Friday, May 11, 2012

Torture - Part 2: Evaluating Acts vs Evaluating Desires

I am currently discussing the issue of torture - attempting to answer a question from the studio audience that asks:


I was wondering what your thoughts about Sam Harris' arguments for torture were. He seems to proffer the view that, by accepting collateral damage from war as inescapable yet demonizing the collateral damage from torture, we are holding to a very irrational double standard. I find his defense of torture emotionally disturbing, but I can't seem to produce a logical argument that counteracts his thesis.


Yesterday, I examined Harris' moral theory. Harris uses a form of act-consequentialism that says that the right act is the act that produces the best consequences. The best consequences, in turn, are cashed out in terms of the well-being of conscious creatures. If torturing a person brings about the best consequences, then torturing him is the right thing to do.

One well-known argument against this type of defense of torture asks, "If it is okay to torture the terrorist to reveal the location of the bomb, is it okay to torture the terrorist's child?" The act-consequentialist would have to say,"Yes, we may torture children whenever (we believe) it would provide the best consequences."

According to many peopke, this utterly discredits act-consequentialust theores. To some critics of Sam Harris, it proves that one does not have to believe in God to embrace a bankrupt account of right and wrong that leads to horrendous consequences. It shows that some secular moralities are as bad or worse than amything found in religion.

A possible response that Harris could make is, "Look, we know that wars result in a great deal of collateral damage. We have blown up buildings with children inside to get at the terrorist who is also in the building. Why, in the case of torture, are we saying that the brutalzation of a child is absolutely to be prohibited?"

I do not know if this is the same argument that the author of the question was referring to - or if Harris has given this answer. However, it fits the description, and I think it is worth investigating.

Let me give an example in which act-consequentialism clearly argues in favor of torturing a child. Not only that, but many of us otherwise opposed to torturing a child might agree that it is necessary in this case.

You and a young child have been abducted by aliens. The aliens offer you a bargain. Either you torture the child to death – making sure that the torture is excruciating and lasts at least four days, or the aliens will release a bacteria on Earth that will cause everybody on Earth to die a slow and agonizing death lasting at least seven days.

Is it then permissible to torture that child?

I suspect some people will say yes. A few absolutists woukd say no and struggle to defend their answer.

However, for my purposes, the answer to that question is not important. I want to focus on a different question.

Should he enjoy it?

From an act-consequentialist point of view, he should enjoy himself. The person who gleefully tortures the child produces better consequences overall than the person who reluctantly tortures the child. In both cases the child is tortured. We may assume that both perform the same actions (perhaps using a script provided by the aliens). The only difference is that, in one case, the torturer experiences inner joy, while in the other he suffers inner turmoil. Joy is better than turmoil, so the former is a better state of affairs than the latter.

Now I have a follw-up question. What do you think a community would be like if it were filledwith people who would gleefully torture a child under these circumstances?

Our desires do not simply turn themselves off and on. In order for an agent to be so constituted that he would cheerfully torture a child in these circumstances, he will have to be te type of person disposed to cheerfully torture a child in these circumstances even as he sits in his house next doir watching televsion, or teaches third graders at the local school, or coaches the soccer team.

We have no good readon to tell these peopke, "I want you to be the type of person who would cheerfully torture a child under these circumstances." Instead, we have a great many reasons to tell these people, "I want you to be so constituted that, in these circumstances, perhaps you will torture the child, but it would be a horrendously traumatic experience for you. In fact, if you were to kill yourself afterwards, I will take that as evidence of your virtue."

Going further, we may have good reason to say to each other, "Given that the alien abduction scenario is never going to happen, and given the huge number of real-world situations in which we have real-world reasons to want people generally to be strongly averse to harming a child, we want people to be so constituted that, even if they were abducted by aliens and given this bargain, they would not be able to go through with it. The alien abduction scenario is not going to happen, but there will be countless real-world interactions in which an eagerness to see a child suffer will have relevance. "

Here, then, is the argument against torture. A wide-spead social acceptance against torture requires lowering the overall aversion to cruelty. This lower aversion to cruelty is going to reveal itself in countless human interactions outside of the torture chamber. It will reveal itself on the school playground, on the streets of Los Angeles during rush hour, at the bar on a Satuday night, at the soccer game, in every interaction between human beings, to one extent or another."

There is a reason why people who are comfortable with torturing the terrorist suddenly grow uncomfortable at the thought of endorsing the act of torturing the terrorist's child. We do not want to be surrounded by people who are capable of that kind of indifference to the suffering of a child. We do not want to promote and encourage others to adopt the attitudes that they can happily go along with torturing the child. Instead, we have good reason to surround ourselves with people who recoil at such a thought.

Is it the case that cultures that embrace torture tend to be cultures where individuals experience more cruelty and less kindness in general?

We seem to have evidence of this type in the issue of capital punishment. Cultures that embrace capital punishment tend to have more murders. This could be because getting people to embrace capital punishment requires lowering their aversion to killing. They have to tolerate, and even cheer, deliberate acts of killing. A culture whose people have less of an aversion to killing is one whose population will find it easier to kill. Thus, its people will commit more murders (all else being equal). One way to reduce the number of murders may well be to create such an aversion to killing that the population recoils even at the thought of capital punishment.

If a culture that embraces torture tends to be more cruel and less kind, then all of the cruelty and the absence of kindness found in a culture that embraces torture would count as reasons not to embrace it. They count as reasons to say of one's neighbors, "I want all of you to be so constituted that, even in this hypothetical case involving the terrorist, you could not participate in this kind of cruelty." We have reason to say this because our neighbors will almost certainly never find themselves in a situation where they need to torture somebody to get information about a bomb. However, they will find themselves in countless situations where their overall kindness and cruelty will play a role.

These points illustrate the key difference between an act-consequentialist moral theory (the right act is the act that produces the best consequences), and a desire-based moral theory (the primary object of moral evaluation are desires - such as kindness and cruelty - and acts are evaluated according to whether they are acts that a person with good desires would perform). The former looks at the collateral damage that may result from an act of torture. The latter looks at the overall social effects of being the type of person who could participate in or celebrate and endorse an act of torture.

In conclusion, I want to note that desirism itself does not dictate an answer to the question of whether torture is legitimate. That answer depends on the outcome of empirical research such as studies linking support for terrorism with cruelty and an absence of kindness. Desirism does not provide a fixed answer on any moral issue - from capital punishment to homosexual marriage to secularism to blowing up houses where a family has gathered because a family member is a suspected terrorist. It provides a way to answer those questions, but it does not dictate any answers.

7 comments:

Gingerbaker said...

"The act-consequentialist would have to say,"Yes, we may torture children whenever (we believe) it would provide the best consequences."

No, you just moved the goalposts off the field and into another stadium. Harris believes torture is justified by the ticking bomb scenario, and he does not go further.

Alonzo Fyfe said...

Gingerbaker

Please note that in the sentence you cited I used the term "act-consequentialist", not "Harris".

Harris' defense of torture in the ticking bomb case IS act-consequentialist. Harris may not be willing to follow act-consequentialism to its logical conclusions, but these are the logical conclusions of act-consequentialism nonetheless.

Jesse Reeve said...

The only difference is that, in one case, the torturer experiences inner joy, while in the other he suffers inner turmoil. Joy is better than turmoil, so the former is a better state of affairs than the latter.
Does Harris say that joy is better than turmoil? Personally, I think that's like claiming that satiety is better than hunger. If my body needs food, I prefer to feel hungry, even if it hurts. I prefer turmoil to joy if turmoil is appropriate to the situation-- such as the situation of having to choose between torturing a child and the end of the human race.

A wide-spead social acceptance against torture requires lowering the overall aversion to cruelty. This lower aversion to cruelty is going to reveal itself in countless human interactions outside of the torture chamber.
I'm not convinced that feelings and acts belong in the same category. Partly for the reasons you describe-- that feelings come from desires that can't be changed nearly as easily as a decision can (at least in terms of cognitive effort required). But even if I stipulate that feelings are a subset of acts, can't an act-consequentialist still treat the effect on those other human interactions as a consequence of her act?

Or is it that I'm being too generous to act-consequentialism? Does act-consequentialism focus on the immediate circumstances of an act only, and disregard their personal and social context?

Alonzo Fyfe said...

Jesse Reeve

I prefer turmoil to joy if turmoil is appropriate to the situation-- such as the situation of having to choose between torturing a child and the end of the human race.

You will need to define what you mean by "appropriate to the situation".

Desirism, as described here, says that being so constituted to feel turmoil in torturing the child is better - morally better, more virtuious, a state that people have reason to want others to be in - than being so constituted so as to feel joy in torturing the child.

So, I also prefer turmoil to joy in this context.

However, in the case where the choice is between joy itself or turmoil itself - the red pill will give you a feeling of joy and the blue pill a feeling of turmoil, and you must take one or the other - joy would in.

[C]an't an act-consequentialist still treat the effect on those other human interactions as a consequence of her act?

It would be a tortured use of language (please pardon the pun).

These other effects would exist even if the act in question is never performed. They come from being the type of person who would experience joy at torturing the child, even if one never finds oneself in a position of being abducted by aliens and told to choose between torturing a child or bringing about the destruction of the world.

It would be hard to justify calling X a consequence of Y under conditions where X exists even when Y never happens.

Jesse Reeve said...

You will need to define what you mean by "appropriate to the situation".

That's a tall order. Tentatively, for an emotion to be appropriate to a situation means that it is consistent with the response of the agent's values to the situation. Maybe "consistent" is a better term to use, less likely to be misunderstood. If Alice supports same-sex marriage, and she hears about a state passing same-sex marriage into law, it is consistent for her to feel happy. If Bob opposes SSM, it is consistent for him to be disappointed when he hears the news.

Whether Alice's or Bob's values are right is a different question from whether their emotions are consistent with their values. I would say that Alice's value judgment is right and Bob's is wrong. But if Alice's emotional response to the news is disappointment, then I would say it is inconsistent, because it doesn't match the relation between her values and beliefs, and if Bob's emotional reaction is disappointment, I would say that's consistent.

Emotions are one of the most powerful motivating forces on humans, so we have reason to keep our emotional responses consistent with our values. A joy pill would be a fun recreational substance. I might even take it while torturing the child in your hypothetical, if I couldn't make myself go through with it without chemical assistance. But a pill that made joy my emotional reaction to torturing-- that, I would avoid like poison.

These other effects would exist even if the act in question is never performed. ... It would be hard to justify calling X a consequence of Y under conditions where X exists even when Y never happens.

Thank you, Alonzo. This is exactly the type of explanation I was looking for. I don't entirely agree; there are are examples of X occurring as a consequence of Y, even if Y never happens. Bob the Bully might choose not to bully Assertive Alice if he believes she will fight back, even if the situation never actually arises. But act-consequentialism doesn't seem to offer a convincing account of acausal influence, so that doesn't save it from this argument.

Justin said...

I heard a good philosophy bites the other day that dealt with this very question. I thought it was a good solution.

Because torturing a prisoner who may know the location of a bomb happens very, very, very rarely (outside of hollywood), it ought to be prohibited and illegal. This is because it would be easy to believe that the circumstances justify that you torture a prisoner, when that's simply not the case. There would be more torture involving, if not innocent, at least guilty people with no useful information.

If you really do have somebody who has the information on the location of a bomb that's going to go off in a short period of time, then go ahead and torture them, understanding that what you're doing is wrong, and you'll face legal retribution. If you manage to avert a disaster by torturing the terrorist, then that can be brought up in your defense, at your trial.

Kristopher said...

thanks everyone this has been a really good discussion of some things that had been rolling around in my head and you really helped to me focus my thinking.