John Oliver discusses torture

[Belz...]

How many studies actually use torture in that study?

Would you agree that you can use data from previous torture to conduct such a study ? You seem to be implying that in order to study torture you need to conduct torture. Please not that I am not necessarily in disagreement with you.

 

[Ziggurat]

No, you're just suggesting that scientists are writing their papers, not so they reflect the research they have done, but to support a desired conclusion. You did not say "inherently dishonest", but that is what you described.

No, Upchurch. There are far more subtle ways that bias can enter into the picture, including in the very design of a study.

Of course scientists are interested in their subjects. Why would they study them if they were not? Interested, however, does not mean that they are biasing their work

In this case, actually, that's exactly what it means. You have confused disinterested (not influenced by considerations of personal advantage) with uninterested (not caring about something). Scientists should ideally be disinterested in the subject of their study.

or that they are being dishonest in the hopes of getting, of all things, grant money.

The thing about having an interest (benefiting from a particular outcome) in the subject of study is that it can influence you whether you intend to be dishonest or not.

 

[Upchurch]

In this case, actually, that's exactly what it means.

Why in this case, specifically?

 

[Belz...]

I think John Oliver is the apotheosis of the Truthy Comedian evolution of journalism, pioneered by John Stewart and Steven Colbert.

I enjoy all three of them but you have to be very careful about how they spin the news, sometimes. Often, in fact.

 

[Ziggurat]

Would you agree that you can use data from previous torture to conduct such a study ?

And how would you do that without controlling for other variables which will be extremely relevant to the outcome?

You seem to be implying that in order to study torture you need to conduct torture.

You would need to if you wanted to control for other variables, yes. And if you don't control for other variables, your results will not be reliable. Our understanding of psychology is far too primitive to be able to assemble a model of interrogation by testing individual components separately, the way that engineers can model a new structure based on the properties of its pieces.

 

[Ziggurat]

Why in this case, specifically?

Because I'm the one who introduced the word, and that's what I meant by "disinterested". I chose that word, and not the more commonly used "uninterested", precisely because I did not mean "uninterested".

 

[Upchurch]

Because I'm the one who introduced the word, and that's what I meant by "disinterested". I chose that word, and not the more commonly used "uninterested", precisely because I did not mean "uninterested".

No, why is this particular study suspect for its financial interests as opposed to, say, any other study ever?

 

[Belz...]

Do you think a funding agency will respond more favorably to a researcher who shows that torture works, or to a researcher who shows that torture doesn't work?

I should think that they will respond to a conclusion that fits the data.

 

[Belz...]

And how would you do that without controlling for other variables which will be extremely relevant to the outcome?

Not very well. But my point is that you _can_ use existing data for a study without the need to torture people. Our knowledge of dinosaurs is pretty damn sketchy but it's functional, for example.

You would need to if you wanted to control for other variables, yes. And if you don't control for other variables, your results will not be reliable.

Not very reliable but they could yield useful conclusions nonetheless.

 

[shuttlt]

Wouldn't evidence for the effectiveness of torture be in the records of fallen regimes that have used torture successfully?

 

[sunmaster14]

I enjoy all three of them but you have to be very careful about how they spin the news, sometimes. Often, in fact.

No kidding. I actually thought there was an audiobook version of the Senate Torture Report, because John Oliver said so, and I wasted a few minutes on Amazon looking for it. Man, if you can't even trust a sanctimonious, mildly amusing, British comic who has more than his fair share of trouble reading lines off the teleprompter without laughing or fumbling his words, who can you trust?

I'll admit he was right that the report is a terrible pick for a book club. I recommended it to my libertarian book club, and the idea seems to have gone over like a lead balloon. Looks like we'll be reading a lawnmower repair manual next month instead.

 

[Ziggurat]

No, why is this particular study suspect for its financial interests as opposed to, say, any other study ever?

Every study is potentially suspect. That's why seeing the quality of data is important. I can't see the quality of the data here, but the impression given by the abstract does not lead me to believe it's very good.

 

[jimbob]

Let's say, works, in the sense that the person will tell you something they'd prefer to keep secret and would do so without the application of torture.

Nobody was arguing that.

True, it is that people will say anything to get it to stop.

There's a 2,000 page intelligence report that says differently. And even Dick Cheney's best examples were debunked.

And there were examples where it was counterproductive.

Also we have plenty of confessions of witchcraft.

 

[Upchurch]

but the impression given by the abstract does not lead me to believe it's very good.

On what are you basing that impression?

 

[Ziggurat]

Not very reliable but they could yield useful conclusions nonetheless.

Yes, we can reach some very broad-brush conclusions by examining the historical use of torture. One of those conclusions is that torture can work. Another conclusion is that torture doesn't always work. That's not really enough to decide how well it can work versus other methods. We simply don't know, and are not willing to find out (and I'm not suggesting we should, either).

And again, none of this is an argument for torture. The point is that we cannot simply dismiss the problem by pretending that it doesn't work, because it does. If you want to categorically prohibit torture, you cannot do so on the pretense that it's ineffective. The primary reason to avoid using torture is morality, not efficacy. It's a bad thing to inflict pain on people. But given the moral imperative to find out information from very bad people to stop them from doing very bad things, that question is thornier than many people like to pretend. But brushing it away with a false claim that torture doesn't work isn't an honest approach to the problem.

 

[Belz...]

And again, none of this is an argument for torture.

I understand your position. I won't confuse arguments for the efficacy of torture with arguments in favour of torture, don't worry.

 

[Upchurch]

The point is that we cannot simply dismiss the problem by pretending that it doesn't work, because it does.

I don't believe anyone has argued that.

 

[shuttlt]

It's a bad thing to inflict pain on people. But given the moral imperative to find out information from very bad people to stop them from doing very bad things, that question is thornier than many people like to pretend. But brushing it away with a false claim that torture doesn't work isn't an honest approach to the problem.

If you ask me, it's not particularly thorny. September 11 was clearly a hugely successful terrorist attack, but it is so noteworthy in part at least because it is such a unique event. Are terrorists an existential threat to any western country? Their real successes in getting large numbers of people killed have been by provoking governments into conventional wars. I just don't see that they are worth sacrificing the moral authority of a country over.

 

[Ziggurat]

I don't believe anyone has argued that.

Then you haven't been paying attention.

John Oliver discusses torture and how it doesn't work

Well that's the nub. By all accounts it's not necessary because it doesn't work.

 

[The Don]

Then you haven't been paying attention.

Yes "it doesn't work" because it's less effective than other interrogation techniques and the noise to signal ratio is too high