Debates about the legality of torture often invoke intuitions about the morality of the practice. The argument is that there may be extreme circumstances in which torture is morally justified, and if so the law should reflect this. (A subsidiary and separate issue is whether a particular given circumstance, e.g. the War on Terror, rises to the level of sufficiently extreme circumstances.)
Many people, especially the type of people who believe in inalienable human rights, have the intuition that torture is always wrong. Other people are not so sure; their intuitions are more utilitarian (it was Bentham, after all, who said “The idea of rights is nonsense and the idea of natural rights is nonsense on stilts”). What if torturing (killing?) one person (or a few people? or a few dozen people?) could save thousands, or millions? Wouldn't that be morally justified?
The most common capsule version of this question bandied about is the ‘terrorist known to have an a-bomb in NY’ (TABNY) scenario: What if the police capture someone 'known' to have a ticking a-bomb secreted somewhere in a major city, and 'know' they have only 24 hours to get the location before it goes off.
I think these hypotheticals have almost no connection with reality: How can the police 'know' the suspect is in fact guilty, and 'know' about the deadline, with sufficient moral certainty to be willing to contemplate torture, and yet not know what they seek?
I think these hypotheticals also elide what we know about torture: that some people don't crack, and that others will say anything, yes, even false things, to make it stop. And how many false leads does the victim have to give before the 24 hours are up?
But never mind that. Let's take it on its own terms. And by its own terms, I mean from the framework of a utilitarian moral calculus, since I doubt that a short blog post is going to convert a utilitarian to a rights-based vision of morality (although there are arguments justifying rights-based morality in consequentialist or utilitarian terms).
Law preforms complex functions in modern society. Among them it gives notice of which actions risk consequences (deterrence) and is to at least a limited extent a moral statement of what the community values, tolerates, or abhors (education).
We also know that rules tend to be violated. Generally speaking, however, if something is permitted we are likely to see more of it. Indeed, as the Medium Lobster recently noted, in some visions of utilitarianism it would be morally proper to torture N people if it would save N+1, or even morally proper to torture infinite numbers of “them” to save one of “us”.
Rules against the torture of suspects/detainees/prisoners are directed at the people who have power over that person. If we as a nation craft a rule that says torture is permitted to serve the greater good, we instruct the police officer/CIA officer/soldier at the sharp end that they should in each case make a personal judgment as to whether the end justifies this means. It is the nature of man, and especially bureaucratic man, that in times of stress people frequently are going to choose to err in the direction of heading off the mass disaster rather than risk being held responsible for failing to prevent it. That means we're likely going to see a lot of torture, indeed “too much” even by a utilitarian calculation. Furthermore, once you open the door to torture, there's no logical reason to think it will only be applied to “them”. What if the suspected 'terrorist with the bomb' is one of “us”?
Utilitarian opponents of a flat no-torture rule nevertheless object that it fails to deal with the rare but possible TABNY case where torture would be justified, and that this failing should be corrected. Here, I think I'll follow the great Charles L. Black, Jr. lead. In an article I wrote on cryptography and the constitution a few years ago I summarized Black's view:
that an “absolute” right against being tortured might nonetheless find room for an exception in the case of “the man who knew where the [atom] bomb [was ticking, but] sat grinning and silent in a chair” far from the place he had planted it. Charles L. Black, Jr., Mr. Justice Black, The Supreme Court, and the Bill of Rights, Harper's, Feb. 1961, at 63, reprinted in The Occasions of Justice: Essays Mostly on Law 89, 99 (1963). Explaining this position in a Constitutional Law class I attended at Yale in 1984, Professor Black stated that he believed torture morally justified in this extreme and hypothetical case. Once the torturer extracted the information required, Black continued, he should at once resign to await trial, pardon, and/or a decoration, as the case might be.
I think Charles Black got it exactly right. I'm not sure that I think torture is ever morally justified or sensible, but I am prepared to accept that in the most extreme circumstances there might be an exception to that rule. But one thing I am certain about: if someone thinks that torture might be morally correct in a given situation, I want the potential torturer to understand that by acting on their view they are putting themselves personally at risk, and that their duty is to turn themselves in as soon as they've extracted what they sought (or failed).
If it turns out that the belief which motivated the torture was justified (and the a-bomb is defused), we may praise them. But if it turns out that the belief was mistaken, and especially if they have tortured an innocent, let them not turn to legal institutions for refuge.