Can Torture Ever Be Moral?

The recent Senate report on the Central Intelligence Agency’s use of torture has been the focus of a national debate about whether torture is ever permissible. This interview, the second in a series on political topics, discusses philosophical ideas that underlie this debate.  My interviewee is Jeff McMahan, White’s Professor of Moral Philosophy at the University of Oxford. He is the author of “The Ethics of Killing.” — Gary Gutting

Gary Gutting: What’s your overall view on the morality of torture?

Jeff McMahan: I think that torture is almost always morally wrong and that, for moral reasons, it ought to be prohibited absolutely in law. Torture has been used to extract confessions, to terrorize people associated with the victims, to punish presumed wrongdoers, and even to gratify and amuse sadists and bullies. These uses are always morally wrong. The only use of torture that has any chance of being morally justified is to gain important information. But even when torture is used to gain information, the torturers are usually wrongdoers seeking information that will help them to achieve their unjust aims. And even when those seeking information have just aims, their victims are often innocent, or lack the information sought, or are sufficiently strong-willed to mislead their torturers, so that the torture is ineffective or counterproductive. Still, both those pursuing unjust aims and those pursuing just aims will continue to be tempted to engage in torture if they can do so with impunity. Hence, torture has been widely practiced, though its use has almost invariably been wrong. This means that the overriding goal of the law ought to be to deter the wrongful use of torture, even at the cost of forbidding the use of torture in those rare cases in which it might be morally justified. The legal prohibition ought therefore to be absolute; for those who think that torture would be advantageous to them will always be tempted to try to exploit any legal permission to use it.

G.G.: But you do agree that torture can, in extreme cases, be moral. Why do you reject the absolute view that any instance of torture is immoral?

J.M.: Torture can be morally justifiable, and even obligatory, when it is wholly defensive – for example, when torturing a wrongdoer would prevent him from seriously harming innocent people. It could do that by forcing a person to reveal the location where he has planted a bomb, or hidden a hostage who will die if not found. It can be morally justifiable to kill a person to prevent him from detonating a bomb that will kill innocent people, or to prevent him from killing an innocent hostage. Since being killed is generally worse than being tortured, it should therefore be justifiable to torture a person to prevent him from killing innocent people. In cases in which torture is defensive in this way, the person tortured is not wronged. Indeed, he could avoid the torture simply by doing what he is morally required to do anyway – namely, disclose the location of the bomb or hostage.

G.G.: Do you worry that even saying that torture can be moral will provide an excuse for immoral torture?

J.M.: Yes, very much. The philosopher Henry Shue has a story of being thanked for his influential 1978 article [“Torture,” Philosophy and Public Affairs 7, no. 2 (1978): 124-43] by a pair of American agents who had tortured people. The article had argued vigorously against torture but conceded at the end that the moral prohibition of torture is not absolute. The agents were grateful for the concession, as that made them feel they could engage in torture without doing wrong. I think this is the explanation of why many people who aren’t absolutists about any other moral issue say they are absolutists about torture. They rightly want to avoid giving any aid or comfort to those who seek to justify torture in the circumstances in which it is actually practiced. But there is a dilemma here, for it can seem morally obtuse, and therefore discrediting, to deny that torture is permissible in those cases in which it obviously is permissible – for example, when it would in fact force a kidnapper to reveal the location of hostages who will otherwise die.

G.G.: Should we treat cases of justified torture the way some say we should treat cases of justified civil disobedience: You may, in extraordinary cases, have a moral right to disobey the law, but then you have to face the legal consequences?

J.M.: I think so. To effectively deter wrongful torture, the law should make anyone contemplating torture feel that if he does so he will be sacrificing himself for the sake of morality. It may indeed be best for the law not even to allow a necessity defense for torture, though I don’t think it should make harsh penalties mandatory. It should be possible for courts to exercise leniency in sentencing if there are cases in which people have engaged in torture with clear moral justification. But potential torturers should not be allowed to think that they can evade punishment through statutory loopholes.

G.G.: We’ve been using the term “torture” without defining it. Is it enough to work with clear cases of what is and of what is not torture (locking a prisoner in a cell versus beating him) or do moralists have to get into fine points about exactly what constitutes torture?

J.M.: Both moralists and legal theorists must go into the fine details. There are many reasons why paradigm instances of torture are objectionable: the sheer awfulness of suffering; the humiliation, terror, and dehumanization; the psychological scarring; the various forms of betrayal – of others, one’s ideals, and oneself – and so on. The moral evaluation of torture in a particular instance may depend on which elements are present and to what degree. Torture is not all or nothing: Some instances are worse than others, and at the lower end of the spectrum torture shades gradually into forms of harming that are horrible but do not rise quite to the level of torture. It is one of the problems of the absolutist view of torture that it has to identify some threshold on the scale that measures the elements of torture, such as suffering, and then claim that nothing, not even the prevention of a billion murders, can justify the infliction of that degree of harm, even on a wrongdoer. But the view does not absolutely prohibit the infliction of the highest degree of harm below the threshold. It has to concede that the infliction of that degree of harm can be permissible, even to prevent harms far less bad than the murder of a billion people. The idea that there is such a threshold is wholly implausible.

G.G.: Why do the absolutists have to specify a precise point at which the infliction of pain becomes torture? Why can’t they admit that there are gray areas but still maintain that, in clear cases, torture is always immoral?

J.M.: If you claim that a certain type of act is wrong no matter what the consequences, I think you must specify the precise conditions for an act to be of that sort. But even if absolutists can claim that there are gray areas in which it is indeterminate whether an act constitutes torture, or in which we are uncertain whether the act is torture, they still have to say whether an act that is in the gray area can be permissible. If there is uncertainty, there is a chance that in fact nothing could possibly justify the act. A principle of reasonable caution suggests that we should treat such an act as absolutely prohibited. But if that is so, we have to know at what exact point the gray area begins and claim that all acts at or beyond that point may not be done, whatever the consequences. Or, if the absolutist accepts that acts in the gray area can be permissible, we then need to know exactly where the gray area ends and absolutely prohibited acts begin. So the suggestion that there might be a gray area does not solve the problem of the precise threshold but merely pushes it back a bit.

G.G.: Absolutists might object that you’re just assuming that actions should be judged by their consequences, whereas they think at least some actions are immoral in themselves, apart from their consequences. Is this just a case of conflicting basic moral intuitions, with no way of resolving the issue?

J.M.: I don’t think so. Thus far I have criticized absolutism about torture on the ground that it has to draw a sharp line between acts that are prohibited, regardless of their consequences, and acts that may or may not be prohibited depending partly on their consequences. But I don’t think that only the consequences matter. I have said that it might be permissible to torture a terrorist to force him to reveal the location of a bomb or a hostage, but that would be quite different from torturing the terrorist’s child as a means of extracting the same information. In planting the bomb or capturing the hostage, the terrorist makes himself morally liable to be harmed as a means of preventing him from harming innocent people. But his child has done nothing to become liable to be tortured as a means of saving the parent’s potential victims.

The most important objection to the claim that all torture is absolutely prohibited is that it does not distinguish between the torture of wrongdoers and the torture of the innocent. The British philosopher Elizabeth Anscombe was a moral absolutist but she appreciated that absolute prohibitions have to take account of whether intended victims are innocent or non-innocent. Thus, while she believed that the intentional killing of innocent people is absolutely prohibited, she was scornful of absolute pacifism, which, she argued, “teaches people to make no distinction between the shedding of innocent blood and the shedding of any human blood…and in this way…has corrupted enormous numbers of people.” Absolutists about torture, who say that it can never be justified, make the same mistake. As I indicated earlier, because most of them believe that it can be permissible to kill a person to prevent him from committing murder and also that it can be less bad for a person to be tortured than to be killed, they should concede that it can be permissible to torture a person to prevent him from committing murder. Apart from the fact that killing is usually worse, the only significant difference between killing in defense of the innocent and torturing in defense of the innocent is that torture can only very, very rarely be used defensively.

G.G.: As a moral philosopher, how do you view the recent controversies about the Senate report on the C.I.A.’s use of torture?

J.M.: It is reasonable to demand that a democratic government act in ways that it can justify to its citizens. The Bush administration lied to its citizens about its torture policy, which was in violation of international law and therefore also in violation of American law, which treats international law as the law of the land. Members of that administration have not shown that any of the many instances of torture in which Americans were involved were morally justified in the way I have suggested that they might in principle be justified. Apologists for torture also cannot point to any terrorist atrocities since Obama abolished the torture policy that might have been prevented had the policy been continued. Given that the policy has been abandoned without any apparent ill effects, there seems to be no reason of national security to continue to conceal facts about the Bush administration’s torture practices. There are, moreover, democratic reasons for exposing them – for example, to deter future administrations from acting in similar ways and to help to dispel illusions that citizens have about the honesty and trustworthiness of their government. Some defenders of torture are outraged by the exposure of the Bush administration’s clandestine criminal acts but not by the fact that these acts were committed. That attitude seems to me fundamentally anti-democratic.

This interview was conducted by email and edited. Other installments in this series can be read here.

Gary Gutting is a professor of philosophy at the University of Notre Dame, and an editor of Notre Dame Philosophical Reviews. He is the author of, most recently, “Thinking the Impossible: French Philosophy since 1960,” and writes regularly for The Stone.

Follow The New York Times Opinion section on Facebook and on Twitter, and sign up for the Opinion Today newsletter.