Aristotle on Torture

I was just reading through Aristotle's Rhetoric and found Aristotle discussing judicial torture. He doesn't apparently take sides about evidence extracted under torture (since this is a book about rhetoric), but nonetheless shows a fairly critical attitude. In the context he is pointing out arguments that, in a law court, could be used to both promote or dismiss such evidence. He ends up pointing out more extensively why such evidence is untrustworthy, saying: "we can destroy [evidence extracted under torture's] value by telling the truth about all kinds of torture generally; for those under compulsion are as likely to give false evidence as true, some being ready to endure everything rather than tell the truth, while others are equally ready to make false charges against others, in the hope of being sooner released from torture"(1377a2-6). What's interesting is that cases of false information extracted under torture were apparently well enough known, even in Aristotle's day, that a person could cite several examples.

This issue has come up recently because some information extracted under the torture of Khalid Sheikh Mohammed eventually led to the discovery of the hiding place of Osama bin Laden. Some people have used this as evidence that torture can have its value. The argument for torture is a form of Utilitarian argument, basically stating that, in certain circumstances torture is justifiable when the benefit of information extracted under torture vastly outweighs the evils of torture. For example, if you torture someone to get information to stop a terrorist plot that will kill thousands of people, then it's justified. Such argument might be a bit tenuous in this case, since it's not clear that Killing Osama will undermine Al Qaeda or make a dent on the threat of terrorism, but there still might be some cases where it applies.

Steve Chapman over at Reason writes about the weakness of this argument. The problem is that this argument relies on a couple key assumptions, that information extracted under torture is reliable and that torture is the best way to extract information. But, the unreliability of torture has been known at least as far back as Ancient Greece. Since it's all too common for innocent people who know nothing to be tortured into giving false information to get the torture to stop, about the only circumstance that torture might be reliable is if you have a suspect that you know is holding some information, but who refuses to divulge.

Even here, torture doesn't appear to be the best strategy. People are human and are generally willing to divulge information with people they trust and bond with. If Stockholm Syndrome is real, then you can gain the trust and empathy even of prisoners and get them to willingly speak openly about what they know. In fact, it can work even for people who really don't feel empathy. Remorseless serial murderer Pedro Lopez confessed to his several hundred murders after police mercilessly beat him into subm... Oh wait. No, they didn't do that. They just put a priest in the cell with him, and the priest earned his trust, and Pedro Lopez confessed. He even took police out to show them all the burial places of the victims he'd killed in the area. Steve Chapman points out that that gaining trust is probably more productive, since it'll lead the subject to divulge more volubly, whereas under torture subjects will try to give as little information as possible. In other words, if you try to torture some kernel of information out of a guy, he might finally divulge it, but he'll completely close up his mouth from then on. Gaining someone's trust and bonding with them does require a bit more finesse and creativity, but it's worth the effort and will generally lead to more information extracted.

The case that torture is horrible and unethical is a pretty easy case to make, but it's sometimes justified on Utilitarian grounds, arguing that there are cases where the ends justify means. Unfortunately, even these arguments seem highly doubtful and should lead us to conclude that torture is something that is never justifiable.

I was just reading through Aristotle's Rhetoric and found Aristotle discussing judicial torture. He doesn't apparently take sides about evidence extracted under torture (since this is a book about rhetoric), but nonetheless shows a fairly critical attitude. In the context he is pointing out arguments that, in a law court, could be used to both promote or dismiss such evidence. He ends up pointing out more extensively why such evidence is untrustworthy, saying: "we can destroy [evidence extracted under torture's] value by telling the truth about all kinds of torture generally; for those under compulsion are as likely to give false evidence as true, some being ready to endure everything rather than tell the truth, while others are equally ready to make false charges against others, in the hope of being sooner released from torture"(1377a2-6). What's interesting is that cases of false information extracted under torture were apparently well enough known, even in Aristotle's day, that a person could cite several examples.

This issue has come up recently because some information extracted under the torture of Khalid Sheikh Mohammed eventually led to the discovery of the hiding place of Osama bin Laden. Some people have used this as evidence that torture can have its value. The argument for torture is a form of Utilitarian argument, basically stating that, in certain circumstances torture is justifiable when the benefit of information extracted under torture vastly outweighs the evils of torture. For example, if you torture someone to get information to stop a terrorist plot that will kill thousands of people, then it's justified. Such argument might be a bit tenuous in this case, since it's not clear that Killing Osama will undermine Al Qaeda or make a dent on the threat of terrorism, but there still might be some cases where it applies.

Steve Chapman over at Reason writes about the weakness of this argument. The problem is that this argument relies on a couple key assumptions, that information extracted under torture is reliable and that torture is the best way to extract information. But, the unreliability of torture has been known at least as far back as Ancient Greece. Since it's all too common for innocent people who know nothing to be tortured into giving false information to get the torture to stop, about the only circumstance that torture might be reliable is if you have a suspect that you know is holding some information, but who refuses to divulge.

Even here, torture doesn't appear to be the best strategy. People are human and are generally willing to divulge information with people they trust and bond with. If Stockholm Syndrome is real, then you can gain the trust and empathy even of prisoners and get them to willingly speak openly about what they know. In fact, it can work even for people who really don't feel empathy. Remorseless serial murderer Pedro Lopez confessed to his several hundred murders after police mercilessly beat him into subm... Oh wait. No, they didn't do that. They just put a priest in the cell with him, and the priest earned his trust, and Pedro Lopez confessed. He even took police out to show them all the burial places of the victims he'd killed in the area. Steve Chapman points out that that gaining trust is probably more productive, since it'll lead the subject to divulge more volubly, whereas under torture subjects will try to give as little information as possible. In other words, if you try to torture some kernel of information out of a guy, he might finally divulge it, but he'll completely close up his mouth from then on. Gaining someone's trust and bonding with them does require a bit more finesse and creativity, but it's worth the effort and will generally lead to more information extracted.

The case that torture is horrible and unethical is a pretty easy case to make, but it's sometimes justified on Utilitarian grounds, arguing that there are cases where the ends justify means. Unfortunately, even these arguments seem highly doubtful and should lead us to conclude that torture is something that is never justifiable.

Aristotle on Torture