Hacker News Reader

Comment by zhaphod | Hacker News Reader

People who coined the term enhanced interrogation should be subjected to the said interrogation. Joking aside, I agree with the overall point the author is making. There is a torture creep in the culture and people are getting insensitive to what it means. Because it feels good to catch the bad guy and do unspeakable things unto him and get the location of the ticking time bomb. What most people don't realize is that torture doesn't work. The person being tortured will give what ever information the torturer is looking for to stop the pain. However, that is an inconvenient fact that people tend to ignore.

View on HN

Sometimes the subject does have the sought-after information, and the interrogator knows the subject has it, and people will die if that information is not extracted. Sometimes asking nicely doesn't work, and neither does saying "Tell me! or I'll say 'Tell me!' again!"

Another inconvenient fact is sometimes it does work, and is an alternative to actual harm.

ETA: I'm surprised at how many here adhere to the notion "you can't ever know if the subject knows". We justifiably incarcerate people in peacetime, and kill people in war, on less information. As another poster noted: if the subject possesses a computer clearly connected with the issue, admits it's his, admits he knows the password, refuses to give the password, reveals the password under enhanced interrogation (with no physical harm done), and the password decrypts life-saving information, far better that than letting innocents die (or killing him putatively after the fact) because "torture isn't nice". Unwavering equalization of torture with crime (and invariant punishment thereof) is just as absurd as invariably criminalizing use of deadly force in defense of self or others. I get that sometimes/often/usually it doesn't work, but this broad insistence it is always criminal is absurd.

Rather than legalizing torture as a result of these sorts of fantasies, one might argue that in such extraordinary cases, the interrogator should be willing to suffer the consequences of committing the crime of torture.

This risks giving them some sort of glorified martyr position that people might subsequently aspire to.

As opposed to risking just making it legal so anyone can do it without a second thought?

As opposed to saying "It's never the right thing to do, no matter how certain you are that this case is different".

I agree, but I was addressing the fantasy that there is a different case, which remains pervasive, despite an abundance of strong arguments against it made by other replies offered to parent.

If by sometimes you mean hypothetically. Still, we don't need to make torture legal.

But I bite. Let's say that thing you describe happens. People who torture should still get hard sentences, say 20 to 25 years. If they are not willing to sacrifice their own life and break the law to get the information, the torture is not justified.

In other words: torture should always be big personal sacrifice for the torturer.

How is the determination made that the subject does have the sough-after information. Because the subject may give what ever information the interrogator is biased to receive.

The following pdf "http://www.cgu.edu/pdffiles/sbos/costanzo_effects_of_interro... makes the case much more eloquently than I can in this comment section. Reading the "Conclusion: Consequences and Alternatives" section is very instructive.

Also the guardian article "http://www.theguardian.com/science/the-lay-scientist/2010/no... is very instructive.

Furthermore, I feel using torture is inherently immoral and we should strive to achieve a better world even when we know that our adversaries are not playing by the rules.

You need to fix your links, remove the quotes at the end of them please.

This is a popular fallacy for supporting torture, but there are a couple things wrong with it.

Firstly, the most likely outcome of torture is false information and multiple sources from the CIA to the FBI and various levels of rank[1] have said that one of the most problematic issues with torture based intel is the extreme amount of time it takes to verify anything because most of it is made up. So in a ticking time bomb scenario, torturing someone you think has information may make you FEEL better, but it is a waste of time (which, by default, is in short supply).

Secondly, and perhaps more importantly amongst this crowd, the logic doesn't work. Statistically speaking, ticking time bomb situations do not actually occur in real life and basing institutionalized acceptance of torture on a fantasy what if scenario is intellectual fraud. Georgetown law's David Luban has an excellent 2005 essay "Liberalism Torture and the Ticking Bomb"[2] that takes a deep dive into the psychology behind this argument as well as the real-world applications. It's linked further down in these comments by @elipsey[3].

[1] http://skeptics.stackexchange.com/questions/4498/does-tortur...

[2] http://scholarship.law.georgetown.edu/cgi/viewcontent.cgi?ar...

[3] https://news.ycombinator.com/item?id=7017729

To be honest, I'm not comfortable with the 'torture is bad because it is ineffective' argument, because it sidesteps the larger moral issue.

For the sake of argument: if it turns out that slicing off a man's left hand will always yield 100% verifiable information, should we still be doing it? My answer to that is still going to be no. The effectiveness of torture is besides the point.

> To be honest, I'm not comfortable with the 'torture is bad because it is ineffective' argument, because it sidesteps the larger moral issue.

I, on the other hand, favor it because it sidesteps the moral issue and presents what is, in fact, a strong and compelling argument against torture that does not rely on first principles that are demonstrably not held by those who would advocate for torture in the first place.

The fact that torture is wasteful even if you ignore any moral considerations is a lot more powerful of an argument against it than one which requires the audience to accept that "torture is wrong" is true as a matter of moral first principles, no matter how much one might wish that the rest of the population would accept that moral principle.

> The effectiveness of torture is besides the point.

Except it's not. Those that are willing to turn to torture don't have a moral issue with it, it would not be just another tool if they did. The effectiveness of the tool is an argument people like that are interested in. In just the same way as those that do have moral issues with torture don't care if it was very effective, your arguments need to be tailored to the audience. Given a varied audience, you hit all the points.

Torture doesn't work for extracting information, it's effectiveness is basically useless. That is, if you are torturing to get information in the first place.

If torture were effective, it would allow the debate to be framed as a tradeoff. If slicing off the man's left hand would allow you to save X lives with 100% certainty, pretty much everyone has a value of X for which they would justify the action -- for most people that value would be quite low.

The problem with these tradeoffs is that there's no clear cut moral high ground on which choice is correct -- it's a murky area. So it's a much more compelling argument if you can make the case that the tradeoff doesn't exist at all b/c torture is ineffective.

http://web.maths.unsw.edu.au/~jim/torturecardozo.pdf

"He asked them where the bomb was. The terrorists-highly dedicated and steeled to resist interrogation-remained silent. [He] asked the question again, advising them that if they did not tell him what he wanted to know, he would kill them. They were unmoved. So [he] took his pistol from his gun belt, pointed it at the forehead of one of them, and shot him dead. The other two, he said, talked immediately; the bomb, which had been placed in a crowded railway station and set to explode during the evening rush hour, was found and defused, and countless lives were saved."

Interestingly, I can find no evidence that this ever occurred outside of the little anecdote in the aforementioned paper, which cites a January 2002 article in The Atlantic.

The article again, is purely anecdotal. The interrogator's name of "Thomas" is just an alias, ostensibly used for "safety reasons".

Well, Luban essay cites newspaper sources as well, it's not a problem per se.

Note that much of the information you cite is old, from technologically backwards times. E.g., the CIA's 1963 interrogation manual:

   Intense pain is quite likely to produce false
   confessions, concocted as a means of escaping
   from distress. A time-consuming delay results,
   while investigation is conducted and the admissions
   are proven untrue. During this respite the interrogatee
   can pull himself together. He may even use the time to
   think up new, more complex ‘admissions’ that take still
   longer to disprove.

A story told by the interrogatee that took a few days to check out then quite possibly could be checked out in a few minutes now, because of our vastly better computer, database, and communications technology.

I think there is a chance there may be a window where torture works reasonably well. You start out where it doesn't work well, for the reason given in the 1963 CIA manual.

Then, as technology gets better, so that you have the databases and data mining applications and communications technology to check out the interrogatee's story in close to real time, torture becomes effective.

Later, as the technology gets even better, and the databases get bigger (ironically, partly due to things like the NSA surveillance), you reach a point where the interrogations are pointless because you aren't getting anything from the interrogatee that you couldn't have gotten quicker with clever queries to the databases you already have.

> Then, as technology gets better, so that you have the databases and data mining applications and communications technology to check out the interrogatee's story in close to real time, torture becomes effective.

This assumes that being able to prove the torture victims story quicker increases the chance of getting a truthful admission.

For starters that presumes that the torture victim knows what you want, which is not a given.

Secondly it presumes that the torture victim will give up the information if only you can keep the pressure up long enough. If it was "that simple" a solution to the problem in the CIA manual would be to simply recommend to continue the torture regardless of admissions, until one of them is proven true, while insisting to the victim that you believe he is lying.

Sometimes does work? You mean, like guessing? I mean, "sometimes" throwing a coin to decide if someone is a criminal will work.

The problem with torture (from a point of view of information) is that the information is NOT reliable AT ALL. Someone under torture will often not only imagine facts, it will actually believe that they are true, and will confess false crimes, especially if presented by the torturer. There is a famous case in Spain in 1910 (a movie was made about it http://en.wikipedia.org/wiki/The_Crime_of_Cuenca) that two men were charged for a murder and tortured until they confess. 12 years later, the "murdered man" reappeared, having spent the time on a nearby town. There are more real life examples showing that people under torture says mainly what the torturer wants, not real life facts. Even in the presented case (the tortured subject truly has some critical information, which is somehow a "tricky case", real life is not that simple), the resulting information can be fake or inaccurate.

Information is more easily obtained in more reliable ways. Aborted terrorist attacks typically aborted in real life because someone will alert the authorities, or someone is being watched, not catching a "bad guy" and torturing it.

No, if the sought information is simple and verifiable, it most definitely works.

If you withholding a password, a torture would likely reveal it. It can be checked on the spot. There's even a relevant xkcd strip.

If you're a merchant hiding gold assets somewhere, a torture might just convince you to point the stash. And on and on, there were countless cases in human history when torture worked, propagating it's ubiquitous use as investigation method.

If the only evidence you have is in a password protected laptop then the case is flimsy yo begin with. You might as well add coin flipping to the case because that's what you are basically doing.

You realize you now reframing the argument to present simple evidence (which is often the most powerful one) as pointless?

There was a kidnapping case in germany where the police was absolutely certain that the suspect was the right one and that the only way to save the kidnapping victim (a bankers eleven year old son) was to torture the suspect. The suspect then admitted the kidnapping, but the victim had been killed shortly after the actual kidnapping.

One of the fundamentals of the german law systems was broken, by a well-meaning person for practically no gain. The kidnapper later sued the police for the torture and won.

Now germany doesn't have a fruit of the poisonous tree doctrine so the conviction for the kidnapping later was based partially on the information gained in that one interrogation, but what if the kidnapper would walk free because of such a breach of law?

[1] http://de.wikipedia.org/wiki/Entf%C3%BChrung_von_Jakob_von_M... (sorry, german only)

> The kidnapper later sued the police for the torture and won. > but what if the kidnapper would walk free because of such a breach of law?

In the U.S., a kidnapper walking free for something like this would result in a nationwide popular movement to insulate police departments from being sued for torturing suspects.

No way, wouldn't ever happen. In the US this is called a coerced confession and it is inadmissible in court as evidence, as the Supreme Court ruled in Brown v. Mississippi.

http://en.wikipedia.org/wiki/Brown_v._Mississippi

The confession would never be allowed, however if there were other evidence to implicate him, such as blood in his house, that would certainly still be admissible in court and enough to convict him. If however, there was no other evidence AT ALL, than he walks free, and he gets to sue to police.

There wouldn't ever be a movement to stop police from torturing suspects. This is generally frowned upon, even if they are guilty.

It's easy to construct Jack Bauer scenarios in fiction, but they are exceedingly unlikely to come up in real life. Real life is a multithreaded narrative - there's never just ONE lead to follow and you can never be that certain the lead you're looking at is the right one.

What made torture "work" in 24 is that Jack Bauer can't trust anybody else and never has more than one lead to follow at a time. Even though he talks to dozens of people who know parts of the big conspiracy, these people all magically disappear but only do so moments after they tell him one key bit of info that will lead him to the next clue. Sometimes they disappear by escaping custody, sometimes they disappear by being killed by the bad guys (or a traitor among the good guys), and sometimes they kill themselves. Through repetition of this pattern, we come to expect that each person who becomes the focus of Jack's attention will have some key bit of info he just has to get out of them. And he has to get that info from them now because he if he doesn't he'll never see them again.

Real life doesn't work that way.

In real life you have dozens, hundreds, or thousands of possible leads at any given time. Because the bad guys aren't supernaturally good at their job, you can go back and ask followup questions. You can check one guy's testimony against the next guy's. And some of the people you think seem like pretty good leads don't know anything useful. Torturing those people will generate false leads which lead you to torture more people who generate more false leads until your entire organization is in the business of serially torturing innocent people until they tell you whatever you seem to want to hear rather than the business of finding out the actual truth.

Each person you torture into generating false leads makes it more likely people will die because it ties up your organizational resources. It also is likely to generate massive blowback - some of the people you torture become or inspire new enemies, whereas people you are nice to might become friends and voluntarily help the investigation.

You have no idea what it is like in life. You have a guess but it is entirly wrong. But I do. Simply put I've been there, and no there not hundreds of leads. The people you 'torture' are not random guys, They were linked to the case in question. And the information they give is always always somthing you know and something new, They of course don't know which is which. Otherwise the information is usless and you will need endless amount of time and money to sort it.

Again. Your ideas about real life are entirely wrong.

One specific example of "torture works" would completely overturn everything that is known about the use of torture by the US. You are so very adamant that one would think you have that example ready to go. And yet you curiously stop before giving any examples, even though that invites thinking you might not be truthful. Strange.

Proving one is the intended target of bigotry is usually unhelpful.

ETA: Downvote demonstrates the point.

Bigotry? Bigotry against torturers? Seriously?

Bigotry is defining a group as inherently evil, and rejecting any sensible discussion with a member thereof on the grounds that someone is (or sympathizes with) one. Never mind whether he may have saved lives thru "enhanced interrogation", your curiosity about what he did exists only insofar as his admission thereof justifies your labeling him as evil (no room for good-faith discourse possibly resulting in "hey, maybe you're not so bad after all").

I find it amazing that there is pro-torture propaganda on TV, never mind that in the real world it is the darkest of the dark crimes against humanity. I find it amazing that the psychologists who designed tortures and the people at Guantanamo who carried out torture are walking free. Shouldn't they be on the run from the same kind of borderless manhunt as WWII era war criminals?

You say I'm not leaving room for good faith discourse. I asked for an example of torture saying a life. I asked because, time after time, in all forums, no such example has ever come up.

The real problem is that torture has been allowed to approach too close to where it's considered OK to use.

I don't see anyone claiming that torturers are inherently evil. I see people claiming that act of torture is inherently evil. I'm perfectly open to the idea that someone who has tortured someone else isn't evil, but torture is always an evil act.

And in your world how do you prove you do not know something ? I guess ... more torture ?

Maybe the fact that you did such things leads you to require to rationalize them. Just about everybody has reasons and is convinced they're correct, that is true even for Charles Manson.

A poem by Erich Fried comes to mind... "don't doubt those who say they're afraid, but be afraid of those who say they know no doubt" -- I guess doubly so if they put torture in quotes.

Where is this "there" that you've been? Are you saying that you, personally, have used torture to resolve a ticking time bomb scenario? Can you be more specific?

Why would it matter that the info they give is "something you know and something new"? How would that stop the "something new" part from being something that was invented to get you to stop torturing them? Of course the most realistic lies will include elements of truth and reference other stuff you think you already know.

And why would merely being "linked to the case in question" mean you can KNOW they have info they aren't willing to volunteer that would help at all, much less that would crack the case? (Sometimes a taxi driver is just a taxi driver.)

"They were linked to the case in question."

Not everyone "linked to the case" is a criminal, terrorist, despicable sub-human scum, or what-have-you.

That is not a popular idea in many areas of the military or law enforcement.

We all have witnessed how effective torture was in Guantanamo, with so many prisoners being tortured and released after they have been found not guilty.

I'm very curious to hear more. Contrary to this article, I don't see any evidence of "pro-torture propaganda", because everyone in this thread and almost every source I find is anti-torture.

The only interesting source arguing for the effectiveness of torture was this: http://wikileaks.org/wiki/Torture%2C_interrogation_and_intel...

It seems like torture would work, not in a "ticking time-bomb: scenario, but in a situation where you had a long time to corroborate intelligence gathered from torture with other sources. Interrogate the captive on stuff you already know about, and if they lie, punish them with worse forms of torture. And if people are prone to losing their minds under torture - you can merely submit them to one session of brutal torture, then leave them for a few weeks or months before resuming normal interrogation - with the threat of more torture if they don't comply.

Indeed, we should not underestimate the effect of police/agencies behaviour on regular citizen. It is something that shocked me when talking to friends freshly immigrated from countries like South America to London.

They have a general mistrust of cops, the more cop they see the less safe they feel. On the other hand, my wife and I, coming from other European countries are very happy to see cops around.

That said, since Tony Blair's "we can hold you in custody for a month without rights", or the we just shoot random people in the back after 7/7 bombing, our attitude is slowly changing. I'm not sure I would volunteer information to the police related to any hot topic without studying the risks first.

Well, consensus seems to be that you're wrong, but I think you're obviously right. However, you left off one of the most important conditions: the interrogator has to be able to tell that you've told him the truth. The scenario where torture would best work is when you pick up someone with an encrypted laptop and want the password that they have. In that situation, it will work.

>but I think you're obviously right

Did you read all the research presented in the article and here in the comments and come up with that conclusion, or are you just talking out of your ass?

What if they didn't have the password? This is a real life example. I'm going to use an example. This lady was stopped on an airplane because she was carrying explosives. Her husband packed her bags and she didn't know she was carrying them. What if it was a really important laptop? Would you torture her? Would you risk giving her a lifetime of mental and physical scars because you thought she had a password she didn't even have?

>In that situation, it will work.

You sure about that? Some people would rather be a martyr and die. Examples are all around.

Waterboarding can cause death by the way.

You're confused. The question is whether torture works. Many people in this thread and elsewhere have claimed that it never does because they wish it were true. I have not seen anyone even claim that torture can be resisted, let alone provide evidence of this claim. Please provide a link. What people argue is that you can't trust the information produced by torture, but this misses the obvious set of cases where trust is not required.

Please note: I am not arguing for the use of torture! You're right that in many cases the victim will not have the sought information. That doesn't mean torture doesn't work.

> Sometimes the subject does have the sought-after information

As far as I know this has never been the case in the history of the use of torture in US interrogations. Before you make statements in favor of torture, you should probably have a specific case in mind or it's just being gratuitously in favor of torture.

> Sometimes the subject does have the sought-after information, and the interrogator knows the subject has it, and people will die if that information is not extracted.

And even then, it is not the most effective method of extracting it.

> Another inconvenient fact is sometimes it does work

Sometimes it might, but there is no set of observable circumstances from which it can be concluded that torture will be expected to work better than other methods, so it is never a reasonable choice even excluding any negative moral considerations (or negative impact of more indirect effects of adopting such methods as policy) applicable to the method.

> and is an alternative to actual harm.

"Enhanced interrogation" is a euphemism for methods that involve actual harm, not an alternative to them.

>(with no physical harm done)

You're going to subscribe to the idea mental torture is somehow less than physical torture? Nope, wrong. Ask anyone who has suffered from a mental illness and ask if their illness was less than a physical illness.

One example.

http://science.howstuffworks.com/five-forms-of-torture.htm#p...

Number 1 - Mock Executions

It's against international law and The U.S. Army Field Manual expressly prohibits soldiers from staging mock executions.

>and the interrogator knows the subject has it

Oh really? How does this interrogator know? With 100% probability?

I'm not familiar with interrogation techniques, but I'm going to go out on a limb and say we study them like we study most things, with DATA and FACTS, not just some comment on the internet saying "they say Tell me! or I'll say 'Tell me!' again!'" because there is a good incentive to get the technique right, and people know more than me.

Sometimes the subject does have the sought-after information, and the interrogator knows the subject has it, and people will die if that information is not extracted.

Given that professional interrogators have come out and said that the 'ticking time bomb' situation does not happen in real life, this argument is pointless. Argue about the kind of tortures that actually happen, don't construct a fantasy scenario to prove a philosophical ideal and then assume that covers the ordinary case.

It's the same way with other things, like the recent privacy revelations regarding the NSA. I'm pretty sure I remember seeing thriller films in the 80s or 90s where the completely over the top, unambiguously evil government plan was spying on citizen's telephone calls. A couple of decades later, that becomes reality, and a huge number of people have no problem with it.

Reply on news.ycombinator.com