February 7, 2003

A thought-experiment on torture and expediency

The New York Times this morning has an interesting story this morning about the trial of Zacarias Moussaoui, who is accused of being a co-conspirator with the al-Qaeda members responsible for the 9/11 attacks. The trial has already had a number of twists and turns, including a rejected attempt by Moussaoui to plead guilty, but more recently, it has become increasingly clear that federal prosecutors are wishing they had never brought the case to a US criminal court in the first place. Now the Times reports that the federal government believes they may have to transfer Moussaoui’s trial to a military tribunal in order to avoid granting his lawyers’ motion that they be allowed to question Ramzi bin al-Shibh, the alleged al-Qaeda intermediary who carried messages between the 9/11 conspirators and al-Qaeda’s top leadership.

Bin al-Shibh, who was captured last October, is allegedly being "interrogated overseas". The Times also asserts that federal officials do not want some of his knowledge about al-Qaeda to become public knowledge and quotes a federal law enforcement official who says that they also do not want to disrupt “psychological games” they are playing with al-Shibh.

Call me a cynic, but here’s how I read this story: we’re allowing al-Shibh to be tortured (or directly torturing him ourselves) and we don’t want him to see the light of day so that this becomes public knowledge.

Time was when these kinds of accusations came up, US federal or local authorities just denied them flatly—mostly because they were in fact untrue, the kind of fevered exaggeration that conspiracy theorists invent casually; in a few cases, especially some notorious instances of police violence, because the accusations were true and actionable criminal behavior in their own right. Now instead what we hear are non-denial denials and sober discussions in the public sphere about whether torture might be justified in some circumstances.

After all, wonder some, what if you could have stopped a terrorist attack that killed tens of thousands of people if only you’d used all means available to get information from a captured suspect?

I want to take that question seriously. It’s an old question given new urgency. This is a version of the Stephen King Dead Zone Johnny Smith thought-experiment, “Would you assassinate Hitler if you found yourself miraculously back in time in 1931 or 1933, even knowing that no one would believe the Holocaust possible and everyone would think you a madman?”

What exactly is it that separates us from al-Qaeda? What are we defending? What are the limits of our defense? What would we not do to stop a terrorist attack that kills tens of thousands?

Because no one would say that everything that might stop a terrorist attack of that magnitude is justified—and I wish there weren’t so many people on the left who seemed to be saying that virtually nothing should be done to forestall such an attack. And yet, there are many commentators who seem to think that all they need to do is wave the magic phrase, “Do you want to be the person who failed to stop the use of a weapon of mass destruction by terrorists when you could have prevented it?” That’s not an argument in favor of a particular action: it’s a platitude that vacates all hope of rational discussion. In some cases, yes, we ALL want to be the person who failed to stop a terrorist attack because the method of stopping one would be worse.

So, I propose a thought-experiment to calibrate our standards, to discover where the lines are that almost none of us would cross, and try to figure out what the basis is for drawing them. Table for the moment the thought that the actions in question would actually not prevent terrorist attacks, or would provoke worse attacks. I’m just asking: where would you draw the line if you knew that a particular action was guaranteed to stop an actual attack?


1. A campaign of total extermination directed at anyone who professes the Islamic faith.
2. Complete closure of the borders of the US, expulsion of dissidents and non-citizens, and suspension of the Constitution.

I fervently hope that no one reading says, “Sure, I’m ok with either of those”, stipulating for the purpose of this exercise that these measures were guaranteed to prevent a terrorist attack (which of course they wouldn’t be). So none of us think that any and all preventative measures are justified.

3. Through a lucky break, US intelligence discovers that a terrorist cell has a small nuclear weapon hidden in a basement in the middle of a medium-sized city in an Arab nation. We know roughly where it is within a ten-block radius, but not precisely. There are fears that the cell may try to move the weapon and we will lose track of it, and it is known that the cell may have the ability to transport the weapon to the US without being detected. The government of the nation in question, while not aiding the terrorists, will also not permit a house-to-house search or initiate one themselves. If we knew that we could detonate the bomb preemptively and remotely in some fashion, killing as many in the Arab city as the bomb would have killed in a US city, but preventing that attack, should we?

I’m hoping that anyone reading would say, “No”. The standard proposed here is that US lives are always and under all circumstances worth more than the lives of non-Americans, and that there is no scale at which the deaths of innocent non-Americans makes action to protect Americans intolerable. If someone told me, “Either those civilians in that city die or you do: the bomb goes off there or it goes off here,” I would say that I would rather die myself. I would rather be murdered than be a murderer.

4. A captured al-Qaeda operative known to possess specific information about a planned attack using weapons of mass destruction resists all interrogation, including psychological torture, beatings and shocks to his genitals. Officials looking into his background find out, however, that he has an unusual and intense aversion to the thought of rats gnawing his face off. So they take him off to Room 011 and put a cage full of hungry rats on his face. He gives in after losing most of a cheek.

Right, I know, sounds familiar. The point is, suppose you think torture is justifiable if it prevents an attack. Is all torture justifiable? How about sawing off someone’s genitals slowly, or stretching them on the rack? If not, why? What’s the dividing line between beatings and amputation, exactly?

I ask this in all seriousness. There isn’t an absolutely clear distinction, for example, between putting someone handcuffed in a dark room with a hood over their head for 24 hours and Pembleton and Bayliss grilling someone in The Box on the TV show Homicide. But if Pembleton and Bayliss could make an al-Qaeda operative spill the beans and save ten thousand lives through psychological trickery and pressure, I would be okay with that. I might feel queasy about the hood-in-the-dark scenario, though, and the full Winston Smith treatment would be intolerable. At that end of the spectrum, I say once again, “I would rather die if the only way to save me is hot pokers and iron maidens”.

At the other end of the spectrum, I suspect even the most lefty among us might say, “Hey, if Encyclopedia Brown manages to trick Bugs Meany into confessing while he’s being kept in the principal’s office, that’s fine.”

In between is where it gets tough. Solitary confinement and a minimal diet for a month? I guess that might be okay if it saved lives. Ten hours of interrogation under a blazing light culminating in the interrogator striking the prisoner three or four times in the face? If that produced the information that saved ten thousand lives, would I really say no? Obviously the problem here is that most of us recognize that we might trust ourselves to make these judgements carefully but we do not necessarily trust others—or governments—to do so, and we also recognize that this is the most slippery of slippery slopes, where a slap in the face one day turns into bamboo shoots under the fingernails the morning after.

5. Al-Qaeda operatives are meeting in a neighborhood in a Somali city, and they have a cache of weaponized anthrax with them. After the meeting, several are going to carry the anthrax to the US. Once they leave, US intelligence isn’t sure it can track them. A surgical strike with cruise missiles kills all the operatives and vaporizes the anthrax safely, but in the attack five civilians who had no association with al-Qaeda—people in the wrong place at the wrong time—also die.

I’m ok with this, because in this case we didn’t attack knowing we were going to kill those civilians in particular and because the lives saved are hugely disproportionate to the lives lost. In this scenario, I would rather that I live and those five innocents die. Those deaths are not really on my conscience: they go on the balance sheet of al-Qaeda, not the US.

The upshot of all these rigged hypotheticals is simply to say that if the US government is torturing bin al-Shibh or countenancing his torture and fears that being known, we can’t just shrug and say, “Whatever saves lives”. Because none of us endorse all possible steps that would pre-empt a terrorist attack, and few of us reject everything that might. Nor can we permit our government to say, “Move along here, none of your business, trust us.”

If we’re going to torture bin al-Shibh, we have the right—the obligation—to make that decision as a whole society, in the light of day. Maybe that’s most important thing that would distinguish us from our enemies, and the most important value we are trying to defend.