It’s been a couple of days since the Senate released the torture report. The discussion in the press seems to concern (a) whether it really might be effective, (b) whether that doesn’t miss the point, that it’s wrong and that we should take the stance “we don’t do that”; (c) whether the partisan bickering about the report–is it accurate? will it hurt us internationally?–will undermine any broader significance it might have; and (d) how other countries might respond to it–with violence, prosecution, admiration, etc.
A few days back I posted on my Facebook page a link to a piece in The New Republic entitled “We Will Never Know Whether Torture Works. That Shouldn’t Matter.” A friend then asked me if it was really true that its effectiveness doesn’t matter. As he put it: “[T]he use of a flamethrower on [a] bunker is to protect the lives of one’s own soldiers [and citizens], while in the classic “ticking bomb” scenario the use of torture is to protect civilian lives. So maybe there’s more symmetry between the two cases than I’ve usually thought. But the difference remains that flamethrowers are effective in clearing bunkers, while torture is of questionable effectiveness at best. Would we consider flamethrowers acceptable were they ineffective, though still horrifying brutal, weapons? I think not. And would we consider torture permissible were it foolproof? Perhaps. So I’m not sure I agree with the article’s conclusion that the question of effectiveness is irrelevant.”
I thought this an excellent way of challenging the thesis that I had, I admit, too reflexively endorsed. So I wanted to explore what the best answers were. I’m no expert in the ethics of torture. I’ve read a bit about it, but it’s not something on which I actively work. Still, it seems to me that one can rehearse the main possible lines of argument quickly. Doing so leads me to this tentative thought, that torture is bad for reasons having to do with its connection to political abuse. But I’d love comments from members of this web site. And I should add: curiously, there has been no post on this topic since Ralph Wedgwood posted in 2008. So maybe it’s time to do it again.
Here, then, are my main thoughts: One reason torture might be prohibited is that it requires the torturer to aim at and track the infliction of pain, which should be an evil. (Think Nagel on the DDE). But there are many problems with this answer. One: there might be medical cases that require something similar (“Does this hurt?…”) but would be acceptable if done for the good of the patient. Why isn’t the good of others sufficient reason as well? Two: if it is so hard on the agent to torture, then give agent’s an option not to do it. That doesn’t require a paternalistic refusal to let them do it if they are willing to sacrifice themselves for the greater good. (Think Scheffler’s criticism of agent-centered restrictions.) Three: the Catholic idea that one simply may not aim at evil seems appealing only if one buys into the whole Catholic system of thought about practical reason. If, instead, one accepts, as I do, the secular premise that deontology must be fundamentally patient focused, rather than based on a concern with the agent and her connection to the good, then this line of argument just won’t get a grip.
Another potential reason is that torturing someone uses him simply as a means of achieving other ends, not his own. This too doesn’t seem convincing. Punishment that aims at deterring others from committing crimes likewise uses some simply as a means. Yet I still believe in the justice of punishment when it is proportionate and the criminal “deserves” it, or has at least forfeited his right not to suffer it. When dealing with a terrorist, one might think they too have forfeited their right not to be used as a means of stopping further harm (at least harm related to terrorism). So while this justification works for a prohibition on torturing the innocent, it does not work for a prohibition on torturing terrorists to get information that would prevent terrorist acts from succeeding. Of course, it is sometimes hard to be sure that one has a real terrorist as the potential torture victim. But it can be hard to be sure that one is punishing the guilty too. Set the standard of proof high enough, and get on with it, one might say.
Another third potential reason is that torture is just too bad a thing to do to a person, worse than death. But is it really? Most who survive it are presumably glad to have survived it. That is, they would rather not have been killed. Nor is that likely to be the present glossing over the horror of the past. I imagine that many would not choose death at the time, or, if they would, they would feel that doing so was just momentary weakness.
With that, I’ve run out of basic hypotheses as to why torture is intrinsically impermissible. Maybe I’ve missed something, but I’m drawn to suggest an extrinsic hypothesis: that it is the sort of tool one does not want to admit into one’s toolbox, because of the potential for abuse. While there may be cases in which it is justifiable, there is probably a slippery slope into being the kind of state that allows torture of domestic enemies. (Nor should a state permit private torture, again, given the potential for abuse.) Such states are common enough, and they are all horrific in terms of basic political freedoms. Moreover, it’s hard to believe that this is just correlation. It seems, rather, that once one starts to think that those who are scary can be tortured (and it’s noteworthy that the public seems more open to torturing “them,” after having seen scary beheading videos), the list of scary people will grow to include the likes of Nixon’s enemies list. Then our democracy, already pretty shaky, will be done.
Perhaps then it is best to say what sounds, on its surface, so thoughtless, and yet is said so often: we should be able to say of ourselves: “we don’t do that.” As a statement of “values” (or better: principle), it is a pretty important place to draw a line. Yes, it might be that the line presupposes that it’s not often a tool of great importance. So effectiveness cannot be kept completely out of the discussion. But as long as we have adequate options in general for getting information (and I both think that we do and that there’s no other plausible reason to endorse it; certainly, it seems unnecessary as a form of punishment), it should be categorically banned. That’s the best I’ve got… either that, or I’m missing something, or Cheney is right.
Comments?
First, if you haven’t already, I strongly encourage you to read David Sussman’s brilliant and frightening “What’s Wrong with Torture” (2005, Philosophy & Public Affairs 33). Sussman argues that torture is uniquely bad because it “forces its victim into the position of colluding against himself through his own affects and emotions, so that he experiences himself as simultaneously powerless and yet actively complicit in his own violation. . . . Torture turns out to be not just an extreme form of cruelty, but the pre-eminent instance of a kind of forced self-betrayal, more akin to rape than other kinds of violence characteristic of warfare or police action.” (p. 4).
Second, I think that any argument for the intrinsic impermissibility of torture is bound to fail. This is because we can always conjure up a scenario under which (1) we are confident that torture will be effective in obtaining some information and (2) that information is necessary to prevent a catastrophe. (To get your intuition going, make the torturee as nasty as you want and the catastrophe as large as you need.) Since such a scenario is possible the act cannot be intrinsically wrong.
But I don’t think that that’s ultimately that interesting of an insight. In the actual world torture doesn’t seem to be an effective way to obtain intelligence. The SSCI report tells us this. Thus it is unlikely, as a practical matter, that a scenario would arise in which its use is permissible.
What is crystal clear, though, is that this program has damaged the U.S.’s moral standing in the world and our ability to exert soft power. And given that the only good argument for torture is a consequentialist one, the harm that the program has done to our national security ought to give future policymakers, contemplating something similar, pause.
Thanks Thomas. I did read David Sussman’s piece when it came out. It may capture something important about the horror of torture, something that goes beyond the pain involved. Yet it still seems less bad than killing, which, again, we allow to be intentionally inflicted.
And note, your second and third paragraphs essentially concede the point that the issue IS its effectiveness. And the problem with citing the SSCI report is that its conclusion is disputed, and not only by the likes of Dick Cheney.
So I come back to my question: does one have to believe the empirical claim that it is unlikely EVER to be effective to adopt a ban on it, or is there some other sort of argument that can be used to ban it, or should we admit that sweeping bans overstate the case against it and be open to allowing it sometimes?
Thanks for the post Alec. Like you, I haven’t read everything on this. Despite your doubts, I thought that there may be many facts about torture that strengthen the prohibition on torture.
First, you rightly claim that it is sometimes permissible to aim at pain in a medical procedure. But that would not show that it is generally unproblematic to aim at pain – it might be more difficult to justify aiming at the pain of one person to benefit others than causing pain without aiming at it. I don’t say this is right, only that your example doesn’t show that it is wrong.
Second, you, I think, are tempted by the view that as death is worse than being tortured, and it is sometimes permissible intentionally to kill a person for the sake of some end, it is sometimes permissible intentionally to torture the person for the sake of the same end. As torture uses the person, we should make other things equal by focusing on intentionally killing a person as a means to an end.
One question is whether we should accept your implicit claim that if it is permissible to do X to a person and doing Y is worse for the person than doing X, other things equal it is also permissible to do Y to the person. I think that this might well be false. For example, lengthy punishment may be no more effective than humiliating punishment, and may be worse for the person punished, but the former seems sometimes permissible when the latter is not, other things equal. So it is open to friends of Wasserman to claim that it is wrong to torture a person where it would be permissible to kill the person as a means to an end, even though killing the person would be worse for her.
More importantly, I would have thought that it is normally wrong to kill terror suspects as a means to save the lives of others where a) we are unsure whether they are terrorists; b) we are unsure that doing this would be effective, both in the individual case and in general when compared with other things that we could do to avert the threat of terrorist violence. You rightly claim that we do use people as a means in deterrent punishment. But, in countries with reasonably humane systems of punishment (not mine, and certainly not yours) a) we have gone through a trial with an appeals process, so we can be much more confident that the person is liable to be used than in the case of terrorist suspects; b) we don’t kill them or directly inflict pain on them; and c) we harm them whilst at the same time aiming at their rehabilitation. In hypothetical scenarios, punishment without these features might be permitted. In the real world I believe punishment without these features is also wrong. Perhaps what the US does in torturing is no worse than what it does in the criminal justice system. Not much comfort in that though!
Great stuff Victor. Thanks so much for engaging. Perhaps you’ve hit on the answer. But let me push back a little and see what you say.
First, indeed, I assume it is “more difficult to justify aiming at the pain of one person to benefit others than causing pain without aiming at it.” But that mixes together two issues: the relevance of aiming at pain and whether it is the pain victim’s or other people’s good that might justify inflicting it. As I said in my original post, I think deontology has to be fundamentally patient focused. I think the fact that one can aim at a patient’s pain when for his own good helps show that. The aim is the same, it’s the patient’s relationship to it that varies. So I put aside as probably misleading, when it comes to explaining basic wrongness, the issue of agent aim.
That leaves the Q of whether one can justify doing X to V for another’s good as for his own (even without his consent). The reason to think not is that it risks running afoul of the means principle or whatever close cousin of it is right. We escape that danger if and only if the person is liable to such treatment because of his acts. But that takes me to my second substantive argument. You come to that last, however, so I’ll move on following your order.
Your next question is really interesting: “whether we should accept your implicit claim that if it is permissible to do X to a person and doing Y is worse [you mean “less bad”] for the person than doing X, other things equal it is also permissible to do Y to the person.” You illustrate with the example of humiliating punishment versus long punishment, where the latter is worse, but the former more likely to be unjustifiable. I’m not sure what to say, except that if this is right, we need to understand better why.
One possibility, which undermines the argument from punishment types to torture and killing, is that humiliation actually is worse than long-term detention. That’s not altogether implausible. Humiliation can be destructive of one’s ability to lead a good life. One might argue that humiliation is a kind of social death, leaving one spiritually isolated and angry. Long-term detention can leave one physically isolated, but still spiritually connected, and that may matter more for a good life. It was a premise of my argument that torture is not, in that way, worse than death.
Suppose humiliation is not that damaging. Suppose its scars are typically less bad than those of long-term detention. Is there still a reason to rule the former out and allow the latter? I don’t deny that there could be. But I’d like the account. If it repeats the puzzle about torture, that’s interesting, but ultimately not enlightening. Further thoughts from you would be most welcome here.
Turning to your final paragraph, where you engage the argument from liability, you introduce two uncertainties: (a) whether V is a terrorist, and (b) whether torture will be effective. You suggest that punishment should be (though it is not in the UK and “certainly” not in the US) better because the trial process lets us be “much more confident” that they are liable to being used. (You say more, but I want to handle the rest separately.) But comparing an ideal punishment system with real torture choices seems unfair. If you are not going to be an abolitionist about punishment–indeed, the kind of thoroughgoing abolitionist who would let all or almost all convicted criminals go free–then you have to say that our flawed trial process, in which the innocent do sometimes get convicted, is justifiable. I’m going to work with that assumption–I’m finishing an article on proof beyond a reasonable doubt when not writing this reply; you can find my reasons there if you want to. But if we can punish despite the risk of doing it to the innocent, then I don’t see why we can’t do the same with torture. The only question is: how high must the standard of proof be? Certainly there are instances, and I’d guess some of the victims of CIA torture fit this bill, in which that standard of proof has been met. That takes care of worry (a).
Worry (b), whether torture is effective, is of course the underlying alternative explanation. I started the post by endorsing my friend’s worry that maybe the deep objection to torture is that it’s just not sufficiently effective to warrant using it. I’m hoping to find an alternative answer. My suggestion was that using it puts a country on a slippery slope to despotism. I’d like to learn of an even stronger answer. But if it isn’t to be found, then maybe we have to hang our opposition hats on effectiveness and slippery slope worries. And insofar as those contingent worries can be assuaged, then we might want to admit that it is sometimes justifiable.
Returning to your last paragraph’s argument, you also push me on the fact that we don’t kill or directly inflict pain on people in idealized punishment systems. I’m not sure that’s right. I’ll grant you your opposition to the death penalty, though not unconditionally. If the death penalty were much more effective than other punishments, then I suspect I would endorse it for the most serious crimes, at least assuming other problems of racism, classism, etc could be addressed. But as for the claim that punishment does not involve inflicting pain… really? OK, we don’t whip people like the Russians did at the time Dostoyevsky wrote “The House of the Dead.” But we certainly do intend to inflict suffering. I don’t see the categorical moral difference.
Maybe you’ll say that my attitude just shows that I wrongly embrace the barbarism of retributivism. Maybe you’ll say that we should not aim even at suffering. At most we can incapacitate. But I don’t think you can really take that line. You think punishment is substantially justified because of its deterrent effects. To deter, it has to be undesirable. Making it undesirable has to be an aim, and a perfectly sensible way of doing that is to make sure that it involves some suffering.
Your last point: we punish while also aiming at rehabilitation… at least in an ideal system. Interesting claim. Let me push back in the spirit of Michael Moore. For whom do we rehabilitate them? For us or for them? If for us, so that they are less likely to be recidivists, fine. But then that doesn’t seem like a deep point that would tell against punishing them without rehabilitation if we weren’t worried about what would happen afterwards. And if there’s no deep commitment to rehab, then I don’t see the contrast with torture. Alternatively, you could be thinking that we owe it to them. Again, there’s a bifurcation that should be considered. We could owe it to them because we think we have treated them unjustly–most criminals come from parts of society where they faced structural injustices. But if that’s the position, then why focus on giving rehab to them? Why not address those resources more generally to fixing the structural injustices. Or you might think we owe it to them paternalistically, as people who are clearly more troubled than the rest. But why paternalize? Why not simply offer help? And if we are in the business of offering help, why prioritize convicts over other poor people. In sum, I don’t really buy point C.
In sum, I’m not yet moved by anything you say in your third paragraph except the point that torture might be too ineffective given the costs. But that, as I said, is the hypothesis I’m looking to avoid falling back on.
Insofar as I’ve missed something, please set me straight! And especially, if you have more to say about the humiliation point, let’s see it.
Thanks Alec for your nice response, and sorry for the slips in the earlier post – fast writing costs accuracy!:
Two points. First, most people think that it can be wrong, even very seriously wrong, to act towards others in some ways that do not make them worse off at all, or even that make them better off. Humiliation seems a good example. It is normally wrong to humiliate a person even if doing so makes the person better off. For this reason, we should be sceptical of the general claim that the stringency of a duty not to affect a person in certain ways depends only on how badly off doing this makes the person. And it follows that we should also be sceptical of the general argument that as we could justify making a person very badly off, other things equal we are permitted to do things to the person that make them less badly off.
In the light of this, take the Sussman view that forcing self-betrayal is especially difficult to justify. It might be argued that the reason for this is not simply that self-betrayal makes a person much worse off. It might be argued that manipulating a person’s will to act against her values is distinctively wrongful independently if it making her worse off. And this seems plausible, given that it often seems clearly wrong to manipulate a person’s will to get her to do things that conflict with what she values, even if doing so makes her better off. Of course, some might claim that terrorists are liable to be treated in this way, or that they have no interest in not being forced to act in a way that conflicts with abhorrent values. So there is plenty of room for argument about whether this is going to justify a stringent prohibition on the torture of known terrorists. But if this argument against torture fails, it does not fail simply because torture does not make a person worse off than killing.
Secondly, on the rehabilitation point, my thought was simply that it is much easier to justify restricting a person’s liberty and inflicting harm on a wrongdoer if we also aim at their rehabilitation. Assuming that wrongdoers still have moral importance, if we can choose between two kinds of punishment that have equal deterrent effects (and even if they are unequal, I think), and one will help the person see the wrongness of their conduct, and help them to lead a better life, we ought to choose that one. That is, at least in part, for the sake of the wrongdoer. As people will tend to dislike having their liberty restricted, or being forced to recognize their wrongdoing, this will often be sufficient for powerful deterrent effects without us aiming at pain or suffering.
Perhaps you might argue that we should give the wrongdoer the choice between these options on the grounds that not to do so would be wrongfully paternalistic. I doubt that. The importance of a person making proper judgements about their wrongdoing might be sufficiently significant to justify restricting their choices. But even if you are right, the fact that we can provide the option of a rehabilitative form of harming makes a significant difference to the permissibility of inflicting the harm. This distinguishes properly practised punishment from torture.
Very helpful Victor. But again I want to push you a bit.
You write: “most people think that it can be wrong, even very seriously wrong, to act towards others in some ways that do not make them worse off at all, or even that make them better off.” You use this to account for the at least pro tanto wrongfulness of humiliation. But I wonder if this isn’t a case of a paternalistic restriction. People don’t like being humiliated. And there is a pro tanto reason case against forcing people to endure what they do not like. Citing their own good to overcome that is paternalistic. There better be a good reason to paternalize, like their deep inability to see what’s in their own good (insanity, immaturity), or there being an emergency (like about to swallow poison, unwittingly, and simply calling out information won’t save them).
If we move from paternalistic reasons to other-regarding reasons, we now have the means principle problem. But then liability justifications get a grip. And in that context, where issues of their welfare are not part of the justification, I wonder if my principle isn’t sound: if you may do X to V to achieve G, and X is worse for V than Y, then you may do Y to V instead.
Turning to Sussman’s point about self-betrayal, I think this point may undermine what you say about Sussman’s point. Yes, it may be impermissible to make a person betray her values for her own good (paternalism, assuming no paternalism exception applies), but if it’s for the good of others, and she’s liable to harm, and the harm of self-betrayal is less than the harm of gruesome death, then…
Re rehab, you say that its justification “is, at least in part, for the sake of the wrongdoer.” Is your thought that this helps with parsimony: we should be as harmless as possible, given the needs of deterrence and incapacitation? If so, I’m on board. But I’d want to structure that justification so that it does no more than give them options to get education, moral engagement, and the like. Otherwise it’s paternalistic, and I don’t see how that’s justified.
Now you take that on in your last paragraph, saying “The importance of a person making proper judgements about their wrongdoing might be sufficiently significant to justify restricting their choices.” I take it that you mean that this is important for us or for the victims, not for the convicted criminal. (Note, Tori McGeer has done some interested psychological work on how important it is To Us that the wrongdoer own up and apologize.) I agree, that it is important for us that criminals recognize the nature of their wrongdoing and repent. It allows us to bring them back into community with us. It seems a proper goal of punishment, given that we usually do not mean to cut them out of our community for the rest of their lives. But I’m not sure how that is supposed to make “a significant difference to the permissibility of inflicting the harm.” It’s not about their welfare under that justification. And I would not count the good to them as something you can sneak into the justification by the back door. Only what comes in the front door gets to count in a justification. (I know, there’s Kamm’s principle of secondary permissibility, but I think that’s different.)
It’s me that should be pushing you!
But still – I find this principle ‘if you may do X to V to achieve G, and X is worse for V than Y, then you may do Y to V instead’ hard to believe. Take harming V, innocent person, as a means to achieve a goal. This seems sometimes permissible, if the harm is modest and the goal is very important. It doesn’t follow that forcing V to betray her values would also be permissible for the sake of that goal, as long as V suffered no more harm than in the first case. Our attachment to the things we value, and hence our interest in not betraying our values, is not simply dependent on how well off or badly off we will be. As I said in the earlier post, this might not apply to those with abhorrent values though.
On rehab – I meant importance to the wrongdoer! It is important for wrongdoers to recognize that what they have done is wrong I think. But more generally, you seem to think something like this: if it is wrong to inflict harm on a wrongdoer for the sake of some goal where this would not result in rehabilitation, it would also be wrong to inflict the same degree of harm on the wrongdoer for the sake of that goal where this would result in rehabilitation. I find that hard to believe. Suppose that a rehabilitated offender will be much better off than a non-rehabilitated offender, as is plausible. Punishment, I think, is easier to justify if it also makes the lives of wrongdoers go better – the harm that we inflict is at least to some degree offset by the benefits of rehabilitation.
I welcome the pushing. But I’m not inclined to grant you your counterexample in your first paragraph. While I agree that an innocent may be used as a means of achieving an important goal if the harm is modest, I’m not sure that forcing her to betray her values wouldn’t count as a more than modest harm. I guess details matter. I can imagine a case where the forcing would be easy, because V is inclined to betray her values anyway. In that case, I’m not so sure that the forcing–say a mild threat–would be wrong if the goal were important. On the other hand, I can imagine a case where the forcing would be hard, because V is deeply committed to her values. In that case, the kind of torture required to turn her against them, and the rupture that would result, seem like more than modest harms. So I put the ball back in your court: give me an example that really suits your point.
It’s worth noting, by the way, that I think we both agree that what matters is objective harm. Hence your point about abhorrent values. That said, I think we also agree that being coerced and betraying one’s values are pro tanto harms. Yes?
Re rehab, I still object to the paternalism in justifying forced rehab with reference to the “importance to the wrongdoer.” Even if it makes the whole punitive experience less bad for V, I don’t think you get to do the subtraction yourself in the justifying process. It’s like this: Suppose, given limits based on proportionality and/or the good to be achieved, I get to inflict 5 units of harm on V to achieve some good. Suppose the punishment is made up of two components. One does her 10 units of harm, but another gives her 8 units of good. She, however, would not choose to receive that good. I don’t think you can justify the punishment on the grounds that it imposes only 2 units of net harm. That’s paternalistic. (Maybe I’m wrong, but that’s how it seems at first blush.)
On the harm that needs to be done to get a person to betray her values – this harm need not be very great, even if the values are deeply held. Threatening a person with something that she is powerfully phobic about might be enough, even if the values are quite deeply held. More generally, a person can be tortured by making her anticipate a great deal of harm without harming her a great deal.
I was hoping to identify something wrong about forcing a person to betray her values that is independent of harm. I have powerful reasons not to betray a friend, for example. This is not because betraying a friend makes me worse off. If I am forced to betray a friend through torture, I think that the gravity of the wrong that is done to me is not best explained by the fact that I am harmed, even though I may be harmed.
On rehab – even if you are right that goods offset harms only if they are accepted (or would be accepted under certain conditions), rehab would still play an important role in justifying punishment. Many of the goods of rehab are accepted, or would be under relevant circumstances – training for jobs, skills, literacy, etc etc. Furthermore, given that others plausibly have a right that offenders recognize their wrongs, the fact that this is also good for offenders surely counts to offset harms – this is a good that the offender is required to accept, even if he does not want to.
On the relevant harm connected with betraying values, you miss half of the picture. The harm isn’t just in the forcing, but also in the betraying of deeply held values. I think the case you want is one in which not much harm is done to get the person to betray deeply held values (threaten with stimulus that the person has an irrational phobia of), the person betrays deeply held values (which would normally be a significant harm), but the values are abhorrent (making their betrayal objectively a good thing).
Assuming that you accept this as a way of fleshing out the example, how does it cut? You were trying to offer a counterexample to my “if you may do X to V to achieve G, and X is worse for V than Y, then you may do Y to V instead.” For the example to work, it would have to be the case that we could do something really harmful to V to achieve G, but we couldn’t threaten her with a phobia to get her to sing, and thereby abandon her deeply held but abhorrent values. Gee, when I spell it out that way, I’m not sure it’s a counterexample!
On your betraying a friend example, why isn’t the right thing to say either that the harm is to the friend or the friendship, the latter harm being a harm to both parties.
I’m going to let the rehab point go.
But I want to add, to refocus the discussion on torture, a point that you made to me offline: the question is not whether torture is so morally problematic that it cannot ever be justified. The question is whether it is (substantially) harder to justify than causing equivalent (or greater) harms, given the good that may be expected to result. If you can make out your rejection of my “principle” then I think you’ll make the case. If…
Thanks Alec,
I don’t know if betraying deeply held values harms me when those values are not abhorrent. I guess the concept of harm is a bit unclear. But I tend to think that betraying something I value is not harmful in itself, though it may cause me harm. I have a powerful reason not to do this, but that reason doesn’t have much to do with my well-being, and I think of harm as a set back to well-being.
Thank you Victor.
I suppose I agree that betraying values (however deeply held) is not *itself* a harm. But I think it correlates strongly, and tends to cause a harm: a sense of self-betrayal that comes with dislike of self, discomfort with self, shame, and guilt. Those are all rather subjective measures of harm, but I suspect it’s also tied to self-destructive behavior: lashing out at others, undermining relationships, sabotaging one’s work, etc. Those correlations are looser. One could use the betrayal as an occasion to reform oneself. But I think that’s a rarer response.
I get the sense you agree. Which leads back to my question: have you given a counterexample?
I now think I have, because forcing a person to betray a value might gravely wrong her, it might not harm her, but even if it does, the harm is less grave than the wrong. So we should expect the stringency of a prohibition on forcing a person to betray values in a way that harms them to be greater than the stringency of a prohibition on harming as a means. That is enough to show that the principle ‘if you may do X to V to achieve G, and X is worse for V than Y, then you may do Y to V instead’ is false. X may be a harm, Y may be a slightly lesser harm, ,but one that results from, or causes, a betrayal of values. Y may then be less harmful than X, but the prohibition on Y may be more stringent than the prohibition on X. So even if inflicting X is permitted, inflictiing Y may be wrong.
THis may be only one counterexample to the principle. It stems from a much more general idea – that the strength of a perosn’s rights against being used as a means for an end depends on a broader range of claims that they might make than simply that they are worse off. Torture may have more than one feature that grounds such complaints. I don’t say this is true, and I don’t have an account that woudl make this work, I only say that it is open to those who think the prohibition on torture is especially stringent to find those reasons. For example, they might argue that respect for agency renders it wrong to turn a person’s agency against itself, even when this does not render the person worse off, or much worse off. I’m not sure whether an account like this is true, I only say that showing that its worse to be killed than to have one’s agency turned against oneself doesn’t show that it’s false.
Great Victor. As a deontological pluralist, I really like the options you articulate in your last entry. I can imagine someone saying that all of the considerations you list, like respect for agency, should be cashed out in terms of harms. But I suspect we agree: such a move would save the principle at the cost of making it trivial. The substantive thoughts need not appeal to an intuitive idea of harm; they appeal, rather, to other bases of claims that one may make.
Whether this really answers the Q: “What’s wrong with torture” I think we also agree to be agnostic. But it might, and it’s really helpful as a more general point.
THANKS.
My own recent thoughts on this revolve around the term, ‘learned helplessness.’ One way to grasp what is wrong with torture without reducing everything to the terms of benefits and harms is to consider that the root of human dignity is moral agency, the ability to make decisions for oneself. Torture attacks dignity at its root, seeking not merely to treat a person as a thing, leaving agency intact, but reducing that person to an objective kind of thinghood, utterly pliable.
As far as I’m concerned, the horror of such an act sets a very high bar for any putative argument from the social benefits of a program of torture.
I was wondering about this as well. The article seems to focus only on tortures effectiveness in getting information. But you also can’t torture people for punishment. If we consider 3 main purposes of legal punishment: Retribution, deterrence and rehabilitation, I think a strong argument for torture can be made for the first 2.
Now if we consider some definitions of torture:
Torture, according to the 1984 United Nations Convention Against Torture, is:
“…any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him, or a third person, information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity. It does not include pain or suffering arising only from, inherent in, or incidental to, lawful sanctions. –UN Convention Against Torture”
(I’m quoting this from wikipedia so deserve to be tortured if they/I got it wrong)
In the US they focus in on the physical pain threat of death or mind altering substances.
“As used in this chapter—
(1) “torture” means an act committed by a person acting under the color of law specifically intended to inflict severe physical or mental pain or suffering (other than pain or suffering incidental to lawful sanctions) upon another person within his custody or physical control;
(2) “severe mental pain or suffering” means the prolonged mental harm caused by or resulting from—
(A) the intentional infliction or threatened infliction of severe physical pain or suffering;
(B) the administration or application, or threatened administration or application, of mind-altering substances or other procedures calculated to disrupt profoundly the senses or the personality;
(C) the threat of imminent death; or
(D) the threat that another person will imminently be subjected to death, severe physical pain or suffering, or the administration or application of mind-altering substances or other procedures calculated to disrupt profoundly the senses or personality; and”
http://www.law.cornell.edu/uscode/text/18/2340
Now consider prolonged imprisonment. Clearly this causes mental suffering. The US code clearly rules out that this would be considered torture. But the UN Convention, not so much. I mean they add the modifier “It does not include pain or suffering arising only from, inherent in, or incidental to, lawful sanctions.” But isn’t this definition supposed to tell us what is and is not a “lawful sanction”??
Ok but here is where I think it gets interesting. If we are to say that
1) torture is wrong and
2) prolonged incarceration can be used *as a deterrent*,
Then what about the imprisonment is doing the deterring, if not the mental suffering it causes? It seems to me that if imprisonment caused no mental suffering it wouldn’t be a deterrent. It seems to me that some sort of suffering is required if any punishment is to have a retributive or deterring effect.
Yes, to deter, punishment must involve some sort of suffering. But torture is consistently defined by reference to something like SEVERE pain and/or suffering. Where and how to draw the line is obviously a difficult matter. And it is even more difficult if we take into account that what many convicts will endure without much trouble, some will find unbearable. Nevertheless, it seems to me that most punishment is designed not to cross the line. Through neglect of basic prison security, I suspect it often does cross the line. That is, prisons often allow prisoners to torture others. But that’s a failure to live up to its own standards–a serious failure, but not a fault inherent in punishment.
A nice insight from Paul Krugman, from his 12/21/14 column, to support my first hypothesis: “One suspects, by the way, that this false notion of power [that conquest pays] was why the architects of war [especially the invasion of Iraq] made torture routine — it wasn’t so much about results as about demonstrating a willingness to do whatever it takes.”