Bias Against Torture

In many situations it would be better to impose a punishment of torture than imprisonment.  The fact that the U.S. justice system rejects torture as a punishment is the result of an anti-torture bias.

Torture has two benefits over imprisonment.  It’s cheaper for the state to impose and it doesn’t prevent the criminal from engaging in useful labors (such as parenting and working at a job) for long periods of time.   To determine who should be tortured as opposed to imprisoned we need to consider the benefits to society of imprisonment. 

Prison serves three purposes: deterrence, retribution and incapacitation.  Fear of prison deters many would-be criminals from committing crimes.  Fear of torture, however, could do likewise.  Imprisoning criminals can satisfy victims’ desires for vengeance and so make victims feel better.  Torturing criminals could, however, also satisfy victims’ desires for retribution.  Finally, prison prevents imprisoned criminals from attacking people who are not in prison.  The primary disadvantage of torture is that it doesn’t result in the incapacitation of criminals and so leaves them free to strike again.

Many convicted criminals, however, don’t pose a risk to society.  Men convicted of securities fraud, for example, are frequently barred from the stock market and so their freedom won’t endanger society.  Because of its far lower cost, the U.S. should torture rather than imprison criminals who don’t need to be removed from society.

Some would argue that it’s excessively cruel to torture criminals.  But both prison and torture impose costs on criminals.  Why is one type of cost crueler than the other?  If a convicted criminal is indifferent between receiving a certain type of torture or being imprisoned for a given period of time then why would it be excessively cruel to torture but not to imprison?

In the U.S. many prisoners face a significant chance of being raped by a fellow inmate.  This high chance doesn’t seem to bother many people, and is often the subject of jokes.  Yet our society considers it barbaric for a criminal justice system to deliberately torture criminals in ways that may well impose less physical and emotional costs than rape does.  I find these conflicting moral views about torture and imprisonment to be irrational.

Comments

[MJV (Finland)]

There's nothing new under the Sun, or somefink. Michel Foucault wrote "Discipline and Punish" in 1977 (see Wikipedia for original work), and the book touches this very theme on a very moralistic way. (however, it doesn't have that much calculus of probabilities, though)

I see that there isn't much lawyers around, otherwise somebody would have had brought up his name earlier. Personally I detest the book (and the man), as it can be seen to advocate for all-out abolitionism, á la "ivory tower"-way. (or then it was "propagandist" work, based on Foucault's beliefs on what the communist system ought to be, instead of what it really was)

[Nick Bostrom]

A disgusting thread to my mind, but I guess it's in the nature of the forum that cold-hearted deliberation must be enabled even on such topics.

I think of moral advancement as proceeding by a strategy of divide and conquer. Slowly and with much effort, moral decency sometimes manages to erect a barrier against some particular kind of cruelty or barbarism. But this progress is always at risk, and short-sighted expediency will easily cause people to give up the moral standards that were achieved after so much sacrifice. This is one reason why it can be important to stand up for particular principles, even if they seem inconsistent with our conduct in other areas. I can easily imagine that a future more enlightened age will regard life-time imprisonment, in prisons with high rates of violence, without any chance of parole and without any attempt at rehabilitation, as a barbaric violation of human rights, and that these ages will develop more humane ways of reducing crime.

I recommend Jonathan Glover's "Humanity: A moral history of the twentieth century" (Routledge) for a fine discussion of the moral resources that we possess, and which we need to cultivate in order to avoid repeating the atrocities of the distant and recent past.

[mtraven]

Every governmental instrument requires a professional class to administer it ...
This is true, and a dangerous fact. Government has a monopoly on violence and employs people whose job is to administer violence, such as police officers, soldiers, and prison guards. These professions are vital, but dangerous if they get too much power (resulting in police states or military coups). That's why we have strict laws and professional cultures that restrict what they can do. But there's always the danger that they will break out of the boundaries society places around them.

So, should we create a profession of torturers, and hope that they stay within their bounds, that they develop a culture which takes pride in their professionalism and their tradition of honor and service? It strikes me as a dubious proposition. Torture destroys the humanity of the torturer just as well as it does that of the victim.

Gene Wolfe drew a convincing fictional portrayal of a torturer's guild in his excellent Book of the New Sun. But even there the ultimate verdict on their efficacy was negative.

[TGGP]

mtraven, would creating a Torturer's Guild be acceptable on the condition that the Jailer's Guild be abolished?

[mtraven]

No, why should it?

I liked Nick Bostrom's comment. We have indeed slowly achieved a somewhat higher standard of civility than our ancestors. Progress has not been monotonic, as the various industrial-scale barbarisms of the 20th century have shown. But in fact we have mostly weaned ourselves from torture, capital punishment, punitive mutilation, and the like. We should not deliberately try to reverse this ratchet.

BTW, in Wolfe's book, economics was one of the arguments in favor of the torturers -- it was just too impractically expensive to jail people. But that was a society in a deep decline.

[Pseudonymous]

I suspect that “our civility” is due to there being less need for such things today.

I like to point out the Jane Austen considered it perfectly normal to send children to war. In Mansfield Park, as an example, the brother of the heroine is sent of into the fleet as a midshipman at the age of twelve. Their family was moderately well off and called in favours, so that this could happen. Everyone considered this a great favour, and by the time the novel ended a younger brother had happily joined him.

This should give some idea about what the alternatives were like. In a family that could afford servants.

Our compassion is to some extent a consequence of the fact that we are in so much less danger ourselves. It is easy to be generous when you are safe and rich.

[Hopefully Anonymous]

Should we declare this whole thread an empiricism-free zone? Eliezer, there might be a useful post for you to write on how people prefer certain topics to be intellectual playgrounds where empiricism isn't introduced. It's interesting to me how people are arguing multiple perspectives, but no side seems interested in making the topic or their position informed by empiricism.

[Anónima]

There certainly is a bias against torture in our occidental world. But this is not true world wide. So, following the suggestion of Hopefully Anonimous,it could be a good idea to compare scientifically the results obtained by countries were torture is forbidden and the results obtained in countries were it's part of the legal system.

To do it well could be difficult: one would need to define what parameters one must measure to assess the success of one method or the other, collect data and so on. And then perhaps all ways of applying torture don't get the same results.

But, without any scientific foundation whatsoever, I suspect that they don't really get a better result in fighting, let's say thieves, by cutting their members than we do by putting them in prison in those countries that judge and condemn people according to the sharia law. But it's of course, a biased conclusion that should be backed with raw data analysis. I'm sorry I haven't the time to do it.

Also I suspect I won't be very wrong if I say that no one here, even the person that wrote the post, would prefer to live in one of those countries, what I find as a strong, if not scientific, argument against the use of torture in a legal system.

[Douglas Knight]

HA,
James Miller seems to me have made an entirely correct argument, admitting exactly where empiricism affects it, although it would be better if he'd gotten it right the first time, rather than admitting new details whenever pressed. But empirics are only relevant to deciding the correct policy, which is not, I think, an appropriate topic for this blog.

What is an appropriate topic is how people and states reach opinions and policies. As Eliezer said, "I wonder why I think that." I'm not sure why I'm against corporal punishment. It's probably based on heuristics that embody empirical claims. But I can't worry if those claims are correct before identifying them. James's last paragraph about people's incoherent views on prison violence has little to do with actual rates of prison violence, but mainly people's perception of it.

[Hopefully Anonymous]

Douglas, I don't see what you claim to see in your post. Specifically, I don't see that Jamese miller "made an entirely correct argument" and I don't see that he was "admitting exactly where empiricism affects it". And I think empirics are relevant for more than "deciding the correct policy", I think they're also relevant for determining whether a particular position is biased or not, and for determining "how people and states reach opinions and policies". Empirics aren't limited to the "actual rates of prison violence", they're also the best way I'm aware of to determine "people's ... views on prison violence", "people's perception of [prison violence] and the degree to which those views are "incoherent".

I don't get this desire for empiricism free intellectual play spaces, but it does seem to be a real phenomenon, and I think Eliezer would be a great person to dissect and get to the heart of this phenomenon.

[Floccina]

I do not know about torture but I have often wondered why some prefer life in prison with no chance of parol over death.

[Douglas Knight]

HA,
In your second comment, you said "torturing...may reduce your and my likelihood of personally being tortured...It's an empirical question though." Certainly, empirics are necessary to answer this question. In theory, a sufficiently controlled experiment could directly answer it, but in practice, our best guess answer must be produced by plugging in more specific empirics into a theoretical framework.

I admit that this conversation did not much increase my understanding of how people and states reach decisions. They are so far from logic that holding up a logical argument for comparison may not be helpful.

You are right that my last sentence does not make sense. James Miller's last paragraph is making an empirical claim. It is hard to draw lines between empirical claims and inference, but it seems to me that that paragraph requires very limited empirical claims about people's beliefs.
What would it mean to be "transparently rooted in empiricism"? That he say how he reached these beliefs?

[TGGP]

I liked Nick Bostrom's comment. We have indeed slowly achieved a somewhat higher standard of civility than our ancestors. Progress has not been monotonic, as the various industrial-scale barbarisms of the 20th century have shown. But in fact we have mostly weaned ourselves from torture, capital punishment, punitive mutilation, and the like. We should not deliberately try to reverse this ratchet.
The Marxists believed communism was the endpoint of dialectical materialist history. Hegel believed the Prussian state was in his time, and Fukuyama believed market democracy was rather recently. I do not know what the future will be like. It is possible (though I do not think likely) that they will have replaced imprisonment with torture and regard our time as barbaric for locking people in jail for years at a time with the most violent people rather than a quick administration (done in a highly professional and scientifically optimized manner) of disincentive. Would you be willing to say to these hypothetical denizens of the future "Yes, our ancestors had completely wrong ideas about morality and crime and punishment. Yes our system was flawed. But no matter what you think your system is much worse than ours!". What argument could you present to this hypothetical man (or woman, or who knows what genders and species may exist) of tomorrow that the unenlightened folks of the past could not similarly?

[Hopefully Anonymous]

Douglas, I generally agree with your most recent comments. I like this line " in practice, our best guess answer must be produced by plugging in more specific empirics into a theoretical framework." To answer your question to me: yes, exactly -that he didn't transparently state how (or even if) he reached his conclusory statements from empirically derived information.

[mtraven]

For one thing, saying there has been gradual improvement is not the same thing as saying we've reached some sort of an endpoint. If I had to imagine the future of penology, my guess is that prisons will largely disappear, since the same functions of restraint and monitoring can easily be done with implantable trackers and electronic monitoring technologies. This is already happening on a small scale.

You suggest torture will be performed in a "highly professional and scientifically optimized manner". Maybe. I'm talking about how torture is practiced now, which is often in a professionalized and highly pseudoscientific manner that corrupts all people and institutions involved with it. You can read about how the field of psychology was co-opted and corrupted by the CIA's Cold War torture experiments in this recent book. Somebody wanted empirical evidence, I suggest you start here.

[Prakash]

If I can take a slight liberty in interpreting this topic as alternatives to present incarceration which might work better, I am in favour of supervised working prisons.

Prison as of now does not allow people to work and pay back their debt and make the victim whole. Now, not all crimes can be compensated this way, but this approach reduces costs and makes sure rehabilitation is on a better track than today because of learning useful skills, and getting a chance to get back into society. Also, the victim receives a compensation.

[Torben]

I think one of the key differences between a lifetime in prison and supervised torture is one akin to what Sam Harris calls the perfect weapon (when distinguishing between the collateral damage of Western air raids and an equal amount of mayhem produced by blowing up a bus with a tummy belt full of nails). Nobody wants rape to happen in prisons; it's an accidental and perhaps unavoidable byproduct. A noteworthy distinction is that it is private citizens and not the state perpetrating these acts. Crucially, we don't think in terms of maximizing economic benefit at all times and (moral) costs.

Say someone came up with a solid, mathematical proof that sending the poorest 15% of the population into forced labor camps was economically beneficial to society as a whole and perhaps even to (most of) the people at risk. Would this be a wise path to pursue, even if some of the laborers preferred this?

Incarcerating people for long periods of time is the least invasive measure we have in deterring and preventing crime. Prisoners' bodies and minds are not breached, their personal integrity is not compromised. We have not dehumanized them, but only taken the minimally intrusive precautions to protect society from their doings.

Finally, European prison systems don't normally dish out as long prison sentences as the US; many rehabilitation efforts are also more succesful. Furthermore, incarceration seems to be the least cost-effective means of preventing crime; improving high school completion rates the most. Why only consider the most provocative alternative to incarceration? It seems to me similar to assume the most provocative explanation of differing cognitive test outcomes. It seems to me a more rational approach is to encourage crime prevention, reduce incaceration rates and prevent prison rape.

[itchy]

Many convicted criminals, however, don’t pose a risk to society. Men convicted of securities fraud, for example, are frequently barred from the stock market and so their freedom won’t endanger society.

So you propose torture only for white-collar criminals? In what way are they not a risk to society upon being freed? It would seem they'd be just as likely to re-commit their chosen crime as a non-white-collar criminal would be. In what way is torture a better deterrent, especially, if, as your comments suggest, the criminal would prefer torture over imprisonment?

And given that the knock on white-collar sentences is that the criminals are shuffled into 'Club Med'-style prisons, why do you think a sentence of torture would amount to more than a slap on the wrist?

Personally, I don't look to the prison system to provide retribution or much of a deterrence. That's for medieval types. Prison is about incapacitation, and, hopefully, rehabilitation. And, yes, I'm willing to pony up for that, just as you're willing to pony up to send a huckster back out on the street to bilk us all out of money through the back door by pretending you've deterred him. Just because it doesn't show up on the budget doesn't mean you didn't pay.

[Scoop]

I think the rehabilitation discussion has it entirely backward. From all that I have read and heard, a stint in prison, surrounded by criminals, is far more likely to harden criminals and encourage criminality than it is to rehabilitate anyone. I'm not sure I support the torture option, but it would seem that keeping criminals apart would be a major argument in its favor -- unless anyone out there has data that argues the other way.

[DaCracka]

As a person who has run afoul of the law once or twice, corporal punishment would be preferrable to incarceration. Primarily because the state would decide an appropriate amount of punishment relative to the crime. In a D.O.C. facility, physical harm is frequently administered by other prisoners as well as gaurds. Judges aren't 100% fair, but I'd trust their judgement overall more than the judgement of gaurds and inmates.

Before joining general population, I was stripped naked, made to take a shower in scalding water, and washed with a harsh soap. I'm sure designed to limit (not eliminate, which would be impossible in a lock-down facility,) antibiotic-resistant staph infections, which were prevelant enough to warrant a pamphlet. Staph costs lives and limbs, which would seem an extreme punichment for any crime short of murder, but is acceptable as a necessary side effect of imprisonment.

Despite someone's thought that prison rape is a fantasy of the writers on Oz, that, also, merited a seperate pamphlet, with suggestions for avoiding involuntary sodomy.

I don't know of any judge who would advocate that violation, even in retribution for a rapists' crimes. Even so, a judge would, presumably, pre determine the appropriate duration, location, and medical concerns involved.

Finally, this experience was BEFORE a trial, before I was "proven guilty in a court of law." Before I faced a jury of my peers.

Torture seems a little less evil, when you look at both sides of the issue. A caning seems downright humanitarian in comparison.

Unless we can garuntee no rapes, infections, and beatings, imprisonment seems worse than torture, because it's inflicted randomly and without concern for benefit or rehabilitation. No-one is stepping in to end this expiriment, and it's high time we did.

(I don't advocate torture, or corporal punishment, but I think the current option is worse, in many ways.)