Norway and the United States have been NATO allies for nearly 70 years. Our troops trained together to defend against invasion by Soviet and Warsaw Pact forces, and have fought and died together opposing Al Qaeda and the Taliban in Afghanistan. We continue to face common threats from people covertly directed or inspired by ISIS, as well as “lone wolves” who murder ostensibly in the service of Christianity or racial supremacy. No doubt our respective intelligence agencies share considerable intelligence about such threats, but some challenging ethical issues regarding the use of human agents (HUMINT) can accompany that cooperation.
In the wake of the devastating 9-11 attacks in the U.S., many government officials decided that at least some prior restraints on the conduct of intelligence and military personnel should be modified or even abandoned, in order to obtain information that might be vital to preventing another terrorist strike. CIA clandestine service officers, military intelligence and special forces, and civilian contractors were given permission to use more forceful and aggressive tactics than had previously been allowed, including interrogation methods intended to produce acute pain and fear, and renditions of detainees to countries notorious for using torture, all of which are problematic under international treaties to say the least.
In this article I will examine the following factors bearing on intelligence interrogation techniques: 1) Non-moral questions about the effectiveness of torture in interrogations, e.g., does it ever “work” in the sense of producing accurate intelligence? 2) What does the law require? What have we pledged to do in relevant international treaties? 3) If the law were silent on torture, or if we were to reconsider our treaty obligations, how should we sort out the moral rights involved? Is a right not to be tortured absolute, or something less strict than that? 4) What would be the probable consequences of legalizing torture in intelligence interrogations? Would the likely harms from that outweigh its potential benefits?
I’ve published several articles on this subject before, most recently a chapter in my book Partly Cloudy: Ethics in War, Espionage, Covert Action, and Interrogation, 2nd ed. (2016). Readers may want to consult that chapter for references and commentary that I’m unable to include in the present article due to space constraints.
Absolute vs. Prima Facie Moral Obligations
In our contemporary efforts to combat terrorists, we face some difficult ethical questions: Do ruthless enemies merit or deserve ruthless countermeasures? Should we uphold high ethical standards even against unlawful combatants who don’t respect them? Or, in order to preserve what we value most, are we morally justified in using tactics (such as torture in interrogations) that strain or even contradict those values?
A potential way to resolve such questions would be to show that at least one of the relevant ethical principles is absolute, i.e., admitting of no exceptions and trumping all other moral considerations. Are there any such absolutes?
Some readers might reply, “Yes, there are absolute moral principles, namely those commanded by God.” In theory every person might be able to hear the voice of God directly, in which case presumably everyone would hear the same general moral commands, or at least ones that were logically consistent. But religious believers often disagree among themselves, not only between religions but also within the same religious tradition, on how one comes to know those commands and what they specifically require. And of course individuals can be mistaken about what God might saying to them directly: today we know that some of those errors can be due to wishful thinking, perceptual biases, etc.
Most of the faithful in Christian, Jewish and Muslim communities in fact tend to place greater trust in their sacred scriptures than in individual revelation. But those scriptures are often inconsistent guides to action, even on a question as fundamental as whether it’s ever right to kill people. To illustrate, if a Christian were trying to assess the ethics of killing, should he or she follow the teachings and example of Jesus who seemed to advocate strict nonviolence, or some Old Testament verses that urged strict retributive justice (including capital punishment), or Joshua who supposedly slaughtered every last inhabitant of Jericho in obedience to God’s order? Hence, religious believers cannot escape the necessity of wrestling with ethical challenges without strict or exclusive reliance on scriptural commands or stories. (For comparative analysis of relevant precepts in Hindu, Buddhist, Jewish, Christian and Muslim traditions, see chapter two of my book Partly Cloudy.)
Some secular philosophers have attempted to unify all moral obligations under one overarching principle or procedure. For example, utilitarian ethical theorists rely exclusively on a teleological/consequentialist principle, seeking the greatest balance of beneficial over harmful consequences among alternative actions or policies; while Immanuel Kant’s theory is strictly deontological/nonconsequentialist, testing every potential moral rule as to whether it can be universalized without logical contradiction. Both utilitarians and Kantians also claim absolute status for their fundamental principles.
The strengths of utilitarian theory lie in its consideration of the well-being of all sentient beings potentially affected by proposed actions, and its goals of ameliorating suffering and enhancing happiness. The chief virtue of Kantian ethics is its respect for individual human autonomy, dignity and worth. Both of those theories have been highly influential, for good reasons. But many other ethicists have identified serious problems in them: utilitarianism is especially vulnerable to concerns about justice and basic rights, while Kantian ethics tends to undervalue the importance of compassion.
A more promising approach was developed by W. David Ross, an important 20th-century British philosopher. He proposed a mixed/pluralistic theory that sees consequentialist, nonconsequentialist and aretaic concerns as important to consider in making moral decisions. Ross argued in The Right and the Good (1930) that there are no absolute moral principles; rather, there is a cluster of prima facie duties, each of which has moral weight and may take precedence over others in different situations. Often our prima facie duties will reinforce one another, e.g., in ruling out cruel or purely egoistic actions. But when our obligations do conflict, there’s no way to establish for all time which will take precedence: Ross believed that we must simply wrestle with every duty relevant to a particular situation and determine which is most weighty, i.e., which prima facie duty is our actual duty then and there. In some cases one’s paramount moral duty will be to promote happiness; in others, to prevent or alleviate harm; in others, to protect rights, etc. The need for moral deliberation and wisdom is simply part of the human condition, in Ross’s view.
The fact that one principle can give way to another does not mean that the less weighty one loses significance entirely. For example, deciding that I must break a promise to meet a friend, in order to save the life of an accident victim whom I encounter on the way, does not mean that my promise to my friend loses all moral value; I still regret breaking the promise even when it’s the right thing to do. But Ross’s principles apply universally, across cultures and apart from individual differences in preferences and tastes; Ross was not a subjective or cultural relativist.
I think that Ross was right in believing that our general moral obligations are a mixture of consequentialist, non-consequentialist and aretaic ones, and that there’s no clear hierarchy among prima facie duties that would apply to every possible ethical dilemma. But there may nonetheless be some absolute moral rules. Consider “Don’t rape,” and “Don’t torture children or animals”: there don’t seem to be any credible exceptions to those rules, where they could justifiably be overridden by more important duties. (Of course, we’d need to define rape and torture in morally neutral ways if the rules I’ve suggested are to avoid begging the question.) The jus-in-bello rule of noncombatant immunity also seems to be a very good candidate for absolute status.
However, it’s often very difficult to state other rules that aren’t vulnerable to counterexamples: Consider “Don’t kill.” – Never in self-defense? Never in defense of other innocent people? Or consider “Don’t lie” – Even to save lives or to prevent other serious harm? Such counterexamples suggest that general rules against killing and lying are not absolute, a point, which has obvious relevance to military and intelligence operations. (Later I’ll explore the question, is “Don’t torture” an absolute duty?)
Principles can sometimes be strengthened, though, by incorporating exceptional cases into them. E.g., we might modify “Don’t lie” in this way: “It’s wrong to deceive people, unless a) they have forfeited their right to know the truth, or b) they lack the ability to make rational decisions, and telling them the whole truth would clearly hurt them more than it would help them.” Of course, these specific exceptions are controversial, and it would take more complicated arguments to support them.
Does Torture Work?
The empirical, non-moral question of whether torture works as a method of intelligence interrogation is obviously important to the moral controversy, since if torture never worked then it would be silly to use it in HUMINT. Unfortunately, experts are not entirely united on the empirical question of whether torture can be effective.
On the one hand, historical studies (e.g. by John Langbein) of the practice of torture show that it’s a highly unreliable way of producing accurate information. A CIA (“KUBARK”) interrogation manual written in 1963 asserted, “Intense pain is quite likely to produce false confessions, concocted as a means of escaping from distress,” and former CIA case officer Glenn Carle wrote (in The Interrogator, 2011), “Physical coercion—torture—has nothing to do with a useful interrogation. Torture (…) does not work. It is patently stupid, an offense to any understanding of how the mind works.” The U.S. military and FBI have taught their interrogators for decades that torture is not effective in eliciting truthful statements. And recent experiments with volunteers on the effects of isolation and sensory deprivation—conditions forcibly inflicted on some of our country’s detainees since 9-11—indicate that significant memory loss and suggestibility can occur after only 48 hours, meaning that statements made by detainees subjected to such treatment are likely to be unreliable and misleading, even if they intend by that point to cooperate with their captors.
A detailed memoir called The Interrogators (2004), written by an Army non-commissioned officer who served in military intelligence in Afghanistan, suggests that the most reliable interrogation methods are those echoing classic police and detective work: building rapport with the subject to foster trust; offering incentives for cooperation; gathering independent evidence without the suspect’s knowledge; identifying discrepancies between the suspect’s story and the stories of his accomplices and other witnesses; deceiving the suspect into thinking you know more than you do; and catching the suspect in lies that he can’t sustain. Such methods can be effective even against committed terrorists, and have the added virtue of not crossing the line into torture.
Former FBI agent Ali Soufan described in his book Black Banners (2011) several highly productive interrogations that he conducted of Al Qaeda and other terrorist leaders that employed similar techniques utterly free of violence, threats, or other cruel and inhumane treatment. Soufan further claimed that when other interrogators began using harsh tactics against many of those same detainees, they stopped cooperating and frequently invented false stories.
On the other hand, the 1963 CIA interrogation manual suggested that physical coercion as well as sensory deprivation and verbal threats can sometimes be effective against a recalcitrant subject. It quoted a separate study indicating that “most people who are exposed to coercive procedures will talk and usually reveal some information that they might not have revealed other- wise.” There is also anecdotal evidence that torture has occasionally been effective in eliciting useful intelligence: a Sri Lankan army officer claimed that torture induced some Tamil Tiger detainees to reveal details about planned terrorist acts, and torture seems to have enabled Lebanese and CIA officers to identify who blew up the U.S. Embassy in Beirut in 1983.
But even if those claims are true, that obviously doesn’t prove that torture is reliable always or even most of the time, or that humane interrogation tactics aren’t more reliable in eliciting vital information. Like doctors who continue to use treatments they’re comfortable with well after they’ve been proven scientifically to be less effective than other treatments, some intelligence officers might persist in practicing torture due to laziness, lack of creativity, or willful ignorance of the relative advantages of more humane methods. In a devastating report published in 2014, the U.S. Senate Select Committee on Intelligence concluded convincingly that none of the harsh interrogation techniques adopted by the CIA after 9-11 produced any useful intelligence about terrorist plots that hadn’t already been revealed by more humane methods.
Breaking someone’s will to resist via waterboarding, e.g., doesn’t necessarily mean that everything they say after that point will be trustworthy. Indeed, the more acutely painful the torture is, the more likely the victim would be to say anything (including lies) to end their ordeal.
But then, no method of interrogation is guaranteed to work with every subject. And if intelligence personnel have already tried less questionable methods on a suspected terrorist, without success, they might see torture as a last resort, a technique that might just work when other means have failed. To me it’s highly doubtful that torture would ever be necessary, let alone more effective than humane interrogation methods, in producing timely and reliable intelligence. But if we temporarily ignore moral and legal concerns, it’s understandable if some interrogators might want authorization to use it as a potentially useful HUMINT tool, especially when the stakes are very high and time is of the essence, as in a “ticking bomb” situation or its analogues.
So, if we assume for now that torture might at least occasionally work (even if overwhelming empirical evidence indicates otherwise), what do the relevant laws require?
Legal Restrictions on Coercive Interrogation
Since the law can sometimes permit or require unethical actions, such as racial discrimination under “Jim Crow” or South African apartheid, it can sometimes be ethical to break the law. But a prima facie obligation clearly exists to respect and obey laws that have been established by legitimate political bodies, including treaties ratified by representative governments like ours.
This much is clear: contemporary international treaties prohibit torture as well as inhumane and degrading treatment of detainees or prisoners. The prohibition stated in the U.N. Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (CAT) is clear, comprehensive, and absolute. Torture is defined in the Convention as “any act 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 (…)”. Moreover, the treaty permits no exceptions to its prohibition of torture, even for situations where government officials suspect that a detainee has information that could prevent the loss of many innocent lives.
After 9-11, some of the President Bush’s legal advisors claimed that Common Article 3 of the Geneva Conventions (which also forbids torture and inhumane treatment) did not apply to Al Qaeda suspects. And some of those same legal advisors subsequently defined “torture” under the CAT in a ridiculously narrow way, and the President’s authority as Commander in Chief in a ludicrously expansive way, suggesting that in that role he has the constitutional authority to override federal statutes. As a result, some interrogation techniques that were previously considered illegal were approved by senior U.S. officials, including stress positions, sleep deprivation, face- slapping, removal of clothing, exposure to cold, and waterboarding.
A few years later, in order to remove any ambiguities about U.S. obligations under the Geneva and Torture conventions, Congress enacted the Detainee Treatment Act of 2005. The point of that legislation was further reinforced in a June 2006 ruling by the Supreme Court in Hamdan v. Rumsfeld. The White House was thus forced to accept that the Torture Convention and Common Article 3 of the Geneva Conventions do apply to the conflict with Al Qaeda and other illegal combatants, and that anyone detained by the U.S. anywhere in the world should not be subjected to torture or cruel, inhuman or degrading treatment.
As a result, the U.S. Army’s revised Field Manual (FM) 2-22.3: Human Intelligence Collector Operations (2006) affirms the importance of respecting the Geneva Conventions and the Detainee Treatment Act, not only when questioning enemy prisoners of war but even when interrogating suspected insurgents and terrorists, often categorized as “unlawful enemy combatants.” The manual also reminds Army personnel that they can be prosecuted under the Uniform Code of Military Justice (UCMJ) for cruelty, assault etc.
Although FM 2-22.3 does not explicitly refer to the Torture Convention, its regulations appear to be consistent with it for the most part. (A separate manual on counterinsurgency issued in December 2006 by the Army and Marines does cite that treaty specifically, as does a more comprehensive 2015 Law of War Manual encompassing all U.S. military services.) In light of the notorious uses of beatings, sexual humiliations and intimidating guard dogs that occurred at Abu Ghraib and elsewhere, FM 2-22.3 specifically prohibits the following:
Forcing the detainee to be naked, perform sexual acts, or pose in a sexual manner. Placing hoods or sacks over the head of a detainee; using duct tape over the eyes. Applying beatings, electric shock, burns, or other forms of physical pain. “Waterboarding.” Using military working dogs. Inducing hypothermia or heat injury. Conducting mock executions. Depriving the detainee of necessary food, water, or medical care.
However, some of the interrogation techniques that are authorized elsewhere in FM 2-22.3 seem to be in tension with at least the spirit of the Geneva and Torture conventions. For example, sleep deprivation—which can disorient a resistant subject and impair his ability to stick to a consistent story—is apparently permitted by the manual, within limits: detainees must be allowed at least “four hours of continuous sleep every 24 hours.” But there’s no stipulation as to how many days such a regimen could legitimately be sustained; some reports indicate detainees to have undergone 20-hour interrogations every day for many weeks in a row. Prolonged sleep-deprivation can be highly detrimental to health.
An additional Army-approved approach referred to as “Fear Up” involves identifying a preexisting fear or creating one in the source’s mind, in order to link the reduction of that fear to his cooperation. The manual cautions, “The HUMINT collector must be extremely careful that he does not threaten or coerce a source” lest he violate the UCMJ. But the line separating “Fear Up” from illegal threats is perilously thin, especially in regard to a “Fear Up Harsh” variant.
Another method termed “Pride and Ego Down” is “based on attacking the source’s sense of personal worth (…). In his attempt to redeem his pride, the source will usually involuntarily provide pertinent information in attempting to vindicate himself.” Although this technique is hard to distinguish clearly from treatment that would classify as degrading, FM 2-22.3 states candidly: “The HUMINT collector must remember that his goal is collecting information, not concern with the psychological well being of the source. He will be concerned with the latter only insofar as it helps him obtain the former.”
While neither “Fear Up” nor “Pride and Ego Down” would qualify as torture under the law, since they don’t rise to the level of inflicting severe pain or suffering, they risk qualifying as cruel, inhuman or degrading treatment under both the Geneva and Torture conventions.
On the other hand, whatever military law and regulations currently permit or require, that does not exhaust the moral issues at stake. We might also imagine hypothetically, What if the laws were silent on torture? How should we construe the relevant moral concerns, and sort out any conflicts among them?
Is a Moral Right Not to Be Tortured Absolute?
Torture and other cruel, inhuman and degrading treatment share the following essential characteristics: 1) an intentional infliction of suffering in another person, 2) without that person’s informed consent, and 3) not intended to promote that person’s welfare. The second condition is essential to distinguish cruelty from, say, painful medical experiments that people might freely volunteer to undergo solely to help others. Condition three is needed to exclude painful medical treatments given to children or the mentally retarded for their own good but without their informed consent, since they lack that capacity. But any actions characterized by all three conditions are at least prima facie immoral, because they are clearly in tension with moral principles basic to virtually every serious normative theory today: compassion or concern for the well-being of others (entailing nonmaleficence or “non-harm”), and respect for personal autonomy, dignity and equality. By implication, a right not to be tortured would seem to be among the most fundamental of human rights.
If a moral right not to be tortured were absolute, then there could be no legitimate exceptions to a rule against torture. This ethical stance is implied in the Geneva and Torture conventions, and explicitly affirmed by organizations like Amnesty International, Human Rights Watch, and many legal scholars and moral philosophers. The U.S. military’s 2006 Counterinsurgency manual echoes that view: “Torture and cruel, inhuman and degrading treatment is never a morally permissible option, even if lives depend on gaining information.”
An absolute right not to be tortured is especially compelling in light of the horrifying testimony of torture victims during the past century at the hands of ruthless dictatorships. We need to retain the sense of revulsion and terror that torture evokes, even while examining it philosophically.
Consider also the scope and significance of the jus-in-bello rule of noncombatant immunity: the underlying principle here is that people who pose no physical threat to others should not be targeted in war; this evinces the Latin root of the word “innocent,” i.e., nonthreatening. The principle of noncombatant immunity not only prohibits direct and intentional attacks on civilians, it also forbids harming soldiers who have either surrendered or been incapacitated by their wounds. Respect for this fragile principle might be seen as the most important way to prevent international conflicts from becoming total wars of annihilation, nothing more than a grim series of atrocities. So any step taken to qualify the complete prohibition on torturing unarmed detainees is very alarming in the context of military ethics and law.
However, although it disturbs me to say this, I’m not convinced that torture in interrogation is necessarily or always immoral. This is because an absolute right not to be tortured would entail that nothing that anyone might intentionally do to others could justify torturing them, even actively plotting mass murder, which strikes me as an absurd ethical stance.
Imagine that a senior member of Al Qaeda or ISIS were arrested, and refused to cooperate with his captors in response to humane interrogation tactics. In spite of his having instigated the murder of scores of innocent people, and more importantly his probable involvement in planning many more killings, let’s also imagine that he then were to claim an absolute right not to be tortured, and demand to be treated accordingly. (Christian extremists have historically committed many more murders in the U.S. and Norway than Muslim extremists, but I’ll stick with the Al Qaeda/ISIS scenario here.)
Ignoring for a moment the legal rights accorded to detainees under the Geneva and Torture conventions, I have great difficulty accepting the plausibility of absolute moral claims or demands made by people who have shown complete contempt for the basic rights and well being of others, especially if they have knowledge of ongoing plans to murder the innocent. I’m thus led to hypothesize that a moral right not to be tortured may be something less than absolute, that it might be more sensible to consider it a prima facie right instead, i.e., a right that is clearly established and usually ought to be upheld, but which can be trumped by other moral considerations under certain circumstances. Let me explain. A prima facie right not to be tortured might be qualified in a couple of ways:
a) Perhaps the right could be overridden by the rights of innocent persons not to be murdered, if torture were thought to be the only way to obtain information needed to prevent their murders. That sounds intuitively plausible, but it might also rationalize the torture of innocent people in order to save other innocents, which I think would be fundamentally unjust. Indeed, we would become little better than terrorists if we intentionally tortured the innocent, so I would strongly urge drawing a bright ethical line there.
b) Alternatively, perhaps a moral right not to be tortured could be forfeited, as in the hypothetical example of the captured Al Qaeda or ISIS leader. This would be similar to a deontological rationale for capital punishment: we rightly assume that every person has a prima facie right not to be killed, but we might nonetheless also claim that even that basic right can be forfeited by individuals who commit murder or conspire to do so.
Although torture in both a) and b) would be morally troubling to say the least, only in b) would it not clearly be unjust. In other words, while torture would certainly harm an Al Qaeda or ISIS leader, it wouldn’t necessarily wrong him. (The same could be said of executing a murderer.) By contrast, torturing the innocent would both harm and wrong them.
In sum, only those who could plausibly be said to have forfeited their right not to be tortured could legitimately be subject to that appalling treatment; and, I’d further stipulate, only if necessary to prevent serious harms to innocent persons, when more humane interrogation methods are highly unlikely to produce that result or have already failed.
But concluding that someone has forfeited their right not to be tortured might seem to imply that there would be no moral limits on what their interrogators might be allowed to do. This is so disturbing, even regarding would-be mass-murderers, that we must look more closely at the right in ques- tion and other ethical concerns beyond that.
Perhaps some rights can be forfeited in part but not wholly, to some extent but not completely. For example, when we send convicted criminals to prison, we intentionally deprive them of some of their rights, but not all of them. (E.g., even criminals sentenced to death in the U.S. are protected from “cruel and unusual punishments” under our Constitution, without any obvious logical, moral or legal contradiction.) Similarly, even if terrorists could not credibly claim moral immunity from torture entirely, they would presumably retain the right not to be subjected to all possible forms of torture, or to be tortured merely out of vengeance or spite, or to amuse their captors, or tortured long after they could plausibly know any “actionable” intelligence. (There is a parallel here to the jus-in-bello principle of proportionality.) History shows that once torture is authorized ostensibly to prevent terrorism, it’s all too easy for the practice to become standard in eliciting “confessions” of past crimes and punishing people who were never properly convicted of any crime. Hence, even if torture were morally justified in certain cases, it could not justifiably be conducted in utterly ruthless fashion.
Intelligence Officer Virtues and Vices
One of the professional skills uniquely required of intelligence officers (both civilian and military) who employ HUMINT is an ability to manipulate persons. The degree of manipulation can vary from the subtle blackmail threat latent in a financial relationship with an espionage agent to more obviously coercive measures. The element of control in intelligence operations is directly related to suspicion of the loyalty of the agent. And suspicion is a professional virtue for intelligence officers, especially for those who work in security and counterintelligence, since in theory anyone thought to be trustworthy may in fact be secretly serving the enemy.
The practice of interrogation is a significant component of intelligence work, but also illustrates manipulation in its rawest form. William Johnson, a former CIA counterintelligence officer, offered a glimpse of the ethical risks involved: “Interrogation is such a dirty business that it should be done only by people of the cleanest character. Anyone with sadistic tendencies should not be in the business.” We are reminded, though, of the ease with which people can come to rationalize callousness and cruelty in dealing with perceived enemies. Given the natural human capacity for aggression, combined with the wrong set of biases, incentives and peer pressures, many ordinarily decent people can succumb to sadism or callous cruelty. But torture is not necessarily something conducted solely by sadists. For example, in interrogating terrorist suspects, an intelligence officer might well be driven by the motive of preventing harms to innocent people. That officer might take no great pleasure in inflicting pain or fear, but nonetheless consider it a regrettable but necessary means of seeking enough details about a terrorist plot to nip it in the bud.
Empathy, in everyday moral parlance, is related to compassion. But in intelligence work, the other is considered to be a potential threat to persons and interests that the intelligence officer is sworn to protect. “Knowing one’s enemy” in this role, means understanding the other, but not in the interest of enhancing his or her freedom or well-being: on the contrary, empathy becomes a manipulative tool.
Would authorizing the torture of suspected terrorists, even within strict limits, inevitably corrupt the consciences and character of the personnel we asked to conduct it? There is something so obviously and intrinsically appalling about torture that anyone who hoped to remain a person of integrity, an admirable person, would not use more than the minimum degree of force necessary to obtain vital information. In other words, even if we could show that the person being interrogated had forfeited his right not to be tortured, an ethical interrogator would not consider that a “blank check.”
Then again, having the compassion that we’d expect of a person of conscience might make a professional interrogator less effective than someone who was less benevolent or who could train himself to suppress his compassion when questioning a detainee.
In a memoir entitled Fear Up Harsh (2007), former Army interrogator Tony Lagouranis acknowledged developing over time a tendency in his interrogation of Iraqi detainees to use increasingly harsh methods out of frustration when unable to persuade them to reveal what he thought they knew. He and his peers became obsessed with dominating detainees, achieving power and control over their will. This suggests a psychological slippery slope potentially facing any interrogator, but especially one who resorts to harsh techniques.
Consider what it would take to create a training program for personnel who would be authorized to conduct torture. Presumably we would be logically and practically compelled to authorize a broad range of “scientific” experiments on the relative effectiveness of various forms of torture in interrogation, as well as psychological assessments to determine who among the group of prospective interrogators would be most effective at certain techniques. These projects might require the participation of medical doctors at various stages, which would represent an extreme departure from their core professional ethic of nonmaleficence and respect for persons. The 2014 Senate Intelligence Committee report revealed that medical personnel in fact participated in harsh CIA interrogations over several years.
Some CIA and military interrogators themselves objected to being ordered to employ torture, seeing it as a violation of their professional or personal ethic. If raised again this issue would no doubt generate heated debate among intelligence officers, and perhaps necessitate establishing a “conscience clause” to permit objectors to opt out. In February 2016, after presidential hopeful Donald Trump vowed to bring back waterboarding “and worse” interrogation techniques, several former CIA leaders stated that the Agency would refuse any such orders. Although the claim that “Americans don’t torture, period,” was never comprehensively true, it clearly represents an important ideal or core value that many conscientious professionals would be unable to abandon.
Other Potential Consequences of Legalizing Torture
I must now acknowledge that my earlier hypothetical example of the captured terrorist leader, like the standard “ticking bomb” scenario, assumes greater knowledge of the identity and intent of the subjects of interrogation than their captors typically have. In other words, allowing any torture in interrogation runs the risk in practice of subjecting entirely innocent people to horrific and wholly unjust suffering. Individuals are sometimes erroneously detained by counterinsurgency forces, for instance, as a result of false accusations made against them by fearful or resentful neighbours, or based on flimsy circumstantial evidence, or when rounded up for questioning with other locals to satisfy some arbitrary quota or misguided metric of productivity. So anyone like me who questions whether a right not to be tortured is absolute must take into account the incredibly unjust harms to the innocent that could easily occur if the practice of torture were officially permitted at all.
In the end, I don’t believe that it’s possible to eliminate the chance of accidentally torturing the innocent if interrogational torture were permitted legally, even if conducted by the most conscientious and skilled interrogators we have. Even one instance of that would be a horrific tragedy. But would it be possible to limit that risk significantly, short of a blanket prohibition? And if so, would that make it morally acceptable, perhaps along the lines of the jus-in- bello rule of proportionality which permits harms to noncombatants as long as they are not directly and intentionally targeted?
Alan Dershowitz has famously proposed (in Why Terrorism Works, 2002) requiring intelligence and law-enforcement personnel to obtain a “torture warrant” (like a search warrant) from a judge before being allowed to use torture on a terrorist suspect. He argues that this would make the practice of interrogational torture—which he believes is inevitable—both more accountable and less frequent. Andrew McCarthy has further suggested that judicial authorizations for torture would be more effectively regulated by means of a centralized “national security court,” a single tribunal made up of federal judges.
If we openly permitted torture, even with the restrictive legal procedures advocated by Dershowitz and McCarthy, would the U.S. lose all of what remains of our credibility in the international community on human rights? I would assume so. Alberto Mora, a former general counsel to the U.S. Navy, noted (in Senate testimony) that our country’s adoption of harsh interrogation methods had highly detrimental effects on our relationships with allies:
As [U.S. allies] came to recognize the dimensions of our policy of cruelty, political fissures between us and them began to emerge because none of them would follow our lead into the swamp of legalized abuse, as we should not have wished them to. These fissures only deepened as awareness grew about the effect of our policies on fundamental human rights principles, on the Geneva Conventions, on the Nuremberg precedents, and on the incidence of prisoner abuse worldwide. Respect and political support for the United States and its policies decreased sharply abroad (…). International cooperation, including in the military, intelligence, and law enforcements arenas, diminished as foreign officials became concerned that assisting the U.S. in detainee matters could constitute aiding and abetting criminal conduct in their own countries.
Mora added that “allied nations have refused on occasion to train with us in joint detainee capture and handling operations because of concerns about U.S. detainee policies,” and that “senior NATO officers in Afghanistan have been reported to have left the room when issues of detainee treatment have been raised by U.S. officials, out of fear that they may become complicit in detainee abuse.”
Would we want other countries to follow our example if we legalized torture? Presumably they would be logically permitted to do so in relevantly similar circumstances, so we’d be hard-pressed to persuade them not to imitate us. But their systems of legal checks on abuses of power would not necessarily be as robust as ours (even recognizing that we haven’t prosecuted any senior U.S. officials for authorizing torture). So the number of innocent people tortured around the world would probably grow.
In addition, legalizing torture by U.S. or other NATO intelligence officers would almost certainly undermine our efforts to “win hearts and minds” in countries where we’re battling terrorists or insurgents. Would our own personnel be placed at greater risk of torture if captured or kidnapped overseas? This seems quite likely. Detainee abuses at Abu Ghraib almost certainly motivated hundreds if not thousands of Iraqis to kill American troops. An Iraqi interviewed by Mark Danner in November 2003 (months before those abuses were publicized) vividly and passionately anticipated that result:
For Fallujans it is a shame (…) for the foreigners to put a bag over their heads, to make a man lie down with your shoe on his neck. This is a great shame, you understand? This is a great shame for the whole tribe. It is the duty of that man, and of that tribe, to get revenge on this soldier—to kill that man. Their duty is to attack them, to wash the shame. The shame is a stain, a dirty thing; they have to wash it. No sleep—we cannot sleep until we have revenge. They have to kill soldiers.
The serious concerns discussed in the previous section weigh heavily against changing our laws to permit torture under any circumstances. Even if terrorists have in effect forfeited their moral right not to be tortured, and even if torture might prevent some terrorist attacks, there seem to be overriding moral reasons not to legalize torture in counter-terror interrogations. By analogy, even if we regard some crimes to be so heinous as to deprive their perpetrators of the right not to be executed, we might nonetheless refrain from instituting capital punishment, or establish a moratorium on further executions, out of concern, say, for the risk of inadvertently executing innocent people falsely convicted from sloppy police work, incompetent defense counsel, overzealous prosecutors, or the false or mistaken testimony of witnesses.
Fortunately there are many humane interrogation techniques that skilled military and civilian intelligence personnel know how to use effectively, methods that they can rely on in the overwhelming majority of cases to elicit the intelligence that may be desperately needed to save scores of innocent lives.