« A Personal Touch | Main | The role of lawyers in making the law »

Thursday, October 15, 2009

When Is Interrogation Torture?

This post is adapted from my new draft article, The Experiential Future of the Law.  Comments on the article are welcome at akolber “at” sandiego.edu:

Waterboarded interrogatees are strapped down on an incline, with their feet elevated above their heads, while water is poured over their faces to give the sensation of drowning.  A federal statute prohibits those acting under color of law from engaging in acts “intended to inflict severe physical or mental pain or suffering” on people in their custody or control.  Does waterboarding constitute the sort of intentional infliction of extreme pain or suffering that can qualify as torture?

We often address the question by focusing on the interrogation technique in the abstract.  Generally speaking, waterboarding probably does inflict extreme pain and suffering.  In fact, however, whether a person experiences extreme pain or suffering (and whether we can expect a person to) depends on the relationship between the person and the way he is treated.  A person who developed PTSD from having nearly drowned in a swimming pool as a child will react differently to waterboarding than will a Marine who has advanced training in how to resist waterboarding.  Defacing a religious text may constitute extreme suffering to one person but be largely inconsequential to the next. 

Despite variation in interrogatee sensitivity, we seem inclined to rely on broad categories of prohibited interrogation behavior.  Another way of safeguarding interrogatees, if we had sufficiently reliable technology, would be to limit the maximum distress an interrogatee is permitted to experience, based on certain biological markers.  For example, we can already measure several markers of acute distress, like pulse, blood pressure, levels of stress hormones, galvanic skin response, and others.  We’re even getting better at predicting which sorts of interrogation techniques are likely to lead to long-term negative effects.  (There may be an unacceptable risk of electrocution that makes it impossible to measure distress during waterboarding, but there are lots of other interrogation techniques; so the general point remains.)

The idea of measuring people’s acute distress seems disturbing, but the concern is misplaced.  The harm caused by inflicting some amount of distress is far more troubling than the comparatively slight invasion caused by measuring it.  In fact, if you’re permitted to inflict modest amounts of pain or suffering, then perhaps you are morally obligated to measure the distress inflicted to make sure it does not reach a more extreme level. 

The reason that the measurement of distress seems so troubling is that it forces us to palpably recognize and quantify the nasty business of intentionally making people suffer.  The measuring is not the principle act of harm, however.  Measuring merely forces us to confront and better quantify the real concern—the pain and suffering itself.  If it's impermissible to measure the pain of technique X, you probably shouldn't be doing X in the first place.

There are several reasons why we might rely on “categorical” methods of banning torture rather than engaging in individualized measurements.  I'll name just two.  First, we may not be good enough at measuring acute distress.  (True, though someday we probably will be.)  Second, we may be skeptical of those tasked with measuring distress levels.  We fear that they will simply rubber stamp torture.  (True again, though it is far from clear that we couldn’t adopt procedures to reduce this concern.  If people can be dishonest about measuring distress, they can be dishonest about what categories of interrogation are used.)  In any event, we should not lose sight of the fact that even if we ban certain methods of interrogation, those methods are just rough proxies for our true concern, namely the intentional infliction of extreme pain and suffering. 

Posted by Adam Kolber on October 15, 2009 at 06:52 AM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference When Is Interrogation Torture?:



Inspired by and relying on, in part, two articles of Jeremy Waldron,* let me proffer yet more reasons why we might rely on or prefer a categorical prohibition of torture that at the same time helps account for our concern with "the intentional infliction of extreme pain and suffering."

First of all, I think any discussion of this topic must also reference the related prohibition of cruel, degrading and inhuman treatment. This prohibition reflects the principle of "the respect of persons qua persons" (and the corresponding recognition of the principle of human dignity), and the rejection of all forms of brutality. In the context of interrogation, it likewise entails an understanding of the meaning of (respectful) pressure and the coercion imposed as a prospective penalty for legal non-compliance.

Second, there is something about the aforementioned prohbitions that ineluctably involve the peculiar form of VULNERABILIY of those being interrogated and the related power differential of the particular circumstances or situation. The "special circumstances of vulnerability," as Larry May has argued, give rise here to a duty of "humaneness," for the principle of "humane treatment" is plausibly if not persuasively viewed as the cornerstone of international humanitarian law, above and beyond the traditionally recognized principles of discrimination (or 'distinction'), necessity, and proportionality.

Third, the categorical prohibition might be seen as giving due recognition to the more-than-quaint distinction between mala prohibita and mala in se, in which case the latter trumps the former: "Some things are just wrong, and would be wrong whether positive law prohibited them or not" (Waldron). Relatedly, and to fill this out a bit, a categorical or absolute prohibition involves a more or less implicit reference to some normative background, some “moral minimalism,” or basic principles of natural law (which need not be religious), or the jus cogens norms of international law, or the aforementioned notion of humane treatment in international humanitarian law (which includes principles of compassion and mercy: see Larry May’s War Crimes and Just War, 2007), or fundamental human rights (as should be obvious, these principles and norms are not unrelated to each other), all of which represent our endeavor, in Waldron’s words, to “insulate the prohibition of torture against the contingency of positive law” (e.g., revision, redefinition or repeal).

Such a normative background might even include a (Fuller-like) recognition of the (moral) spirit of law, and thus, as Waldron also argues, the absolute prohibition of torture may play an emblematic role (as what he terms a ‘legal archetype’—think, for example, of the Habeas Corpus statutes as ‘archetypal of the whole orientation of our legal tradition towards liberty’) in its fidelity to the “genius and spirit” of law. This is rather diffrent from a simple deduction from natural law principles, as it can be understood as simply giving shape over time to a collective growth in moral understanding and awareness:

“The spirit of a cluster of laws is not something given; it is something we create, albeit sometimes implicitly, it emerges from the way in which, over time, we treat the laws we have concocted. We begin to see that together the provisions and precedents in question embody a certain principle; our seeing them in that way becomes a shared and settled background feature of the legal landscape; and we begin to construct legal arguments that turn on their embodying that principle.” (Waldron)

In this case, the archetype is animated by a principle of “non-brutality” (cf. the constitutional provision on cruel and unusual punishment and the constitutional requirements for procedural due process). Undermine a legal archetype and you increase the likelihood or probability of calling into question, or corroding the integrity of, a significant domain or body of law (in this case, where law prohibits ‘lesser evils,’ like police brutality, coerced confessions, etc.).

And specifically with regard to the question of intentionality, it's important to keep in mind the fact that "the potential defendant we have to consider is one who already knows he is inflicting considerable pain; that is his intention. It seems to me that the working definition of torture in [the anti-torture] statute already gives him all the warning he needs that there is a huge risk in relying upon some casuistry about ‘severity’ as a defense against allegations of torture" (Waldron). The intent here applies to the attempt to “break the will” of a detainee or prisoner.

In conclusion, I think we should be suspicious, in at least a prima facie way, with any attempt to give the prohibition of torture a definitional precision on par with at least some kinds of legal rule, as it appears to invariably raise the question of the suspect or troubling motivation that would animate any such attempt. Such suspicion is warranted, in part, if only because “the use of torture is not an area in which human motives are trustworthy,” a conclusion we might draw both from history and moral psychology.

And I look forward to reading your paper.

*Please see Jeremy Waldron’s “Torture and Positive Law: Jurisprudence for the White House,” Columbia Law Review, Vol. 105, No. 6 (2005): 1681-1750, and his “Cruel, Inhuman, and Degrading Treatment: The Words Themselves,” New York University School of Law, Public Law Research Paper No. 08-36 (November 2008): 1-47. Available: http://ssrn.com/abstract=1278604

If I may, I would like to exploit this opportunity to let students fairly new to this topic that I have a bibliography on the "moral, legal and political dimensions of torture" available here: http://ratiojuris.blogspot.com/2009/04/torture-moral-legal-political.html

Posted by: Patrick S. O'Donnell | Oct 15, 2009 10:05:17 AM

Without getting into the merits of whether you should or shouldn't forcibly interrogate (I happen to think you shouldn't), I think you've done an excellent job of explaining the legal and potentially moral obligations of those who do so or think others should do so.

Advocates for forcible interrogation that may or may not be torture have to be able to explain how they know that their chosen techniques are not torture under the statute and how they plan to prevent torturing from inadvertently occurring or happening through an excess of zeal. Without those two points, their argument boils down to "we like people screaming and we don't care whether it works or not."

Posted by: Matthew Reid Krell | Oct 15, 2009 11:45:01 AM

Let me echo Patrick's comments - not only do we ban torture (or tortures, individually defined) categorically becuase of its effect on the "recipient," we ban (or we should understand it to be banned) as a general statement on the relation between the government and individuals.

Our society would look and feel very different if individuals knew that any and all "interrogation" techniques could be applied, so long as they did not reach a particular distress level. And it would feel markedly different if the "non-sensitive" citizens among us knew (or at least suspected) that they could get the rack and screw a lot longer than their foppish neighbor.

Now, I put "interrogation" in quotation marks because physically brutual methods (of essentially any stripe) are never really for the purpose of interrogation. To paraphrase others, the point of brutality is to brutalize, not to glean information. The notion that interrogation for the purpose of eliciting usable information for the prevention of illegal behavior can be effectuated by physically brutal methods has, I thought, long been discredited.

Posted by: John | Oct 15, 2009 2:08:12 PM

what do we do with the thin-skinned egg-shell (etc) interogee, who treated in a dignified manner nonetheless loses his composure and is physically bearing the signs of great distress despite nothing actually being done to him physically besides being put in a room and facing some questions? Does that benign questioning become torture also, simply based on the experiential account of the law, assuming it's all measured? Or does your account serve simply as a limit?

(Sorry if you address this in the paper; i've downloaded it but haven't read it yet...)

Posted by: Dan Markel | Oct 15, 2009 2:28:18 PM

Thanks for all of your thoughtful comments! (And thanks especially to Matthew for your kind comments. You're right that I think monitoring distress levels, if approached properly, can help protect interrogatees. If a police station had an independent monitor of all of its interrogations, we might applaud its due care in looking out for the well-being of interrogatees.)

My point in the post is not to resolve all of the many interesting and important issues related to enhanced interrogation that are out there or to offer any general theory about the necessary or sufficient conditions for using some interrogation technique.

My point was just that the torture statute prohibits the intentional infliction of extreme pain or suffering. So, there are certain things we are prohibited from doing to interrogatees because we know (or, presumably, are willfully ignorant that) such actions will cause extreme pain or suffering. In some cases, especially if the technology were cheap and available, failing to monitor an interrogatee's distress seems like a kind of willful ignorance. I absolutely agree, though, that there are other reasons, aside from the couple reasons that I list, why we don't rely solely on such technologies to protect interrogatees.

To the extent that some of the comments above defended more objective conceptions of pain and suffering, I think we will be hard pressed to explain something like "brutalization" without reference to the interrogatee's experiential states (including, importantly, the belief and intentional states of the person who inflicts the negative experiential states).

Dan asks a very interesting question about extremely sensitive interrogatees. In terms of addressing your question in the context of the statute, we also have to describe the actors' beliefs/intentions with respect to the hypersensitivity. So, suppose we have an interrogatee who has severe phobic reactions (with panic attacks, subsequent PTSD symptoms, etc.) to water in styrofoam cups. And suppose we know this about the interrogatee. Must we remove such cups from the interrogation room or else risk violating the torture statute?

As a matter of positive law, I don't know the answer and would be curious to hear from someone who does (I'm sure there are cases of religious preferences that raise similar concerns, though we may treat religious preferences in atypical ways.) I think a consequentialist has a pretty clear answer. It depends on the value of proceeding with the interrogation with the cup in the room relative to proceeding without it. I'm not sure how non-consequentialists would resolve the question, but however they would do so, I think they would be mistaken if they did not at least take into consideration the interrogatee's hypersensitivity.

Thanks again!

Posted by: Adam Kolber | Oct 15, 2009 7:01:56 PM

Adam - you say:

To the extent that some of the comments above defended more objective conceptions of pain and suffering, I think we will be hard pressed to explain something like "brutalization" without reference to the interrogatee's experiential states (including, importantly, the belief and intentional states of the person who inflicts the negative experiential states).

Can you explain why categorical limitations on "interrogation" tactics based on brutality of method are necessarily "without reference to the interrogatee's experiential astate"? From my view, there is reference to the interrogatee's experiencial state in making the categorical rule - it's just not calibrated to each and every individual case. What you claim is that the categorical reference is not, essentially, granular enough; that doesn't mean the categorical rules are adopted without any reference to the interrogattee's state.

Posted by: John | Oct 16, 2009 9:22:36 AM

Hi John,

I absolutely agree that we develop the categorical rules based on generalizations about interrogatee reactions. I also agree that the issue is about granularity. If what motivates the rule is a concern to prevent extreme pain and suffering, we might come up with a categorical rule because it's the best we can do. But it will likely be both overinclusive and underinclusive (like all rules are). In a world with technologies that allow more fine-grained determinations of pain and suffering, it seems to me that we can do a better job of preventing the harm we seek to prevent by phrasing the "categorical" rule, if you will, in terms of the levels of distress of particular individuals.

Posted by: Adam Kolber | Oct 16, 2009 4:15:03 PM

Adam -

That makes sense - just wanted to make sure you weren't throwing out the baby with the bathwater. Much thought over many years has been put into the project of making categorical rules in reference to perceieved levels of distress, which you seem to agree with.

Posted by: John | Oct 16, 2009 4:30:18 PM

Post a comment