Norman Geras recommends as a "must-read" a review by Stephen de Wijze of Karen Greenberg's The Torture Debate in America. I'm afraid I have to join Shuggy in being less than impressed with the piece. I can provisionally agree with De Wijze's conclusions, but I do take serious issue with a couple of the arguments he makes getting there, and I agree with Shuggy's later assessment that De Wijze is too soft on the arguments of the proponents of condoning torture. In particular, I disagreed with certain points he made concerning the "Ticking Bomb Scenario" (aka "TBS"), first mooted by Niklas Luhmann, a German sociologist, but made famous in America, in the wake of the 9/11 attacks, by Alan Dershowitz. De Wijze describes the TBS as follows:
[T]he core argument adopts the following formula. 1. The security forces have captured a terrorist who knows of a bomb which will detonate in the very near future 2. If the terrorist can be forced to reveal the whereabouts of the bomb the detonation can be prevented 3. The only way to achieve (2) is to use torture and thereby save many innocent civilian lives.
In a footnote, De Wijze links to the entry on torture in the Stanford Encyclopedia of Philosophy, which gives this example of the ticking bomb scenario:
A terrorist group has planted a small nuclear device with a timing mechanism in London and it is about to go off. If it does it will kill thousands and make a large part of the city uninhabitable for decades. One of the terrorists has been captured by the police, and if he can be made to disclose the location of the device then the police can probably disarm it and thereby save the lives of thousands. The police know the terrorist in question. They know he has orchestrated terrorist attacks, albeit non-nuclear ones, in the past. Moreover, on the basis of intercepted mobile phone calls and e-mails the police know that this attack is under way in some location in London and that he is the leader of the group. Unfortunately, the terrorist is refusing to talk and time is slipping away. However, the police know that there is a reasonable chance that he will talk, if tortured. Moreover, all their other sources of information have dried up. Furthermore, there is no other way to avoid catastrophe; evacuation of the city, for example, cannot be undertaken in the limited time available. Torture is not normally used by the police, and indeed it is unlawful to use it.
De Wijze continues:
Consequently, the TBS reveals a situation where the absolute prohibition against torture is shown to be wrong and opens the way to developing the conditions under which torture can be legally and morally justified in a liberal democratic society.
Many opponents of any legalization of torture, myself included, acknowledge that the Ticking Bomb Scenario provides a compelling argument for the use of torture in the circumstances described, but respond that the specific circumstances described do not, and will not, occur in the real world, that the scenario is therefore ultimately hypothetical, and not applicable to any practical discussion of torture. De Wijze is ready for us:
Firstly, there is no question that real life examples of TBS do occur. Perhaps, and thankfully, not as frequently as advocates of justified torture tend to suggest, but that they do arise is undeniable. Consider the following case reported in the Israeli newspaper Ha'aretz.
A suicide bombing was narrowly averted in Haifa (Israel) yesterday morning when the would-be perpetrator was arrested shortly before carrying it out. The drama ... began when police, acting on specific intelligence information about a planned attack in Haifa, arrested a number of Palestinians who had been staying in the city illegally. One 18-year-old from Samaria was arrested ... and when questioned, he admitted that he had hidden explosives for use in a suicide attack [and] directed [police] to an abandoned building where they found a belt containing several bombs that the Palestinian had planned to strap to his body and set off.... 
It is not clear from the newspaper report what techniques or form of questioning was used to elicit the information that prevented a suicide attack, but given the time scale and possible disastrous consequences of not obtaining this information, here we have a genuine actual example of a TBS.
Footnote 24 reads as follows:
This claim will be rejected by those who insist that any genuine case of TBS requires that we must be certain that (a) the right person is tortured (i.e. the terrorist who knows of the bomb's location), and that (b) this torture will provide the needed vital information to prevent the bomb's detonation. However, Luban and others insist that these certainties can never exist in the real world of counter-terrorism, hence there can never be a realistic case of the TBS. These conditions are too strong. The Israeli example does cross the required threshold for a genuine TBS since what is needed in all such cases is a reasonable expectation rather than cast-iron certainties. (This principle rightly is adopted within the criminal justice system – where the guilt of those accused must be proved beyond reasonable doubt rather than with absolute certainty. If the latter was demanded then no one -or very few indeed - could be justly convicted of committing a crime.)
This will simply not do. Given the global taboo on torture, to the extent that even governments which practice it routinely deny doing so vehemently, the claim that torture might be justified in certain circumstances is an extraordinary one, and extraordinary claims require extraordinary evidence. Indeed, De Wijze's framing of the Ticking Bomb Scenario leaves little room for doubt: he writes "the security forces have captured a terrorist who knows of a bomb." The detainee is—not "might be," is—a terrorist who knows—not "might know," knows—the location of the bomb. The example from the Stanford Encyclopedia of Philosophy uses similar language. Thus, De Wijze has established a high standard of certainty on the part of the security forces (presumably to make the Ticking Bomb Scenario all the more compelling), but then complains that those same conditions are "too strong." This is known as "shifting the goalposts," and it's dishonest.
In fairness to the Ticking Bomb Scenario, it is not always presented with as high a degree of certainty as De Wijze does. A leader in The Economist in January 2003 put it as follows:
Suppose you know that there is a bomb about to go off which could claim thousands of victims. You have good reason to believe that a prisoner knows where it is, and that torture may force him to tell. Would you allow him to be tortured? Most people, however reluctantly, answer “yes”.
This is a significantly more qualified wording, but nevertheless still quite compelling. But does the example of the planned Haifa attack measure up even to this watered-down version? Well, no. Let's look at that Haaretz article again:
[P]olice, acting on specific intelligence information about a planned attack in Haifa, arrested a number of Palestinians who had been staying in the city illegally. One 18-year-old from Samaria was arrested ... and when questioned, he admitted that he had hidden explosives for use in a suicide attack [...]
In this situation, the police did not limit itself to picking up suspects whom they had reason to believe might be involved in the planned bombing and interrogating them for the purpose of discovering the location of the bomb. Rather, they rounded up a bunch of potential suspects and initially subjected them to interrogation to determine whether any of them even had anything to do with the planned attack. Haaretz reports that one of those arrested turned out to be involved in the planned attack (in fact, he was the prospective bomber), but since no mention is made of accomplices, we can reasonably assume that the remainder of the Palestinians hauled in had nothing to do with it. Since the Israeli security forces did not know initially which (if any) of the arrestees was involved, it's more that likely that all of them were subjected to the same treatment. This treatment may or may not have involved "physical coercion" (the Israelis prefer not to use the word "torture") but either way, it does not support the contention that this situation met the conditions set out in the Ticking Bomb Scenario. For, given the wording of the Haaretz article, there are two possible scenarios: either the bomber was made to confess without resorting to torture, which negates the element of the Ticking Bomb Scenario that torture is the only way by which the information might be extracted. Alternatively, the Israeli security forces subjected all the arrestees to "physical coercion" (not implausible, given the urgency of the situation and the Shin Beit's reputation of using torture as a matter of routine on Palestinian detainees), in which case the information came at the price of the human rights of not only the bomber, but also of the remaining arrestees, all of whom were innocent of involvement in the bombing.
It also deserves mention in this context that the death toll of any single Palestinian terrorist bombing in Israel is often in the single digits, and has to my knowledge never exceeded 30 (the Park Hotel in Netanya on 27-Mar-2002). I acknowledge that this sounds callous, but part of the emotional appeal of the Ticking Bomb Scenario is that it balances the human rights of one person (reasonably believed to be a terrorist) against the lives of thousands of civilians. "Thousands" is the number used in the Stanford example as well as The Economist's, and De Wijze himself even adds another digit, stating: "It requires a rather fanatical devotion to moral principles to insist that no violation is permitted even at the cost of tens of thousands of lives" (emphasis mine). Even conceding that point, it requires significantly less fanaticism to insist that it is impermissible to torture one person—even one reasonably believed to be a terrorist—in order to preserve maybe two dozen lives. It is even less fanatical (I would say not at all) to insist that the safeguarding of one or two dozen lives does not justify the torture of a comparable number of detainees, all but one or two whom are almost certainly innocent of involvement in the bomb plot.
As I said, downplaying the number of casualties sounds callous; that is precisely the effect the "numbers game" is supposed to have. To illustrate, a speaker who wishes to establish that the Allied bombing of Dresden was a wicked act may invoke the estimated death toll of 200,000 or more. His interlocutor may point out that that figure has long been discredited, and that reputable sources place the number of dead between 25,000 and 35,000; significantly fewer than the first speaker's figure. In response, the first speaker will wave away the discrepancy, stating that the exact figures don't matter, it's still a lot, and how can the other person be so callous as to reduce human lives to mere numbers? This is plainly dishonest: the original speaker invoked the higher number for a reason, namely to elicit a more vehement emotional response from his audience; a response he evidently did not feel he would achieve had he cited the lower casualty estimate, otherwise why did he not cite that figure in the first place? Frankly, I find this tactic despicable. So it is with the Ticking Bomb Scenario as presented by De Wijze; having initially used hypothetical fatality projections ranging in the multiple thousands to make his case, it is an act of rank dishonesty to tacitly reduce that figure to less than a percent of what it was and then pretend we're still talking about the same (type of) scenario.
If anything, the Haifa example perfectly illustrates the dangers of "mission creep" of which critics of the Ticking Bomb Scenario warn, namely that once torture is deemed permissible, albeit initially only in exceptional circumstances and subject to stringent restrictions, it does not take long for those circumstances and restrictions to erode. Again, De Wijze himself goes from torturing one man who is reasonably believed to be involved in the bomb plot for the sake of saving tens of thousands of lives, to torturing an unspecified number of detainees, the majority of whom (perhaps all) are almost certainly innocent of involvement, for the sake of saving two or three dozen lives at most; all this in the space of half a page. From my perspective, De Wijze's piece alone forms a perfect example of why the Ticking Bomb Scenario is both intellectually and morally bankrupt.