David Luban

When the U.S. Department of Justice’s secret torture memos were released in 2009, journalist Kathleen Parker wrote in the Washington Post:

“Several years ago, I asked a veteran journalist for advice. ‘I’m trying to figure out if I have an ethical conflict’, I began.

“‘If you have to ask, you do’, he said …”

Apply the same construct to torture. If we have to ask, it probably is.

Along the same lines, Jeremy Waldron observes that the prohibition on torture is not like a tax regulation, which needs precision because we expect even blameless taxpayers to push to the limits of the law. It is more like the prohibitions on domestic violence and sexual harassment, where you have no business demanding precise guidance on exactly how far you can go.

For Waldron, the prohibition on torture represents a legal archetype:

“a particular provision in a system of norms which has a significance going beyond its immediate normative content, a significance stemming from the fact that it sums up or makes vivid to us the point, purpose, principle, or policy of a whole area of law.”

The prohibition of torture is archetypal of the fundamental commitment of law not to rule through brutality or savagery. To paw through the law against torture looking for loopholes undermines that fundamental commitment of the law itself.

I agree with Waldron’s analysis and Parker’s rule of thumb. Indeed, I have argued at length that the fundamental dishonesty in the torture memos came from pretending that “torture,” “severe pain” and “cruel, inhuman, and degrading treatment” are purely technical concepts that demand minute legal line-drawing.

It was the fundamental trick that allowed the torture lawyers, quoting Kathleen Parker again, to “[torture] the English language trying to justify the unjustifiable.” Pursuing this line of thought, one might suspect that to demand a definition of torture is merely the opening gambit in a game of loophole lawyering. Why muddy the waters with definitions that invite pettifoggery?

One reason is that a definition of torture can help us identify what the most important evils associated with torture are. There are, after all, other reasons to define a concept besides sorting which items do or don’t fall under it. We define concepts to learn something about them by seeing how they hang together with our other concepts. That is a fundamentally philosophical aim, distinct from the legal definition of torture. It is my aim here.

The defining feature of torture, I will argue, is the use of pain or suffering as a communicative medium for displaying the absolute mastery of the torturer and the absolute helplessness of the victim. This non-legal definition – focusing not only on the painfulness of torture, but on its inextricable link with subjugation and humiliation – aims to identify the essential features that place torture among the greatest affronts to human dignity. It shows why torture violates a moral archetype.

Defining torture: a new start

The Convention Against Torture (CAT) provides the standard legal definition of torture, as the intentional infliction of severe mental or physical pain or suffering by an agent of the state. CAT also asserts that no war or emergency can ever justify torture – it places torture in an infernal class by itself. States have never said about killing anything remotely as categorical as CAT’s declaration that nothing justifies torture. After all, more than half the world’s people live in countries that have the death penalty. Killing in war is lawful, and even peacetime human rights treaties prohibit only “arbitrary” killings by the state, not all killings.

CAT’s definition of torture sheds no light on what makes torture unique among the misfortunes we visit on each other. For legal purposes, explaining that intuition is unnecessary, and too much theorizing would actually weaken the force of the prohibition by inviting controversy.

Thus, my observation that the legal definition sheds no light on why the ban is uniquely categorical is not a criticism of that definition. It does mean that to understand the evils of torture, we need to move past CAT’s definition. The definition centres on pain and suffering, but the evil of torture cannot be reduced to sensations alone. Consider this example: women whom no one would deem irrational often choose natural childbirth despite the pain. Its association with a joyful event transforms excruciating pain into an experience they do not regard as evil.

This example suggests that the evil of physical torture lies in the linkage between suffering and its context. Natural childbirth links pain with creation, love, and the miracle of human life. Torture links pain with fear, uncertainty and the horror of being wholly in the power of a maleficent enemy.

The fact that it is custodial torture matters crucially. Being in the clutches of the enemy and at his mercy – and understanding through the suffering he inflicts that he has no mercy – is an essential part of torture’s evil.

The Stoics recognized this special character of torture. Seneca observed in his Epistles that even the Stoic sage quails when he sees the instruments of torture displayed, and “the spectacle overcomes those who would have patiently withstood the suffering.” This last observation should be read carefully. Seneca means, quite literally, that the terror is more unbearable than the suffering itself. That is why displaying the instruments is actually the commencement of the torture. Seneca admits that diseases can be as painful as torture, but

“that which shakes us most is the dread which hangs over us from our neighbour’s ascendancy [literally ‘another’s power’, aliena potential]; for it is accompanied by great outcry and uproar … Picture to yourself under this head the prison, the cross, the rack, the hook, and … all the other contrivances devised by cruelty … It is not surprising that our greatest terror is of such a fate.”

Notably, Seneca emphasizes suffering amplified by imagination, and he associates the horrors of torture with being in another’s power and having that power put on spectacular display. But imagination does not merely amplify suffering, it transforms its content; that is, the way we experience pain or suffering – in this case, as absolute subordination to another person who uses pain to announce that subordination. Pain now becomes a medium of communication. Bearing these observations in mind, I shall define torture as follows:

Communicative Definition of TortureTorture of someone in the torturer’s custody or physical control is the assertion of unlimited power over absolute helplessness, communicated through the infliction of severe pain or suffering on the victim that the victim is meant to understand as the display of the torturer’s limitless power and the victim’s absolute helplessness.

I label this definition “communicative” because it focuses on the infliction of pain or suffering to announce, or communicate, the total subordination of the victim to the torturer. The communicative definition encompasses mental as well as physical torture. So long as the mental suffering is severe, and it is inflicted in a context that communicates the absolute dominance of the torturer and helplessness of the victim, it qualifies as torture.

The communicative definition is more complicated than the legal definition. I wish to reiterate that it is not intended to replace the legal definition, which serves its own purposes. Indeed, the communicative definition would be useless for legal purposes, if for no other reason than the near impossibility of proving in a court of law through ordinary evidence that an act is a communication of unlimited power over absolute helplessness using pain as a medium, and intended to be such.

The definition offered here means to capture a distinctive feature of torture, which lies in the connection of pain and suffering with subordination and helplessness as two sides of the same experience. The pain of torture is not only an experience in itself, it is a medium of communication; it is contentful pain – in this respect wholly unlike pain resulting from natural causes like illness or injury, whose noteworthy experienced characteristic is its senselessness, its lack of meaningful content. By contrast, what the pain of torture communicates is the absolute subordination of the victim to the enemy’s sovereign cruelty.

Making, taking, breaking

Does that mean that if the victim resists successfully – perhaps even laughing defiantly in the torturer’s face – the act is not torture, because the torturer has not absolutely dominated the victim? Of course not. An assertion of absolute dominance can fail to induce absolute dominance, just as a factual assertion can be disbelieved and rejected.

That observation highlights one important piece to this communicative model of torture. The torturer uses pain and suffering to send a message, but the aim is not merely to send the message. The torturer uses pain and suffering to communicate total domination in order to make it happen that he totally dominates the victim. The communication succeeds only when the victim takes up the message by believing it. What completes the making true is the taking as true. The victim, we may assume, will fight desperately to defy the message of total subjugation, using all the limited resources a helpless captive has available: will, courage and spirit (which are importantly different from each other); indignation and rage; and religious faith, if the victim has it.

These psychic resources may succeed for a while in resisting the torturer’s message, and therefore falsifying it and keeping the victim intact. But eventually they will almost certainly fail. Physical and mental pain exhausts the will, as everyone who has ever suffered prolonged pain or depression knows all too well. They discourage and dispirit the sufferer; they undermine the faith of the faithful. In the Gospels, the cross causes even Jesus to ask in anguish why God has forsaken him.

The torturer’s control over the victim’s pain and suffering automatically makes the torturer the most passionately interesting person in the universe to the victim, the one whose opinions are of greatest immediate significance. It matters crucially that the torturer has custody of the victim, cutting him off from any social contact that might contradict the message of radical hierarchy and radical worthlessness that torture sends. We take our cues from other people, and few of our beliefs, even perceptual beliefs, could survive perpetual contradiction by those around us. That holds especially of our belief in our own worth, which to an incalculable extent rests on the esteem in which we are held by others.

By cutting the victim off from any society other than the torturer’s, the torturer undermines the victim’s confidence in his own worth, which relies on the reinforcement of those around us. Losing confidence in their own worth, torture victims stand at risk of losing their final resource of resistance: anger and indignation at their treatment.

When the torturer succeeds, the victim takes on board the message of radical humiliation, and we then say that the victim is broken. Psychologists talk about it as “learned helplessness” and Stockholm syndrome, when the victim becomes docile and cooperative. It is the golden moment that interrogators aim at. Orwell brilliantly represents that moment in 1984, when Winston Smith betrays his lover Julia and in her place genuinely loves Big Brother. The broken Smith is, in an important sense, no longer fully human.

The evils of torture

One essential corollary of this definition is that the pain of torture cannot be separated, even conceptually, from the communication of absolute dominance over absolute helplessness. The victim experiences the pain not only as physical or mental sensations, but as a humiliation of the self at the hands of an infinitely cruel other.

It follows that the pain and suffering of torture cannot be understood reductively or “medically” as neural events of a specified intensity (“severe,” or merely “serious,” as one U.S. statute confusingly puts it, or not even serious), apart from the contents of the message they convey. Severity must be understood not as a quantitative measure of negative neural stimulation, but as an index of how hard it is to ignore or deny the message the pain or suffering communicates.

A second corollary is a set of consequences that yields a fuller understanding of the evils of torture. The definition centres on the communicative or message-sending function of custodial torture. We can elaborate seven specific evils (seven deadly sins?) that follow from the definition:

• the form of the message: inflicting mental or physical pain or suffering;
• the content of the message: “I, the torturer, am everything; you, the victim, are nothing; I am the master, you are the dog”;
• the means or manner by which torture sends the message: compelling the victim’s own senses and emotions to turn against her;
• the intended effect of the message: breaking the will, courage and spirit of the victim;
• the moral falsehood of the message of radical inequality and subordination, coupled with
• the self-fulfilling (or, in philosophical language, “performative” and “truth-making”) character of the message: by using pain and suffering to convey the message that the victim is broken, torture wears the victim down and makes it so;
• the totalitarian relationship torture establishes between the torturer and the victim, in which the torturer becomes sovereign over the most intimate sensations and emotions of the victim.

Notice that the legal definition of torture focuses only on the first item on this list – the infliction of pain or suffering. It rightly belongs first; obviously, the pain and suffering are the fundamental evil of torture. And, as I suggested above, there are good reasons for focusing the legal definition on the most tangible and provable of the evils. But noticing the multiple dimensions of evil curled up in the concept of torture is essential if we wish to understand why the world condemns torture so categorically.

The evils on this list are not the only basis for regarding torture as an affront to morality. For example, I have omitted from this list the crucial evil of torture that most concerns health professionals who treat torture survivors:

• the lasting aftereffects of the torture, including depression, PTSD, fearfulness and fundamental loss of trust in the world.

In tandem with these aftereffects, we can add – as an evil associated with the act of torture rather than the experience of torture:

• the torturer’s indifference to those lasting after-effects.

Both surely belong on the list of torture’s evils, but they do not follow specifically from the definition of torture offered here. They are collateral consequences of torture. The same is true of:

• the corruption of the torturer, as well as the physicians, psychologists, lawyers and other professionals connected with the torture enterprise, including military and law enforcement personnel and political leaders;
• the further corruption induced by the inevitable cover-ups; and
• the tendency of institutionalized torture to metastasize from cases in which it is supposedly “justified” to borderline cases, and then across the border – as the border itself gets redrawn because, as a consequence of torture’s corrupting effects, the baseline sense of what is normal and what is not changes in the minds and practices of those on the torture team.

Cataloguing these evils does not by itself prove that under no circumstances could torture ever be justified. It is always possible to invent imaginary cases in which all the evils summed together still turn out to be the lesser evil. But the character of evil associated with torture places it in a category similar to slavery: the category of practices about which the question “do the benefits outweigh the costs?” seems out of order – even if earlier civilizations saw nothing odd about the question and even assumed the answer was yes.

Torture – like slavery – belongs in the category of the unthinkable. This obviously does not mean it cannot be thought, only that such thoughts are radically at odds with central parts of our moral framework: most basically, assumptions about human worth and human equality.

Returning to Jeremy Waldron’s conception of a legal archetype, with which I began, we might say that just as the prohibition on torture is a legal archetype, the abolition of torture is a moral archetype, with significance beyond its immediate normative content. That is because to put the question of torture back into play compels us to put other central values into play as well: not only the rejection of cruelty, but also of totalitarian domination, along with belief in the indecency of humiliating and degrading other people, rejection of radical human hierarchy with broken men and women at the bottom, and the perverseness of creating a class of corrupt professionals whose job is to assist in all the other evils and then to cover it all up.

To reject the ban on torture would cast doubt on even minimum modern conceptions of human equality, human dignity, liberal government and even simple decency.

David Luban is University Professor in Law and Philosophy at Georgetown University.