Monday, December 10, 2007

Torture Hearing

Yours truly gave testimony on torture this morning to the Helsinki Commission, chaired by Sen. Ben Cardin (D-MD) and Rep. Alcee Hastings (D-FL). Here's the transcript of my statement passed out to everyone at the hearing.
Thank you, Mr. Chairman, for the opportunity to speak to the Helsinki Commission....

Historically, physical and psychological torture has been used to suppress dissent, force renunciation of beliefs, extract confessions, punish, force denunciation of others, intimidate a population, humiliate, and gather information. All torturers claim a state of necessity. Recently, some have advanced the claim of significant information once again to justify torture. They argue that the information gained from torture is of greater moral significance than the torture of human beings. Torture, they say, is a necessary evil in the battle against a greater evil. The entire claim is based on the premise that there exists information of great moral significance, that it is discoverable only through torture, and that this legitimizes the use of torture.

As with most ethical issues, how the problem is articulated is of crucial importance. Today, torture is commonly justified by appeal to a state of necessity emblematic in the proverbial ticking time-bomb hypothesis. This frames the issue wrongly from the outset, however, and grounds it in a state of fear. The ticking time bomb example, so corrosive of our moral imagination in the public discourse, provides a crude utilitarian justification for the use of torture: torturing one bad man versus saving many innocent people. This may serve to trump the basic claims of the absolute prohibitionist. But why stop with the one bad man, on this view? If the potential information is of great moral significance, why not torture the one man’s children or everyone in his village? To assume this normative framework appears to allow for the most extensive abuses committed in the name of uncovering the morally significant information presumed a priori.

A better understanding of what is entailed in seeking morally significant information through torture thoroughly belies the information-gathering justification on both efficacy grounds and moral grounds.

Torture “works” in that torture victims speak. The information gained is notoriously unreliable, however, as noted since the time of Aristotle. Accounts of torture from the Inquisitions exhibit how the most delirious tales were elicited from the victims. This information served to confirm the prior beliefs of the torturers. Bad weather, for instance, was thought at the time to be caused by airborne demons in consort with human “witches.” In the delirium of torture, torture victims – those accused of being witches – confirmed these beliefs while providing the names of other “witches” who would reconfirm both the preposterous prior beliefs and the inquisitors’ authority. The information was, of course, not true. Yet, it was meaningful information in that it fit extant prior beliefs in a historical context framed as a medieval version of the state of necessity.

If information, today, must be of great moral significance to justify torture, how would we know it was of such moral significance? First, torturing for information requires the institutionalization of torture. Many commentators have noted this. There must be trained torturers and thus also trainers, a legal and administrative apparatus, a cadre of doctors and lawyers and data analysts, and so on. Non-torturous intelligence-gathering and interrogation activities already require similar institutionalization. The conspicuous difference is that the latter does not demand by its nature that each human link in the apparatus suspend its moral decency. Moreover, many intelligence professionals and interrogators state that there are much better methods of gaining actionable intelligence than through torture, even when conducted under time constraints.

Second, since raw information from an individual torture victim is unreliable, information that rises to the level of morally significant information is highly unlikely ever to be gained from an individual victim alone. Torture must be used broadly. On occasion, the torturers might have prior expectations that the prisoner indeed possesses important information. The justification of morally significant information demands prior knowledge that the torture victim possesses this information. It also demands that the information be actionable such that a serious, imminent threat is actually prevented. It is exceedingly difficult, however, if not impossible, to judge the information gained from torture to be morally significant until that greater evil is indeed prevented. This combined knowledge prior to the act of torture might very well obviate any perceived need to torture. More likely than the time bomb case of torturing one bad person is the case of torturing many innocent people in search of what might hypothetically justify the act of torture.

How does one know when one has true information? When one seeks to justify torture by gaining important information one presupposes that such information exists, and will be discovered only through a morally heinous practice. The information must thus be previously unknown in order to justify using torture. Yet, its moral significance must also be previously known in order to justify the act. It is not meaningful information until one has tortured, gained information, and then verified it. This is where information may become meaningful. Meaningful information may then fit with prior beliefs, assumptions, and modes of interpretation (and in the present case, recall, the context is a state of necessity). But it is not necessarily true information, as illustrated briefly in the example from the Inquisition. Furthermore, the victim’s guilt need never be resolved.

The logic of acquiring true information as opposed to merely meaningful information suggests a more extensive practice. Drew Sullivan, an investigative journalist currently based in Bosnia, recently recounted to me his time spent on the Thai border with Burmese journalists and refugees. Each of the journalists had been tortured by the Myanmar government. In discussions with Mr. Sullivan and others about their torture, the victims explained that during their ordeals they were often confronted by the torturers with information – true and false – derived from the previous tortures of other victims, often relatives or friends who had been tortured many months earlier. It became clear that the military regime of Myanmar maintains a database comprised of information gained through torture. Of course, information from individual torture victims must be correlated with information from other victims and verified or falsified in order to be serviceable. The Myanmar government tortured many people in order to evaluate various individual bits of information and compare them with other bits of information in order to build a coherent account of actual information. All data from the individual torture victims – whether “good” or “bad” information – are logged into the database. The database then serves to uncover patterns in the mass of information and misinformation.

A principal incentive raised by the argument for torture as a means of gathering information is precisely what the Myanmar example suggests. It is ultimately to seek patterns of information rather than attempt to verify or falsify individual bits of data, especially under time and resource constraints. Comprehensive sets of data-points yield more complex patterns. The more extensive the practice and institution, the more successful torture will be. If torture is used indiscriminately and broadly, more complex patterns and a better understanding of what is meaningful in the information will be obtained. Patterns of information by themselves are meaningless, but they serve to corroborate and verify partial bits of information and infer other patterns. They also serve to eliminate or falsify outlying bits of information, the information gained from those innocent of any perceived wrongdoing. A descriptive narrative may be interpreted and assembled from the resulting patterns and regularities.

This is now a far cry, however, from an argument based on the moral tradeoff between torturing the one in order to save the many. At no point has meaningful information risen to the level of morally significant information that justifies torture. In the numbers game of the information-gathering justification (symbolized by the time-bomb), as the number of torture victims grows, the moral justification diminishes, although this element is not included by proponents of the argument. The use of torture as an instrument for gaining morally significant information thus contains its own absurdity. We end up with a swelling institution in search of its moral justification, causing increasing damage to innocents and ourselves, all in search of the supreme moral justification – the time bomb – only to find that, in the end, it is we who have become the moral equivalent of the time bomb.

I have limited this statement to discussion of the justification of torture as an information-gathering instrument because its current proponents state that the information is of greater moral significance than the torture of human beings. Every ethical and religious tradition, however, views torture as abhorrent. The other purposes of torture listed at the beginning of this statement are plainly beyond the bounds of all morality, although the slippery slope of torture often leads to such purposes, as exhibited in the photographs from Abu Ghraib. Since the currently proposed moral justification for torture as information-gathering is itself morally unjustifiable, we are better off treating the prohibition of torture as morally absolute.

The laws of a liberal democracy must clearly and firmly reflect these moral considerations, even if a scenario as portrayed in the time-bomb example should ever arise in actuality. In such a highly implausible case, in which all the conditions of prior knowledge of the victim’s guilt are equally in place, those who choose to torture must nonetheless face the consequences of severe legal sanction. Later judges of the torturers may decide to consider mitigation in their case, but mitigation cannot be determined in advance by a presumed state of necessity. Since liberal democracy – indeed the entire liberal political tradition – is grounded on universal principles of individual autonomy and dignity, to institutionalize their violation is to attack the very foundations of liberal democracy.


Anonymous said...

Torture is worse than pointless. They should be summarily executed if discovered under arms.

troutsky said...

That we are even having the debate again speaks volumes. Executed? They?

darkwaterhermit said...

Thank you for your excellent post. I have been shocked at the trend of utilitarian justification for not only war and violence, but even for the use of torture. I teach a college ethics class, and have seen my students steadily succumb to this line of thinking. Your blog will become a required read for my students as I add torture to my list of moral dilemmas. Who would of thought that torture would now have to be amongst the other topics discussed? I hope your presentation to our government officials caused them to think of the consequences of state sponsored torture.

MT said...

The Myanmar database scenario may be real but like the ticking time bomb it is extreme and misleading. You talk about a database in which all the information is obtained by torture and suggest that it is within this we must find a pattern establishing moral significance, which makes the prospect of finding this ex post facto moral justification look extremely dubious. But information by which to assess the veracity and to discover moral significance of torture testimony need not itself be such testimony. It could be printed on the front page of yesterday's paper. It could be the voluntary testimony of a vendor who sold Osama a hot dog on a particular day. It could be from undercover informants or surveillance. It could be the testimony of a relatively uncommitted and/or tangentially involved conspirator. It may be from an interrogation in which the subject was simply tricked or duped--approaches that a prudent torturing institution is bound to try always before torture, given the unreliability of the information torture will be known to have elicited in the past. I think you are supposing an internecine conflict like a civil war or with deep cold war style co-infiltration of the adversarial societies and in a context of such widespread intimidation nobody would dare incriminate anybody else except under torture. It's such conditions that have provided us perhaps all our examples we know of institutionalized torture, I suppose, and it even may be that torture fosters such conditions. But it's not obvious to me that they are any more germane to the moral question at hand than the ticking time bomb.

Not to mention, I bet you lost all the Republicans when you invoked at the end the "liberal political tradition."

Good that you said all that though!