Spring, 2005
8 Green Bag 2d 311
REVIEW: Ticking Bombs & Catastrophes: SANFORD LEVINSON, ED.: TORTURE: A COLLECTION: OXFORD 2004
Rosa Brooks
THERE ARE INFINITE VARIATIONS of the hypothetical, but most go something like this: a notorious terrorist has hidden a nuclear bomb in New York, scheduled to detonate in a few short hours. You -- a courageous federal agent -- capture the terrorist, but he sneeringly refuses to reveal the whereabouts of the bomb. If the bomb detonates, millions of innocents will die. Is it morally justifiable for you to torture the terrorist to force him to reveal the bomb's whereabouts?
Sound familiar? It should, if you've been following the post-9/11 debate about torture and other "harsh" or "coercive" interrogation practices, in which the "ticking bomb" hypothetical recurs with numbing frequency -- usually offered up to suggest that sometimes the ends really do justify the exceptionally unpleasant means.
The ticking-bomb hypothetical also haunts most of the essays in Sanford Levinson's new edited volume, Torture: A Collection. With contributions from a dozen prominent legal scholars and several major philosophers, Levinson's rich volume takes on issues ranging from the philosophical questions surrounding torture (Henry Shue, Michael Walzer, Jean Bethke Elshtain) to the history of torture in Europe and America (John Langbein, Jerome Skolnick) to the legal debate about torture (Oren Gross, Oona Hathaway, Alan Dershowitz). Other essays are by Chilean poet and novelist Ariel Dorfman, Harvard literary critic Elaine Scarry, and legal scholars Miriam Gur-Arye, Fionnuala Ni Aolain, Mark Osiel, John Parry, Richard Posner, and Richard Weisberg.
The essays in Torture are, almost without exception, excellent. Each is thoughtful and readable; many are provocative; some are both profound and moving. Together, they offer an excellent introduction to the troubling issue of torture, which remains, as many of the book's contributors note, both universally condemned and widely practiced throughout the modern world.
Nonetheless, there is something both curious and disturbing about the very project of assembling such a learned collection of essays about torture. Levinson acknowledges this in his introduction. Everyone, he observes, agrees that torture is a profound evil. It is illegal under both U.S. and international law, and one of the most universally condemned of practices. So why, all of a sudden, a new book on torture? "What, a reader may well ask, is there to think about or contemplate?" (24)
Levinson offers his own ambivalent answer: torture is evil, but "this being acknowledged, . . . [there is] an increasingly important debate over the possibility that torture, at least in some carefully specified circumstances, might be a lesser evil' than some other greater evil' that menaces society." (24) Later, he goes even further, noting unhappily that "it turns out to be surprisingly hard to avoid granting some element of legitimacy' to torture. . . ."(35)
Levinson starts from the position of the intellectually and morally tormented liberal, fearing that after September 11, intellectual honesty requires him to concede that his liberal instincts may have led him astray. Thus the slightly hand-wringing quality of the book and many of its essays. But the hand-wringing only makes sense if one accepts two fundamental and related premises, one being that the so-called ticking bomb scenario has important legal and policy implications, and the second being that after 9/11, the threat confronting us is so great as to merit a reexamination of our longstanding opposition to torture. I want to question both of these assumptions.
I. Ticking Bombs
As David Luban has observed, the so-called ticking bomb hypothetical "has become the alpha and the omega of our thinking about torture." n1 Certainly, some variant of the ticking bomb hypothetical recurs in at least half of the essays in Torture, and casts a shadow over even those essays that do not explicitly discuss it.
n1 David Luban, Liberalism and the Unpleasant Question of Torture, 5. Unpublished manuscript.
Alan Dershowitz is perhaps most aggressive in his insistence that the ticking bomb scenario requires us to radically reconsider absolute prohibitions on torture. As Dershowitz infamously argued in a Los Angeles Times op-ed shortly after 9/11, and argues again in his contribution to Levinson's volume, surely most of us would agree that it is better to torture one bad guy than allow millions of innocents to die. If we agree that torture should and will at times be used by state agents, Dershowitz insists, then we should consider "torture warrants" or some similar mechanism for ensuring judicial oversight of the practice. Anything less, he asserts, is "hypocritical posturing," fundamentally incompatible with the rule of law. (257)
Although almost all of the other contributors to Torture decry Dershowitz's notion of "torture warrants," most allow the ticking bomb hypothetical to set the terms of the debate, even as they struggle against the lessons Dershowitz draws from it. Oren Gross acknowledges the slim likelihood of such an extreme situation arising in real life, and Elaine Scarry notes that we almost never have the sort of perfect information (about the suspect's guilt, the bomb's reality, etc.) that the hypothetical takes for granted. Levinson also points to the related "slippery slope" issues: if torturing the captured terrorist does not produce the needed information, can we up the pressure by torturing the terrorist's wife? His innocent child? How far will we go down the path of cold consequentialist logic?
These are all important criticisms, but there is another and still more fundamental flaw in arguments that are based on ticking-bomb hypotheticals. The major flaw is the assumption that it matters -- either to morality or to law -- that in certain circumstances, almost every single human being would use torture to avoid some alternative perceived as worse.
As Luban puts it, the ticking bomb scenario is used "to force the liberal prohibitionist to admit that yes, even he or she would agree to torture in at least this one situation . . . [then] she's down in the mud . . . and the only question left is how much further down she will go. Dialectically, getting the prohibitionist to address the ticking time-bomb is like getting the vegetarian to eat just one little oyster because it has no nervous system. Once she does that -- gotcha!" n2
n2 Ibid.
But as Luban's comments suggest, the rhetorical power of the ticking bomb scenario can easily overwhelm logic. In fact, the ticking bomb scenario tells us little of any real import, and it is a waste of energy to fight it. Those who believe that the law's absolute prohibition on torture is necessary (as I do) should concede at the outset that it is child's play to construct hypotheticals in which most of us would use torture.
I readily concede that I would use torture were I the protagonist in the ticking bomb scenario. In fact, it would take far, far less than a nuclear bomb in New York to get me to torture someone. Take another common variant of the ticking bomb scenario, the "kidnapped child" hypothetical, in which we are asked if we would torture a captured kidnap suspect to learn the hidden location of a child who will soon die if not rescued. Here too, I don't doubt that I would readily use torture. For that matter, if I believed torturing someone would prevent my own child from suffering, I would turn to torture in a heartbeat.
But there is no contradiction between an acknowledgment that one would use torture in some situations and a simultaneous conviction that torture is an absolutely and inherently wrong. This is because the lesson of the various ticking bomb scenarios has nothing to do with human morality, and everything to do with human psychology. Ironically, ticking bomb hypotheticals teach us nothing we cannot already deduce from the practice of torture itself. The lesson of both is that there are pressures too terrible for most humans to withstand.
Interrogators turn to torture because they understand this. It is a truth of human psychology that almost all people can be broken, given enough physical or psychological pressure. Every person has a different breaking point: we each experience physical pain differently, and can tolerate it to different degrees, and our varying psychological defenses allow us to withstand varying degrees of stress, fear, and mental pressure. But under intense physical torture, the most courageous political dissident and the most hardened terrorist may alike betray their own families. Under intense psychological pressure (in the context of threats to torture or kill their children, for instance), each may betray their friends and political comrades.
This is why the various ticking bomb scenarios merely restate what is already evident from the practice of torture itself: that under enough pressure, virtually all of us will betray even those values we hold most dear -- ideology, loyalty, friendship. Ironically, the imagined protagonist in the ticking bomb scenarios is in a position structurally parallel to that of the captured suspect. Call her "the Good Torturer." Her opposition to torture is a value to which she is absolutely committed. Yet she is under unbearably intense and urgent psychological pressure: if she cannot quickly induce the suspect to talk, something she holds dear will be inexorably, agonizingly destroyed.
In the bomb-under-New York hypothetical, the Good Torturer faces the imminent death of millions of innocents, among whom, no doubt, are some of her own loved ones or friends. In the kidnapping hypothetical, an innocent child -- her own, in some versions of the hypothetical -- will suffer and die if she does not use torture. Nonetheless, our imagined protagonist knows that if she tortures the suspect, she will betray her own deeply held values. But she cannot withstand the pressure she is under. She breaks, and turns to torture, in the hope that her victim will similarly break, and in his desire to end his own physical agony, he will tell her what she needs to end her own psychological agony.
To ask whether the interrogator's use of torture was "justified" or "the lesser evil" in such situations misses the point. To understand this, consider for a moment the "good" torture victim: not the sinister terrorist, but the courageous dissident, captured by brutal agents of the repressive regime under which his country suffers. Brave as he is, he is only human, and subject to the usual human weaknesses. If he ultimately buckles under agonizing torture and betrays his comrades and his family, we do not condemn him: we recognize that we ourselves could do no better. But neither do we use his tragedy to argue that betraying one's loved ones and ideals is sometimes "justified" or the lesser evil. The courageous dissident's betrayal is still a betrayal; his tragedy always a tragedy. We should forgive him for his betrayal of what he held dear, and we can hope that, with time, he will come to forgive himself as well. But we feel no need to reconsider our aversion to betrayal.
The ticking bomb hypothetical thus forms no basis for questioning either our moral intuition that torture is wrong or our legal commitment to prohibiting and punishing torture. Under enough pressure, and almost all of us would do any number of terrible things: steal, kill, betray our comrades, use torture ourselves. But it does not therefore follow that we should have a moral code or legal regime which ceases to contain ex ante prohibitions on stealing, killing, treason, or torture.
Neither law nor morality is based on a real-time snapshot of humans as we actually are. As Ariel Dorfman puts it in his foreword, the prohibition on torture is based on the hope that there may come "a day when humanity lives up to the best image of itself reflected in the mirror of the future." (6) In this sense, the various ticking bomb hypotheticals are based on a false premise: that just because numerous terrible acts are psychologically inevitable, we ought to reconsider our prohibitionist instincts, and perhaps even permit them ex ante.
Acts permitted by law probably occur with greater frequency than acts absolutely prohibited by law, so if we think it best that certain acts occur only exceptionally rarely, if at all, it is wise to prohibit them. Unquestionably, as Alan Dershowitz notes, this leaves us open to charges of hypocrisy. But why should we be troubled by this? The law is full of hypocrisy, and could not be otherwise. In this sense, there is nothing "different" about the legal challenge presented by torture.
Dershowitz is correct to argue that there is danger in allowing too great a gap to open between the law on the books and the practices that we know will occur in reality. If the law prohibits too many widely accepted practices, it may come to seem hypocritical or irrelevant, and resistance not just to seemingly out-of-step legal requirements but to all legal requirements may increase. (We can see this happening, to some extent and in some communities, when we look at heightened penalties for minor drug crimes, for instance.) But this objection misunderstands the complex dialectical relationship between law and social norms.
The law can unquestionably be out of sync with evolving social norms, and given the ways in which law is made, to some extent it is inevitably at least slightly out of sync.
But in societies with robust rule of law traditions, social norms do not evolve independently of the law. In the United States, where our understanding of ourselves has much to do with the notion that we are a society committed to the rule of law, even most skeptics accept that law plays some role in shaping popular perceptions of accepted practices. Law has many functions in a complex society, and one is to structure collective interpretations of violence. How the law regards torture affects how ordinary people regard torture.
It is worth noting that if the thought of torture leaves us uniquely chilled, it is not primarily because of what it does to victims -- the taboo against torture is far more absolute than the taboo against killing -- but because of what it does to perpetrators. We believe torture is wrong because we believe it wrong to seek to break the will of another, wrong to violate his or her autonomy. n3 Why we consider this uniquely appalling is a separate question, and one I won't try to answer. If we oppose torture, it is, and should be, because we believe that the act of torture necessarily involves a deep distortion of the torturer's commitment to a shared vision of human dignity.
n3 Cf. the 5th Amendment's insistence that no one "shall be compelled in any criminal case to be a witness against himself."
Back to the ticking time bomb and the "good" torturer. Assume the torture pays off: the suspect reveals the location of the hidden bomb, which is disarmed in the nick of time, saving New York from nuclear annihilation. Is it wrong that the "good torturer," who betrayed his or her own ideals when confronted with awful pressures, may face possible criminal charges?
Both moral theory and criminal law accept that ordinary people and political leaders alike sometimes face morally terrible choices. Thus Michael Walzer, in his classic discussion of "the problem of dirty hands,'" notes that though "a particular act of government . . . may be exactly the right thing to do in utilitarian terms [it may still] leave the man who does it guilty of a moral wrong." (61 -- 62) Jean Bethke Elshtain agrees, situating the debate in the Christian theological tradition; she suggests that if the imperatives of political leadership appear to require torture in some exceptional circumstance, the "dirty-handed politician" must still "stand before God and seek forgiveness."
If we want torture to occur only in the most terrible and exceptional of circumstances, we would do well to maintain its taboo status through every means at our disposal. But treating torture as a grave crime does not require that all torturers be convicted and harshly punished, any more than treating homicide as a crime requires that every killer be convicted and harshly punished. At times, all choices are terrible, and the existence of criminal defenses such as duress and necessity reflects this awareness, as do doctrines of mitigation and clemency. As Oren Gross and Miriam Gur-Arye suggest, some or all of these doctrines might be appropriately invoked should true "ticking-bomb" scenarios ever arise.
II. Catastrophes
Ticking-bomb hypotheticals, though always a stock feature of undergraduate ethics courses, were not part of serious American political discourse until after September II. As Levinson notes, however, the debate about torture rapidly gained legitimacy after the attacks. Before September 11, there would have been little reason for a new book on the moral and legal issues surrounding torture.
Not that there were not plenty of torture-related issues worth discussing: its history in law, culture and practice; efforts to monitor, reduce, and punish its use, and so on. But prior to September 11, most of us would have said that the law on torture was crystal clear, and appropriately reflected torture's status as absolutely prohibited. Prior to September 11, it is difficult to imagine that any serious intellectuals or politicians would have disagreed.
The current debate about torture presupposes that after 9/11, something is "different," so very different that we must reopen a previously closed issue (with obligatory murmurs of reluctance). But what is different? Terrorism does present tactical challenges that differ from the tactical challenges presented by wars between states or by threats from traditionally organized insurgent groups. But the claim that terrorism requires us to rethink the absoluteness of prohibitions on torture cannot be cast in terms of the differences between terrorism and state-sponsored violence. Yes, terrorists might hide bombs in cities or plan hijackings. But states with whom we are at war could do the same, or aim ballistic missiles at us, or plan surprise assaults on our armies or our cities.
Terrorists may organize themselves differently than traditional armies and use slightly different techniques, but it would be frivolous to claim that they can do the U.S. (or any other state) greater harm than traditional war. Just over 3,000 people were killed in the 9/11 attacks. In all other recent years, total deaths (not just of Americans) caused by international terrorism worldwide have never exceeded (or even approached) 1,000 a year. n4 Compare this to the Vietnam War (58,000 U.S. soldiers dead), the Korean War (37,000 U.S. soldiers dead), World War I (116,000 U.S. soldiers dead), or World War 11 (405,000 U.S. soldiers dead). n5
n4 See "Patterns of Global Terrorism," U.S. Dept. of State, www.state.gov/s/ct/rls/pgtrpt/.
n5 See http://web1.whs.osd.mil/mmid/casualty/WCPRINCIPAL.pdf.
If one wishes to claim that terrorism is a "worse" threat because it is directed mainly at civilian targets, then think of the civilian deaths caused by traditional armed conflicts between states: World War I killed about 6.5 million civilians. n6 World War Two killed 40 million. n7 Vietnam left somewhere between two and four million civilians dead. n8 If the U.S. is unaccustomed to the experience of having its own civilian casualties, this is a function of geographic good fortune (most of our recent wars were fought far from home). During the American Civil War, more than half a million Union and Confederate soldiers died, n9 and an unknown (but probably large) number of civilians also died.
n6 See http://en.wikipedia.org/wiki/World_War_I_casualties#Civilians_killed.
n7 See http://en.wikipedia.org/wiki/World_War_II_casualties#Civilians_killed.
n8 See http://en.wikipedia.org/wiki/Vietnam_War.
n9 See http://www.cwc.lsu.edu/cwc/other/stats/warcost.htm.
Terrorism is bad. We may even choose, with George W. Bush, to call it "evil," although this is not an analytically useful category. But there is no meaningful sense in which terrorism is "worse" or "more evil" than various other intentional harms humans have inflicted on one another during the last half century. Is al Qaeda worse than the Nazis? Worse than Stalin? Worse than the Khmer Rouge? Worse than Rwanda's genocidaires? Worse than the Bosnian-Serb militias?
If we believe that torture can produce useful intelligence information, it would presumably have been helpful for the U.S. government to formally authorize torture in any number of prior conflicts. Torturing German POWs might have revealed critical details of Nazi war plans, for instance, thereby averting countless allied military and civilian deaths. Yet serious American intellectuals and politicians did not argue that torture was necessary to combat any of these numerous other grave threats. Indeed, although torture was undoubtedly used from time to time by American soldiers and other officials, we defined ourselves precisely by our moral distance from our foes. Nazis, Stalinists, and the like were our enemies in part because their ideologies justified torture and other atrocities. Official sanction of torture is what made "them" them and made us "us."
One of the earliest legal prohibitions on torture as a means of national defense came from the one conflict which most directly imperiled the survival of our nation, the American Civil War. In General Order 100, "Instructions For The Government Of Armies Of The United States In The Field" (better known as the Lieber Code of 1863, which became the foundation for the modern law of armed conflict, including the Geneva Conventions), the directions to the Union army were unambiguous. "Military necessity does not admit of cruelty -- that is, the infliction of suffering for the sake of suffering or for revenge, nor of maiming or wounding except in fight, nor of torture to extort confessions." n10
n10 See http://www.civilwarhome.com/liebercode.htm.
Nothing is different now than it was in 1863. If anything, we should be much wiser now than we were then: after all, we have the terrible history of the 20th century to remind us that "just a little" officially sanctioned torture always becomes just a little bit more, and then a little bit more again, until the exception swallows the rule.
The various claims that torture might be justified to prevent catastrophes all assume that we will know a catastrophe when we see one. But as the comparison of death rates from various wars suggests, catastrophe is relative. A threat to one of my children would be catastrophe enough for me, but presumably not to others. How bad would a catastrophe have to be to justify torture from some "objective" standpoint? How many deaths would constitute a catastrophe? If torture was not officially deemed necessary by the U.S. military in any prior war, why on earth would torture be necessary to prevent another 9/11?
This, of course, is the problem with catastrophe-based exceptions, as Oren Gross notes in his own essay. We cannot define catastrophe quantitatively, because threats to "us" will always seem more catastrophic than threats to "them." But "we matter more than you do" cannot be an ethical basis for using torture.
Catastrophe-based exceptions to prohibitions on torture also rest on the assumption that there exists an individual and collective imperative to prevent death and suffering whenever it is possible to do so. This is behind the narrow consequentialist logic of the ticking-bomb scenario: if torturing one person could prevent 1,000 people from dying, then a public official has a duty to use torture. But do we, as a society, actually believe that any of us, whether public officials or private citizens, are required to take every action that could conceivably save lives?
It's a pleasant myth, but no more than that. As a society, we continually sacrifice the lives and well-being of the few for the mere convenience of the many. Approximately 40,000 Americans die in automobile accidents each year. We call them "accidents," and make sense of them as such, but they are "accidents" only in the sense that we do not intend them to occur. We know they will occur, however, and we also know that this particular death rate is a product of policies that we could change, if we cared enough. (The U.S. death rate from car accidents is nearly twice that of Japan and three times the rate in England). n11 We could lower speed limits, require better safety features in cars, and improve the quality of our roads. On the whole, though, we choose not to. To put it starkly, as a society, we consider 40,000 auto deaths per year a death level that we are willing to tolerate, in exchange for various conveniences and cost savings.
n11 See http://www.scotland.gov.uk/library2/doc05/ras-11.htm.
As a matter of ethics, as opposed to politics, there is no more a requirement that we do "absolutely everything" to prevent American deaths from terrorism than there is a requirement that we do absolutely everything to prevent deaths from automobile accidents. With torture as with other policy measures that could conceivably save American lives, we should indeed make a cost-benefit analysis. While the "cost" of using torture may appear low in the context of the "exceptional" ticking bomb scenario (cost = temporary agony for one bad guy; benefit = saved lives of many good guys), the problem with torture is that precisely because catastrophe is always relative, torture, if accepted by law ex ante in any situation, would quickly become routinized.
A proper cost-benefit analysis recognizes the empirical validity of this claim: societies that sanction torture in "exceptional" circumstances quickly begin to use it more and more. n12 The torture game involves repeat players, and torture ultimately becomes embedded in institutions and structures. As a result, the costs of torture increase exponentially, not linearly. As John Langbein notes, "Efforts to accommodate torture within the Western legal history have a long history . . . [but] across the centuries it became clear that the safeguards were far from safe." (93) His own study of torture in the European legal tradition leads him to a bleak conclusion: "History's most important lesson is that it has not been possible to make torture compatible with truth." (101)
n12 Consider the Israeli experience.
The point is not merely that the slope is slippery: The slope is made of black ice, and the lesson both of social psychology n13 and recent history is that the descent is blindingly fast. Societies can become brutalized in the blink of an eye: it took stunningly little time for ordinary Germans, ordinary Serbs, and ordinary Rwandans to see atrocity and genocide as the acceptable norm, not the unimaginable exception.
n13 Consider the Milgram experiments.
The story of the banality of evil is not new. As Richard Weisberg's ominous essay on lawyers in Vichy France underlines, it always begins with public hysteria; with troubling hypotheticals advanced by influential elites; with small, almost imperceptible legal evasions; with dry bureaucratic euphemisms. The path runs through Guantanamo, Bagram, and abu Ghraib, but does not end there. Still, abu Ghraib alone should be a reminder that we Americans are no more immune from the plunge into the moral abyss than we are immune from the violence that is endemic to our world.
III. Conclusion
The essays in Torture were mostly completed before the abu Ghraib photos sparked a world-wide scandal in the spring of 2004, and before the hemorrhage of internal government memos showed us all how quickly torture and torture-like practices became routinized in post 9/11 U.S. policy. This does not make Torture dated, since its focus is on more general philosophical and legal questions. Nonetheless, I wonder if Levinson would undertake precisely the same project if he were starting today, or if his contributors would take the same approaches.
The question is unanswerable, but I suspect that the shock of the last year's revelations has led many liberals to reconsider their initial post 9/11 willingness to reopen the issue of torture. Nothing, after all, really changed after 9/11, except that we Americans have discovered our lack of immunity to the violence and terror that has always been a daily possibility for many others around the globe.
In those first appalled months after September 11, we reacted with child-like egocentrism, and our willingness to engage in a hand-wringing reexamination of torture reflects, more than anything, our sense of helplessness and vulnerability. (After all, torture in the modern era has been a weapon of the weak: states that make widespread use of torture are mainly poor states with badly-trained police and judiciaries. When you lack well-trained investigators, fancy forensics labs, and the like, torture is a tempting shortcut to solving crime and cracking conspiracies.) After 9/11 we share a baffled sense that there are people out there who want to hurt us, but our insularity and ignorance -- our dearth of "human intelligence," as they say -- makes it hard to understand or prevent efforts to do us harm. How tempting to imagine that torture -- just a little tiny bit of torture, in the most carefully limited circumstances -- may be the silver bullet that can save us all! The price, however, is too high.
"I cannot praise a fugitive and cloistered virtue, unexercised and unbreathed," said Milton. But virtue is always a fugitive from vice. The goal of every "civilized" society must surely be to build and sustain the cloisters that permit the routine exercise of virtue. Unheroic, perhaps: but most humans are not particularly heroic; they do what peers and authority figures expect of them. Without a clear and constantly reaffirmed prohibition on torture, we will quickly become a society that accepts far worse.
As Richard Weisberg suggests, lawyers and judges have a particular responsibility to sustain the cloisters that protect the exercise of virtue. In his famous dissenting opinion in Korematsu, Justice Robert Jackson (later the chief American prosecutor at Nuremberg) commented that judicial legitimation of acts taken in the name of national security necessity holds particular dangers:
A military order, however unconstitutional, is not apt to last longer than the military emergency. . . . But once a judicial opinion rationalizes such an order . . . the principle then lies about like a loaded weapon ready for the hand of any authority that can bring forward a plausible claim of an urgent need. Every repetition imbeds that principle more deeply in our law and thinking and expands it to new purposes. . . . A military commander may overstep the bounds of constitutionality, and it is an incident. But if we review and approve, that passing incident becomes the doctrine of the Constitution. There it has a generative power of its own, and all that it creates will be in its own image. n14
n14 Korematsu v. United States, 323 US 214 (1944).
This reasoning was valid before September 11, and it is every bit as valid today.