercive interrogation. It seems clear from the dire experience of Abu Ghraib that outright prohibition of both torture and coercive interrogation is the only way to proceed. Rules for interrogations, with penalties in the uniform code of military justice, should be mandatory.
Absolute prohibition, however, is easy. Enforcement is hard, and even rules and punishment for infraction are not enough. The crucial element for enforcement of rules and procedures against abuse of detainees is habeas corpus, the legal requirement of any detaining power in a democracy to produce detainees before a court of law and justify detention to a duly appointed legal authority. As long as the US—or any state, for that matter—has the power to detain at pleasure and in secret, abuse of detainees is inevitable. International pressure, domestic mobilisation and, finally, congressional legislation are all necessary to stop the practice of "ghost detainees," whose identities remain concealed and who may be held outside the US, inside the US, or in third countries. It should be mandatory that every single detainee held by the US, whether a citizen or not, be publicly known. If operational necessity—keeping the enemy from knowing who is in custody—requires secrecy, disclosure of their names to congress and the courts can be undertaken in camera. It should also be mandatory that every detainee of the US, whether citizen or not, whether held onshore or offshore, should have habeas corpus access to a federal court, together with the legal capacity to make representations to that court about treatment and detention.
I am not so naive as to suppose that federal court review of detention will always provide effective remedies for detainees. But evidence of the impact of recent supreme and federal court rulings on the tribunal review process at Guantánamo, and on ordinary treatment of the detainees, does suggest that court review and access, however imperfect, is the only reliable way to keep detention under the rule of law.
So I end up supporting an absolute and unconditional ban on both torture and those forms of coercive interrogation that involve stress and duress, and I believe that enforcement of such a ban should be up to the military justice system plus the federal courts. I also believe that the training of interrogators can be improved by executive order and that the training must rigorously exclude stress and duress methods.
Two significant problems remain. First of all, there is the problem of the exceptional case, one where lives can be saved by the application of physical methods that amount to torture. "Ticking bomb cases" cannot be wished away. They might arise especially where an American or European city faced the threat of WMD. An outright ban on torture and coercive interrogation leave a conscientious security officer with little choice but to disobey the ban. In this event, as the Israeli supreme court has said, even a conscientious agent acting in good faith to save lives should be charged with a criminal offence and be required to stand trial. At trial, a defence of necessity could be entered in mitigation of sentence, but not to absolve or acquit. This is the only solution I can see that remains consistent with an absolute ban on torture and coercive interrogation. Let us not pretend that the enforcement of this rule would be easy. Where the threat could be shown to be genuine, it seems evident that few legal systems would punish such a conscientious offender. So an outright ban on torture creates the problem of the conscientious offender. This is a small price to pay for a ban on torture.
Does an outright ban on torture and coercive interrogation meet the test of realism? Would an absolute ban on torture and coercive interrogation using stress and duress so diminish the effectiveness of our intelligence-gathering that it would diminish public safety? It is often said—and I argued so myself—that neither coercive interrogation nor torture is necessary, since entirely lawful interrogation can secure just as effective results. There must be some truth to this. Israeli interrogators have given interviews assuring the Israeli public that physical duress is unnecessary. But we are grasping at straws if we think this is the entire truth. As Posner and others have tartly pointed out, if torture and coercion are both as useless as critics pretend, why are they used so much? While some abuse and outright torture can be attributed to individual sadism, poor supervision and so on, it must be the case that other acts of torture occur because interrogators believe, in good faith, that torture is the only way to extract information in a timely fashion. It must also be the case that if experienced interrogators come to this conclusion, they do so on the basis of experience. The argument that torture and coercion do not work is contradicted by the dire frequency with which both practices occur. I submit that we would not be "waterboarding" Khalid Sheikh Mohammed—immersing him in water until he experiences the torment of nearly drowning—if our intelligence operatives did not believe it was necessary to crack open the al Qaeda network that he commanded. Indeed, Mark Bowden points to a Time report in March 2003 that Sheikh Mohammed had "given US interrogators the names and descriptions of about a dozen key al Qaeda operatives believed to be plotting terrorist attacks." We must at least entertain the possibility that the operatives working on Sheikh Mohammed in our name are engaging not in gratuitous sadism but in the genuine belief that this form of torture—and it does qualify as such—makes all the difference.
If they are right, then those who support an absolute ban on torture had better be honest enough to admit that moral prohibition comes at a price. It is possible, at least in theory, that subjecting interrogators to rules that outlaw torture and coercive interrogation, backed up by punishment if they go too far, will create an interrogation regime that allows some interrogation subjects to resist divulging information and prevents our intelligence services from timely access to information that may save lives.
If there is a significant cost to an outright ban on coercive interrogation and torture, what can possibly justify it? Many of the arguments that human rights activists make in justification amount to the claim that torture shames their moral identity as human beings and as citizens, and that they do not wish such acts to be committed in their names. Other citizens in a democracy may not value their own moral scruple over the collective interest in having accurate security information, even if collected by dubious means. It may be obvious to human rights activists how to adjudicate these claims, but it is not obvious to me. That is, I do not see any trumping argument on behalf of the rights and dignity of security detainees that makes their claims prevail over the security interests (and human right to life) of the majority. The best I can do is to relate the ban on torture to the political identity of the democracies we are trying to defend—by claiming that democracies limit the powers that governments can justly exercise over the human beings under their power, and that these limits include an absolute ban on subjecting individuals to forms of pain that strip them of their dignity, identity and even sanity.
We cannot torture, in other words, because of who we are. This is the best I can do, but those of us who believe this had better admit that many of our fellow citizens are bound to disagree. It is in the nature of democracy itself that fellow citizens will define their identity in ways that privilege security over liberty and thus reluctantly endorse torture in their name. If we are against torture, we are committed to arguing with our fellow citizens, not treating those who defend torture as moral monsters. Those of us who oppose torture should also be honest enough to admit that we may have to pay a price for our own convictions. Ex ante, of course, I cannot tell how high this price might be. Ex post—following another terrorist attack that might have been prevented through the exercise of coercive interrogation—the price of my scruple might simply seem too high. This is a risk I am prepared to take, but frankly, a majority of fellow citizens is unlikely to concur.
© Michael Ignatieff, 2006. Extracted from "Torture: Does it Make Us Safer? Is it Ever OK?," edited by Kenneth Roth and Minky Worden, £12.99, published by The New Press in conjunction with Human Rights Watch, 2006
A TIMELINE OF TORTURE
- 1948 Universal declaration of human rights states that "no one shall be subjected to torture, or to cruel, inhuman or degrading treatment."
- 1950 Third Geneva convention outlaws use of torture on prisoners of war.
- 1987 UN convention against torture comes into force. Ratified by Britain in 1988 and the US in 1994.
- 1998 The European convention on human rights, which prohibits torture under all circumstances, is incorporated into British law.
- 2002 US department for justice advises the CIA that torturing al Qaeda prisoners "may be justified" and that international laws banning torture "may be unconstitutional if applied to interrogations."
- 2004 Leaked Red Cross report into American abuse of detainees in Iraq claims that some treatment was "tantamount to torture."
- 2004 First photos of detainee abuse at Abu Ghraib prison released and published.
- 2005 Law lords rule that evidence that may have been obtained through torture is inadmissable in British courts.
