It should not be surprising that the long awaited release in December 2014 of the Senate Select Committee on Intelligence (SSCI) Report on the CIA’s Detention and Interrogation did not bring a conclusive end to the debate over the use of torture or enhanced interrogation techniques by the United States. To be sure, John Brennan, the Director of the CIA, acknowledged that the report correctly identified numerous and significant problems with the CIA’s handling of detainees and interrogations in the aftermath of the September 11 attacks. Brennan was also emphatic in stating his own belief that “enhanced interrogation techniques are not an appropriate method to obtain intelligence and that their use impairs our ability to continue to play a leadership role in the world.” But Brennan also restated the CIA’s long-held objection to the SSCI report’s “unqualified assertions that the overall detention and interrogation program did not produce unique intelligence that led terrorist plots to be disrupted, terrorists to be captured, or lives to be saved.”
H-Diplo | ISSF Forum, No. 5 (2015)
A production of H-Diplo with the journals Security Studies, International Security, Journal of Strategic Studies, and the International Studies Association’s Security Studies Section (ISSS).
H-Diplo/ISSF Forum Editors: James McAllister and Diane Labrosse
H-Diplo/ISSF Web and Production Editor: George Fujii
Commissioned for H-Diplo/ISSF by James McAllister
Introduction by James McAllister
H-Diplo/ISSF Forum on the Senate Select Committee on Intelligence (SSCI) Report and the United States’ Post-9/11 Policy on Torture
Published by H-Diplo/ISSF on 16 February 2015
- Introduction by James McAllister, Williams College. 2
- Review by Jonathan Kirshner, Cornell University. 4
- Review by Austin Long, Columbia University. 10
- Review by Robert A. Pape, University of Chicago. 18
- Review by Joshua Rovner, Southern Methodist University. 43
© 2015 The Authors.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 United States License.
It should not be surprising that the long awaited release in December 2014 of the Senate Select Committee on Intelligence (SSCI) Report on the CIA’s Detention and Interrogation did not bring a conclusive end to the debate over the use of torture or enhanced interrogation techniques by the United States. To be sure, John Brennan, the Director of the CIA, acknowledged that the report correctly identified numerous and significant problems with the CIA’s handling of detainees and interrogations in the aftermath of the September 11 attacks. Brennan was also emphatic in stating his own belief that “enhanced interrogation techniques are not an appropriate method to obtain intelligence and that their use impairs our ability to continue to play a leadership role in the world.” But Brennan also restated the CIA’s long-held objection to the SSCI report’s “unqualified assertions that the overall detention and interrogation program did not produce unique intelligence that led terrorist plots to be disrupted, terrorists to be captured, or lives to be saved.”
H-Diplo/ISSF is delighted to present a special forum on torture and international relations. Our four contributors to this forum have all published extensively on issues related to intelligence, terrorism, strategy, and the nature of international relations among other topics. In his essay, Jonathan Kirshner explores the often ignored but fundamental questions that must be addressed in any analysis of whether ‘torture works.’ Exploring in detail several of the cases examined in the SSCI report, Austin Long argues that the SSSCI report presents a flawed critique of the CIA’s actions after September 11. In his lengthy essay, Robert Pape examines the relationship between short-term tactical benefits and the long-term strategic objectives of American foreign policy. He concludes that “the highly limited tactical benefits are easily outweighed by the heavy strategic costs.” Joshua Rovner also addresses the tradeoff between tactical and strategic considerations in his essay. In his view, “the long-term fallout is uncertain. It is usually easier to evaluate tactical results than strategic consequences, and it does not help that there are no obvious historical analogies to guide our judgment in this case.”
H-Diplo/ISSF thanks all of the contributors for taking the time to contribute to this timely and important debate. We also thank Robert Jervis for organizing this forum. His forthcoming essay in Foreign Affairs on the SSCI report will surely be of interest to everyone interested in the ongoing debate over torture and the CIA.
Jonathan Kirshner is the Stephen and Barbara Friedman Professor of International Political Economy in the Department of Government and Director of Reppy Institute for Peace and Conflict Studies, Cornell University. He is the author of Currency and Coercion: The Political Economy of International Monetary Power, and Appeasing Bankers: Financial Caution on the Road to War, which won the 2009 best book award from the International Security Studies Section of the International Studies Association. His most recent books include Hollywood’s Last Golden Age: Politics, Society and the Seventies Film in America, and American Power after the Financial Crisis.
Austin Long is an Assistant Professor at the School of International and Public Affairs and a Member of the Saltzman Institute of War and Peace Studies and the Harriman Institute of Russian, Eurasian, and Eastern European Studies at Columbia University. Long received his B.S. from the Sam Nunn School of International Affairs at the Georgia Institute of Technology and his Ph.D. in Political Science from the Massachusetts Institute of Technology.
Robert Pape is Professor of Political Science at the University of Chicago and director of the Chicago Project on Security and Terrorism. He is a leading expert on the culture of suicide terrorism and his books include Dying to Win: The Strategic Logic of Suicide Terrorism and Cutting the Fuse: The Explosion of Suicide Terrorism and How to Stop It.
Joshua Rovner is the John Goodwin Tower Distinguished Chair in International Politics and National Security at Southern Methodist University. He is also associate professor of political science, Director of Studies at the Tower Center for Political Studies, and Director of the Security and Strategy Program at SMU. Rovner is the author of Fixing the Facts: National Security and the Politics of Intelligence (Cornell University Press, 2011). His recent work on intelligence also includes “Intelligence in the Twitter Age,” International Journal of Intelligence and Counterintelligence, 26/2 (Summer 2013); and “Is Politicization Ever a Good Thing?” Intelligence and National Security, 28/1 (Spring 2013).
Who am I?’ ‘Why am I here?’ Readers of a certain age will recall these as the questions asked by James Stockdale during the Vice-Presidential debate in 1992. Stockdale knew something about torture—the Medal of Honor recipient spent over seven years in a brutal North Vietnamese prison where he was subject to grievous mistreatment at the hands of his captors. I, however, know next to nothing about torture. I am not well-versed in the academic literature that has studied it, and I am in no position to adjudicate detailed disputes between the Senate Select Committee on Intelligence Report and its critics.
My contribution to this forum, then, is limited to the question of how, as students of international politics, we are to understand whether or not torture ‘works.’ And here I have something to say—indeed, to insist: we can only understand whether something ‘works’ in its broader political context. This is not a rhetorical cliché; it is an analytical law, and one of the basic lessons of any freshman class in International Relations. It applies to diplomatic maneuvers, economic sanctions, coercive diplomacy, military invasions, nuclear war . . . and to the policy of torturing captives held prisoner. This foundational, political metric is associated with Carl von Clausewitz: we evaluate the success (or failure) of any action based on whether it brings us closer to (or farther from) achieving the overarching political goals for which the measure was introduced. But in many if not most instances, the torture apologists—certainly the most prominent among them in the public sphere—disregard this essential point. As a result, both the torture enthusiasts as well as those more objective scholars who would evaluate the recent American practice of torture outside of that broader, essential context, risk proffering or implying misguided or even counterproductive policy advice.
Torture is a tricky issue for scholarly endeavors, because its moral burdens are so great they cannot be easily set aside. Torture is the Spanish Inquisition, torture is the Gestapo. It is impossible to sweep this under the rug, or to fail to acknowledge the essential intimacy of these relationships. Nevertheless, this does not mean that normative revulsion cannot be separated from dispassionate analysis. Economists can debate the economic benefits of slavery—one could even hold that it was a vital contributor to American economic development—without endorsing slavery. Similarly, in international politics we study terrible behavior all the time. And we can investigate, and should investigate, things like whether bombing (or avoiding) civilian targets in warfare is effective in contributing to military victory, even if we hold the view that purposefully targeting (that is, killing) civilians is wrong.
But this vital distinction does not clear the normative decks. It is possible to assess torture without reference or appeal to a personal moral code, but it is not possible to assess torture without taking into account its interpretation by other actors. A scholar can withhold moral judgment of a slaveholder for the purpose of analysis, but it is still necessary to understand how that identity, “slaveholder,” is received and understood by contemporaries with whom the slaveholder is interacting. Similarly, advancing goals in world politics is a strategic, recursive, interactive process. Actions taken by one state are interpreted by others through normative lenses, and this informs the political responses of other states, for better or worse. When one tortures, one is properly understood by others to be a torturer. This counts.
None of this—the insistence that actions are undertaken to advance political goals and can thus only be assessed in that context, and that actions are interpreted by others through normatively tinted political lenses—is special to torture. It pertains generally to the use and misuse of the concept of what ‘works’ in international politics. For example, the U.S. and Canada often become involved in disputes over fishing rights; at times these arguments get heated before a negotiated settlement is reached. In theory, during one of these confrontations, the U.S. could threaten to carpet-bomb successive Canadian cities. And those threats might ‘work’—that is, the dispute would be settled on American terms. But most students of international relations would still think it unwise to resort to such tactics, even if they had no moral qualms about them, and even if we knew for sure they would be effective, and achieve the desired results—more fish for us. Because the resort to the tactic likely would have significant, negative effects on other international political goals of the U.S., and to fail to account for those effects in assessing the tactic of ‘threatened carpet-bombing’ is to be a bad analyst of international relations.
As a more practical illustration of the imperatives of this perspective, imagine that towards the end of 1967, the U.S. won the Vietnam war by dropping three small, tactical nuclear weapons on the North. What would have been the result? The benefits are large and clear. The U.S would have won the war. The action would have saved American lives (more than 34,000 over the next three years). In addition, given the vast number of civilian causalities over the course of the war, it might even have arguably ‘saved’ more civilian lives than it incinerated. But would the resort to nuclear weapons have ‘worked’? An analysis from the perspective of international relations requires us to rephrase the question: would U.S. international political goals have been advanced by that action? The answer is no. It is not possible to understand the Vietnam War outside of its Cold War context. Vietnam itself mattered little—never has the phrase ‘geopolitical backwater’ been so often or accurately invoked—it mattered only for its meaning in the larger international political struggle between the United States and the Soviet Union. ‘Winning’ the Vietnam War through the introduction of nuclear weapons would have almost certainly been a geopolitical catastrophe for the Americans, and any assessment of the efficacy of nuclear weapons in that conflict—especially one that implied that nuclear weapons would ‘work’—would be derelict if it failed to take that into account.
This holds true for torture as well. If we want to consider whether torture will advance American interests, it requires assessing the full range of costs and benefits that result from embracing the policy. At the tactical level, a superficial look at the evidence suggests that it is mixed. It is certainly the case that, at this micro-level, torture can sometimes ‘work’—that is, yield actionable information that you are better off having than not having. Countless historically-informed films (if not the contested ones that might leap first to mind) are compelling on this point. The Confession (Costa-Gavras, 1970), showed how torture (what its defenders would call ‘enhanced interrogation techniques’), worked very well on the eve of the Czech show trials in 1951; The Battle of Algiers (Pontecorvo, 1966), illustrated with documentary-like detail how torture (what its defenders called ‘torture’), yielded information that led to the exposure of terrorist cells. These dramatizations reflect an underlying reality—it is clear that the ruthless application of violence against the human body can achieve such results. (Torture advocates can also point to more recent successes. In 1984 CIA Beirut station Chief William Buckley was kidnapped, savagely and extensively tortured, and eventually gave his captors valuable information about CIA operations.)
But even at the tactical level, these waters are muddier than they might first appear. It is important to keep in mind that a policy of torture yields other outcomes as well. As the net is cast, it is not only knowledgeable kingpins who are reeled in. Often one is torturing suspects, who may have useful information. This means that some of the people being tortured are in fact innocent, and (worse, from the torturer’s perspective) they do not have any useful information to reveal. Both the innocent and the guilty (and all those many in-between) face tremendous incentives to stop the torture, and will, all of them, say whatever it is they think will stop it. (Nor is this process quick and easy—the CIA tortured some prisoners again and again over a period of months and months.) Separating out the actionable wheat from the agony-induced chaff is not easy, a problem likely to be greater the more widespread the adoption of the policy. It is said that a talented torturer can cut some of these corners—that good torture is not so much about the fantasy of stopping ticking time-bombs as seen on TV shows like 24, but about asking questions designed to corroborate the known, or to glean hints from what types of information the prisoner seems most eager to conceal. But you go to the dungeon with the torturers you have, not the torturers you want, and the more widespread the torture, the thinner the talent pool will be.
Net assessments of tactical costs and benefits of torture are thus complicated, and do not lend themselves to easy or obvious measurements. Words and phrases used to tout benefits, such as that torture was ‘effective’ or provided ‘valuable information’ are (perhaps necessarily) quite squishy, and consistent with minimalist and underspecified claims. And even if the benefits of the successes could be plausibly tabulated, they would have to be properly discounted against the possibilities the information could have been otherwise obtained, and against the costs of actions taken or not taken as a result of failures (yes, sometimes torture works, but sometimes it does not).
But even if we can point to some tactical successes, and even if, hypothetically, we could at some point reach the assessment that, overall, torture was relatively effective in gathering information, we would still not yet have asked, or properly answered, the central, relevant question: are American interests advanced or undermined by embracing torture? For students of international relations—again, even when checking their own riotously clanging moral alarm bells at the door—the question remains: what are the international political consequences of the U.S. adopting the following policy, which can be dubbed the torture doctrine: ’When we capture people whom we suspect of having information that we want to know, and they do not want to tell us the information or claim not to have that information, we will torture them until we are satisfied that we have learned everything from them that we possibly can.’
One issue that quickly arises in this broader context is how it affects U.S. claims that America is a force for good in the world, and that its policies are not simply in the ruthless service of its own self-interest, the rights and concerns of others be damned. The U.S. often stands accused that its righteous rhetoric masks a more Machiavellian track record. Embracing torture as policy makes it much harder to claim that there is a difference between ‘us’ and ‘them.’ But does it matter if there is such a difference? Or is behavior in world politics indeed nothing more than a no-holds-barred, amoral clash of naked, competing interests, with claims to decency and justice a fluttering veil over malevolence, and the U.S. is little more than Syria with a smile? My own view is that there is a difference, and that it matters; and as this pertains to torture my perspective is similar to one articulated by the historian Tony Judt with regard to the practice of preventive war. It was unwise “for democracies to make war unprompted,” he argued. “Because that is corrosive, not only of the exemplary quality of democracies—without which they cannot lecture dictatorships—but it is also internally corrosive of what democracies are supposed to be about.”
The introduction of corrosion reveals an additional pathology, the virulence of which is not to be underestimated. Once again, embracing ‘torture’ does not get you only those few moments where, through torture, a bad guy tells you a useful thing. You get all the torture, including the torture of suspects, including the torture of innocents. You get all of that, and you get more of it. Because to adopt a policy of torture is to establish a culture of torture, a norm of torture, an understanding that, like TV mob boss Tony Soprano, torture is part of what we do to get what we want. This will inevitably cascade through the U.S. military and affect the practice of its foreign policy. Along the military dimension, torture will become more common down the chain of command and in the field—as stressed actors in dangerous situations ask and answer the implicit questions: if we know it works, and we know there is nothing wrong with it, why not here and now, when we desperately need it? As for U.S. diplomatic efforts on behalf of human rights, and of the assessment of American behavior and motives more generally, the deleterious effects of the torture doctrine would appear self-evident. Once again, the key issue here is not to insist that our own morals guide policy, but to understand that the moral sense of others will shape their reception of our behavior. For most of humanity, torture—worse still, a cold-blooded, calculated, authorized torture—is very high on the list of barbarous acts. The erosion of U. S. political influence and influence and moral authority on the world stage could not but follow.
In sum, to assess whether torture ‘works,’ as with the application of military force, the only relevant metric is whether it advanced the political goals for which it was introduced, not whether it helped to gain a transient tactical advantage or win a particular military battle. (Recent U.S. history is painfully replete with examples of impressive military prowess that nevertheless failed, at great cost, to achieve the political goals for which it was introduced.) It is simply not sufficient to consider specific claims or offer assessments about the decontextualized effectiveness of particular moments of torture; it is necessary to evaluate whether any benefits gained from torture outweigh the costs associated with adopting the torture doctrine.
This International Relations approach does not require agreement with my own view that the costs outweigh the benefits—some analysts might pore over the same factors and reach a different conclusion. But it does insist that to assess whether torture ‘works,’ one must consider the full range of its costs and benefits, and always and everywhere in the context of the full range of the international political objectives articulated by the U.S. To fall short of that is to carpet bomb Canada and to ‘nuke’ Vietnam, and to subsequently sit around in bitter political poverty like the German Generals after World War I who insisted, like so many before and after, that the war was never lost on the battlefield. Failing to account for politics is untutored and sloppy analysis that will routinely yield misguided prescriptions.
To be sure, assessing the full range of ‘international political consequences’ is not easy, and things like the costs of undermining claims the U.S. can make to the legitimacy and purpose of its actions are hard to measure. But even the simplistic tactical evaluation of torture faces daunting measurement challenges (as noted, it is utterly incomplete to simply point to an example of an apparent success), and, much more important, the issue of ease of measurement is beside the point. Failing to account for factors because they are less obviously measurable would be like trying to assess the merits of the Tuskegee Syphilis Study (the non-consensual U.S. study of untreated syphilis in black men from 1932-1972) by acknowledging it was wrong but weighing the virtue of its findings solely by looking at the (scant) medical knowledge gained and without any consideration of the cost, meaning, and consequence of the entire enterprise. Such an evaluation (‘some American lives were saved’), that left uncounted the context of racial politics, relations, and legacies within the U.S. because they are hard to measure, would be shoddy, incomplete, and implicitly suggestive of dismal public policy. Measurement problems do not allow us to pretend that important factors do not exist or weigh decisively; they may suggest that distinct styles of argumentation and analysis, emphasizing reason, deduction, and haggling over contested, imperfect empirical clues, may be necessary.
In the darkest hours of the Second World War, Ernest Hemmingway wrote to a friend, “If we are fighting for what we believe in we might as well always keep believing in what we have believed.” The stakes in this debate are high, and they need to be understood. We are not playing the parlor game of ‘can you imagine a set of circumstances under which you would be willing to approve of torture?’ This is a red herring, and distracts from the singular, enormous question on the table: is it in the U.S. interest to adopt the torture doctrine? We should have this discussion, but we need to have it right.
Nearly four decades ago, Robert Jervis highlighted the human tendency to avoid value trade-offs as one of the major contributing factors to misperception in international relations. He noted that this tendency drove observers to “irrational consistency,” a state of mind in which, despite evidence to the contrary, all good things went together. This state of mind obviated the need to face painful contradictions in potential policies as the immoral must also be ineffective.
The Senate Select Committee on Intelligence “Committee Study of the Central Intelligence Agency’s Detention and Interrogation Program,” (hereafter SSCI study) appears to be a classic example of this sort of irrational consistency. In documenting the CIA’s program, which even the CIA has acknowledged had a variety of flaws, the SSCI study seems determined to conclude that enhanced interrogation is not only morally wrong but also ineffective at producing intelligence. This is unfortunate as a more measured critique would have increased the study’s impact.
Instead, the SSCI study opts for the easy way out, judging in virtually every instance in which CIA claimed enhanced interrogation techniques (EITs) yielded important intelligence the Agency overstated the value of that intelligence. The study draws this conclusion because in many instances the intelligence community already possessed some intelligence related to the topic at hand when detainees provided information. The study therefore concludes the non-detainee intelligence was enough and that detainee interrogation was irrelevant.
This interpretation of the role existing intelligence plays in interrogation literally stands the logic of elicitation and vetting on its head. If an interrogator or interviewer enters a room and knows nothing about the topic of interest, then the subject is free to manufacture a story at will. This is true of any method of elicitation, from simple interviewing to the harshest form of torture.
This essay provides an overview of the importance of intelligence already in possession of the interrogator (the intelligence background) and then specifically reviews instances in the SSCI report that makes the utility of intelligence background clear. It then notes areas in which enhanced interrogation techniques were used in the absence of good intelligence background or, worse, in instances in which the intelligence background was wrong. It concludes by observing that while the SSCI report is flawed in its desire to show enhanced interrogation never works, it is equally true that harsh measures are no panacea in interrogation.
At the risk of spoiling a good film, readers familiar with the 1995 movie The Usual Suspects have seen the importance of intelligence background in action. In the film, Verbal Kint, played by Kevin Spacey, spins an extemporaneous tale to deceive his hapless interrogator, who believes him to be a weak small-time con artist. Without an accurate understanding of who he was interrogating and what Kint knew, the interrogator (played by Chazz Palminteri) was unable to determine that Kint’s story was manufactured from the names of items and individuals in the office where the interrogation was taking place. After releasing Kint the interrogator has an epiphany that the story was manufactured, but in real life sudden realizations of deception are much less common.
Conversely, when interrogators have extensive knowledge of the subject being interrogated and the topic of interrogation the ability to elicit truthful response is greatly enhanced. Hanns Joachim Scharff, a Luftwaffe intelligence officer responsible for debriefing captured Allied pilots, is widely held to be one of the master interrogators of World War II. He used a variety of ‘soft’ rapport building techniques rather than torture but this rapport building was built on a foundation of encyclopedic knowledge of U.S. air units, techniques, operations, and the like. By leading captured airmen to believe he already knew everything about their units and operations, Scharff could elicit useful information that filled in gaps in his intelligence picture. The airmen both believed it would be pointless to try to deceive Scharff, as he could identify when they were being deceptive, and felt less concerned about being candid as Scharff already knew all.
Indeed this technique (or approach in the vernacular of U.S. interrogators) is referred to as ‘We Know All.’ In 2001-2002, U.S. interrogators found this approach to be very effective against alleged Taliban members in Afghanistan, but only after they were able to build up a very significant body of intelligence on detainees. In contrast, many of their initial interrogations were extraordinarily frustrating as they knew very little about any of their subjects. Most of those subjects claimed to either have merely been in the wrong place at the wrong time when detained, or that they were coerced into joining the Taliban. Without any background intelligence it was difficult if not impossible to verify these stories, regardless of the interrogation approach. In some cases the interrogators found that moderately coercive techniques, including sleep deprivation, were often more effective than less coercive techniques, as they could cause detainees to slip up in their denials. 
The Scharff example underscores that with a good intelligence background on the detainee and the topic of interrogation subtle, non-coercive techniques can be extraordinarily effective. The Afghanistan example demonstrates that without good intelligence, coercive techniques may offer some advantage over non-coercive techniques but are frequently not decisive. What then of the combination of good background intelligence and highly coercive techniques?
The French experience in Algeria, particularly the so-called Battle of the Casbah in Algiers, is perhaps the most well-known example of the efficacy of this combination in terms of acquiring information. According to one of the main orchestrators of very brutal techniques, including electric shocks, in Algiers, these techniques were effective in part due to their brutality and in part due to access to existing intelligence. Most notable in this regard was a secret file on criminals and insurgent sympathizers that the police had maintained, to which the military was given access.
This combination of ruthless and brutal techniques with an existing set of intelligence proved very effective. Even subsequent scholars deeply critical of the French methods acknowledge the tactical efficacy of the methods. One noted that the overall French effort (which included more than just interrogation) “… did indeed achieve impressive results in less than three months.”
The central component of success in these instances of successful interrogation, regardless of method or approach, is the intelligence background. A 2009 Intelligence Science Board study on intelligence interviewing (a term selected to be broader and less controversial than interrogation) notes that a technique or approach:
…if done professionally and with the benefit of current and emerging knowledge, occurs in context. Context includes the physical environment or setting in which the detained person resides and in which interviews take place. It also includes the interpersonal environment established by the interviewer and the intelligence interviewing team. Further, context encompasses the informational environment; it includes all the facts and perspectives about the detainee possessed by the intelligence interviewing team… As interviews and interactions take place, information from other intelligence systems is immediately provided to the team. Communications intercepts, and immediate efforts to verify the detainee’s statements and to follow up on leads, increase the “information power” of the team. Information power in turn aids interviewers to develop a productive operational accord with the detainee.
As a more personal example of the importance of the intelligence background, in the course of research I interviewed Iraqi detainees in the fall of 2007. One of the interview subjects was an Iraqi farmer who had been detained, sent to long-term detention at a U.S-run theater internment facility, and then released on four occasions. Yet he was always detained again- a curious case, as he was both unlucky enough to keep getting caught but lucky enough not to have been killed or injured.
Over the course of the interview (conducted through an interpreter) the detainee was cooperative if a bit resigned. As the interview drew to a close, I asked the detainee very directly what explained this curious pattern. He told a story that would be familiar to those who have read Stathis Kalyvas’ seminal The Logic of Violence in Civil War. The detainee claimed to have had a neighbor with whom he had a dispute over land boundaries. After the 2003 invasion the neighbor joined the new Iraqi police and promptly informed the American unit responsible for the area that the detainee was in fact an insurgent. After being detained, the detainee was interrogated, provided no useful intelligence (how could he as he knew no insurgents?) and was then sent to a theater internment facility. He was then released after a few months when it became clear he had no hostile intentions. When he returned home, his neighbor promptly informed on him again- to a new American unit as the old one had rotated out- and away he went again.
This story sounded highly plausible and would have been no more or less plausible had the detainee offered it under highly coercive interrogation rather than a relatively relaxed interview (the interview protocol allowed the detainee to have tea, cookies, and cigarettes). Yet regardless of how the story was extracted, plausibility is not the same as vetting. Fortunately there was some intelligence background on the detainee in the form of detention records.
After the interview a review of those records provided some confirmation of the story. The fourth time the detainee was detained (i.e. the instance that led to his detention before the interview) the detainee was alleged to have fired a mortar at a U.S. base north of Baghdad. Yet on the date of the alleged attack the detainee was still being held from his previous (i.e. third) detention. So unless the detainee escaped detention, traveled halfway across Iraq to fire a mortar, and then returned to detention undetected it seems unlikely he actually fired the mortar on that date. While the story could still be a lie, this background allowed at least some vetting of the story.
With this understanding of the importance of intelligence background in place, it becomes clear that the SSCI interpretation of the importance of information gained as a result of enhanced interrogation is wrong-headed in many places. For example, the SSCI report argues that “…the use of the CIA’s enhanced interrogation techniques against KSM [Khalid Sheikh Mohammed] did not result in the ‘discovery’ of KSM’s ‘Second Wave’ plotting.” The report’s rationale for this claim is that KSM did not provide intelligence on this Second Wave until days after interrogators stopped using EITs and only when confronted with the fact that one of those he recruited for the second wave had been arrested.
First, the fact that KSM provided information after EITs were used is an important part of the context that the Intelligence Science Board report highlights. EITs almost certainly cast a long shadow over subsequent interrogations even if they were not used in those interrogations. The fact that KSM believed further use of EITs was possible if he was caught lying cannot be ignored in evaluating the efficacy of EITs.
Second, the fact that KSM only believed he would be caught lying when the CIA demonstrated that it likely already knew at least some of the right answer is entirely in keeping with the importance of intelligence background. For example, the SSCI report cites two CIA products that argue KSM was only forthcoming with information he believed CIA already possessed. One CIA report noted “… KSM only admitted knowledge of this operation upon learning of Masran’s detention, we assess he is not telling all he knows, but rather is providing information he believes we already possess… CIA officers at the detention site where KSM was being interrogated then wrote in a cable, ‘[t]he core problem, once again, is the appearance that KSM gave up this critical information only after being presented with the idea that we might already know something about it.”
The SSCI report cites numerous other instances of CIA officers believing KSM only provided information when he believed he could not lie about it. In the middle of the period of enhanced interrogation the report notes:
On March 18, 2003, KSM was confronted with the reporting of Majid Khan, who was then in the custody of a foreign government, regarding plotting against gas stations inside the United States, information that KSM had not previously discussed. In assessing the session, DETENTION SITE BLUE [the interrogation facility] personnel noted that “KSM will selectively lie, provide partial truths, and misdirect when he believes he will not be found out and held accountable.” On the other hand, they wrote that “KSM appears more inclined to make accurate disclosures when he believes people, emails, or other source material are available to the USG for checking his responses.”
It is an open question whether KSM would have provided this information without the use of EITs. Even the CIA admits that it is possible he might have. CIA Director John Brennan noted “[t]he answer to this question is and will forever remain unknowable.” During enhanced interrogation some CIA personnel present were equally uncertain that it was necessary, as the SSCI report notes. It quotes an email between CIA personnel:
“Truthfully, though, I don’t recall that the WB [waterboard] produced anything actionable in AZ [Abu Zubaydah] any earlier than another technique might have. This may be different with KSM, but that is still as much a statement of faith as anything else – since we don’t seem to study the question as we go… it’s been many more days of constant WB repetitions, with the evidence of progress through most of them not being actionable intel but rather that ‘he looks like he’s weakening.’ The WB may actually be the best; just don’t like to base it on religion.”
Thus it is entirely possible enhanced interrogation was not necessary. But ‘not necessary’ is not the same as ‘not effective.’ The fact that KSM only provided some intelligence when he believed he could not lie about it in no way obviates the success of the interrogation- whatever the method.
The same pattern appears again and again in the SSCI report- if the CIA used other sources and methods to vet interrogation intelligence or vice versa it somehow vitiates the importance of the intelligence. In the case of the unraveling of a plot in Southeast Asia, the CIA used a variety of sources to connect various actors. From interrogations of Majid Khan, noted above, the CIA knew KSM had tasked Khan to deliver money to an individual known as Zubair. When confronted with Majid Khan’s capture in March 2003 (and after being told that as a result of the capture “… the CIA had ‘stacks and stacks of emails,’ and that CIA officers were going to do a “test of his honesty”…) KSM admitted the money was going to a major figure in the plot known by the nom de guerre Hambali. This information provided the crucial confirmation of a connection between Hambali and Zubair, even though KSM denied any knowledge of Zubair.
This connection had been previously suspected, so for the SSCI report KSM’s confirmation was unneeded and it may indeed appear that way in hindsight. Yet for CIA officers combing through a variety of sources, this confirmation was almost undoubtedly highly valuable. When CIA officers discovered that a human intelligence source they were developing was contacted by a phone number connected to Zubair in May 2003, the confirmation provided by KSM provided exactly the sort of cue needed to sort and prioritize intelligence collection.
Yet despite the successes of CIA interrogation when the intelligence background was good, there are clearly a number of instances when enhanced interrogation was used fruitlessly due to inadequate or simply incorrect intelligence background. The SSCI report highlights several instances where interrogators in the field and officers at CIA headquarters disagreed about the compliance of detainees. In these instances interrogators often believed that detainees were fully compliant while headquarters believed, with the limited intelligence background available in the months after September 11, 2001, that the detainee had to know more.
For example, in the case of Abd al-Rahim al-Nashiri, CIA interrogators quickly concluded after relatively limited application of enhanced interrogation techniques that al-Nashiri was compliant. Headquarters disagreed, and despite strong recommendations from interrogators in the field, sent new interrogators to resume enhanced interrogation of al-Nashiri, who never provided any of the intelligence sought by headquarters.
A similar pattern emerged with Ramzi bin Al-Shibh. After interrogators concluded he was cooperative, headquarters continued to insist he must know more. In response, CIA interrogators asked for an amplification of the intelligence background in order to improve their interrogation. Specifically the SSCI report notes
“[o]n February 11, 2003, interrogators asked CIA Headquarters for questions that ALEC Station was ‘85 percent certain [bin al-Shibh] will be able to answer,’ in order to verify bin al-Shibh’s level of cooperation.” While the SSCI report does not indicate whether such questions were forthcoming, the conclusion of the reply sent from ALEC Station on February 13 suggests not: “We think Binalshibh is uniquely positioned to give us much needed critical information to help us thwart large-scale attacks inside the United States, and we want to do our utmost to get it as soon as possible. Good luck.”
The foregoing is not to say Bin Al-Shibh did not provide good intelligence. Indeed the SSCI report notes he provided a great deal when in a foreign government’s custody. Unsurprisingly a CIA officer who was involved in Bin Al-Shibh’s interrogation by the foreign government noted “… the most effective interrogation tool was having information available to confront him when he tried to mislead or provide incomplete information.” Thus Bin Al-Shibh was productive when the intelligence background was good but all the enhanced interrogation in the world could not make him tell things he did not know, leading to the plaintive request for questions interrogators could ask that he could actually answer. Despite ALEC Station’s fervent hope, wishing interrogators good luck with enhanced interrogation is no substitute for an accurate intelligence background.
It is unfortunate that the SSCI report strove so mightily for irrational consistency. If the report had focused less on making a case that ‘torture does not work’ and instead focused on the limits and scope conditions enabling interrogation (enhanced or not) it could have demonstrated that EITs are no panacea. This conclusion, eminently demonstrable with facts even the CIA does not seem to dispute, could then have led to a clear assessment of the costs and benefits of such methods.
Moreover the SSCI report could nonetheless have made the case EITs are immoral even if the benefits outweigh the costs. This would have been a courageous stand, but one that would have involved a significant value trade-off. All good things do not go together and by arguing that they do the SSCI report confuses and polarizes an important area of policy. Policymakers and scholars alike must confront hard choices about harsh measures; it is best that they do so with the clearest possible view of those measures.
Since September 11, 2001, the United States has carried out a policy of torture in order to achieve strategic objectives in the war on terror. This policy has subjected prisoners to physical and mental abuse long prohibited in international treaties and recognized as torture when used against U.S. military personnel. This policy involved both formal and regulated abuse, as well as informal and unregulated abuse, which are related to each other, and therefore need to be considered jointly.
The “enhanced interrogation techniques” (EITs), the formal and regulated aspects of the U.S. torture policy, were officially condoned in legal memoranda prepared by the U.S. Justice Department in 2002 for use by U.S. military and CIA personnel. Rather than remaining a limited and tightly regulated program, multiple reports by the U.S. government and human rights organizations reveal that prisoners were subjected to informal and unregulated abusive acts throughout the U.S. detention program, at facilities in Guantanamo Bay in Cuba, Bagram in Afghanistan, Abu Ghraib in Iraq, and numerous black sites in undisclosed locations.
In light of the prolific use of these methods, the recent Senate Select Committee on Intelligence (SSCI) report, which focuses on the formally-authorized use of EITs on 39 individuals, is alarmingly narrow. Indeed, since 2002, at least 86 detainees died in U.S. custody, with 27 of these cases criminal homicides by the U.S. government’s own admission, and estimates of the total number of people subjected to torture range from many hundreds to thousands. Rendition, the policy of remanding detainees to third-party countries for presumably harsh interrogations, is also not considered in the SSCI report.
Torture is widely considered as immoral, with the most common exception of a ‘ticking time-bomb’ scenario, where the abuse of one person with knowledge of an imminent attack is justified as the only means of preventing harm to others. Although many American officials have referred to this scenario to justify their actions, the use of U.S. torture in the war on terror goes well beyond situations of a ticking time-bomb. The August 1, 2002 U.S. Justice Department memorandum advising the President, the CIA, and the Defense Department on “Standards of Conduct for Interrogation” explicitly refused to limit the use of enhanced interrogation techniques to situations of definite knowledge that “an attack is immediately about to occur” or any other considerations of “timing” of plots. Instead, prisoners “may be hurt in an interrogation because they are part of the mechanism that has set the attack in motion” and whenever it is “believed necessary” to “avoid a greater” harm, a belief that need not be reasonable so long those carrying out the abuse “honestly believe … the use of force was necessary.”  The use of torture only requires that the individual is part of a terrorist group, although what counts as membership in these murky networks is left undefined.
The main reason American policy makers are attracted to torture is the presumption that it is effective, where the harm to one or a few is justified by the prevention of harm to many more. However, what is at issue is not the rare use of torture under short and exceptional circumstances where there is reasonable knowledge that an individual has information about an on-going plot that is likely almost immediately to kill vast numbers of people. What is at issue is a program of regulated and unregulated torture involving the consistent abuse of hundreds to thousands of people over prolonged periods of time when there is little, if any, hard evidence that imminent harm is about to occur. The key question is whether the overall program of torture is effective as a weapon in the war on terror.
This article assesses the strategic value of U.S. torture and finds that the highly limited tactical benefits are easily outweighed by the heavy strategic costs. Thus far, the question of the effectiveness of the U.S. torture program has been mainly addressed narrowly, in terms of tactical benefits, with the result that only half of the equation has been considered. A full assessment of the value of a systematic policy of torture requires balancing tactical benefits with strategic costs, captured in a simple equation:
Value of Torture = Tactical Benefits – Strategic Costs
Today’s policy makers, journalists and scholars have tended to ask whether torture is tactically effective, namely whether abusive interrogation tactics extract reliable information that could not have been obtained through other means. Although uncertainty exists, the currently available evidence suggests that the tactical benefits of torture have been low. Even those with access to classified intelligence find no clear evidence that torture led to information that directly thwarted specific plots and saved lives in a way that non-coercive interrogation could not. In contrast, there are good reasons to believe that the strategic costs of torture on national security have been high. Evidence suggests that U.S. torture has fueled recruitment by violent Islamist groups, produced attacks on American citizens, and strengthened domestic opposition to American military policies, including the war on terror itself.
Reviewing the recently released Senate report and CIA reply, the evidence for the tactical effectiveness of torture in extracting reliable information from prisoners is, at best, weak. The heart of the report is a detailed assessment of twenty “best” cases for torture, those cited publicly by the Bush administration as evidence that torture, including waterboarding and other “enhanced interrogation techniques,” produced crucial information that disrupted dangerous plots and led the U.S. to Osama Bin Laden.
The Senate Committee report charges that U.S. torture is “not an effective means of acquiring intelligence or gaining cooperation from detainees” and that non-coercive interrogation techniques are more effective at obtaining reliable information. The Senate report presents detailed time-lines of the interrogations related to these twenty cases, using the CIA’s own classified records of these events, and finds no smoking-gun evidence of a clear-cut tactical success – that is, no evidence that torture prevented a ticking time-bomb situation – and that it was non-coercive interrogation that produced the crucial information in every case where reliable information produced at all.
In its response to the report, the CIA argues that torture generated unique information that, when added to the overall mosaic of other information, significantly advanced U.S. knowledge of anti-American terrorist networks and plots. Consequently, the CIA asserts that the U.S. detention and interrogation program as a whole was vital to American interests and saved lives by producing “unique intelligence” that disrupted terrorist plots and captured terrorists. However, the CIA does not take a clear position on whether coercive techniques played a critical role in these results. The crucial problem is one of inference, justifying the argument that information could not have been obtained through other, non-coercive, means. To this effect, CIA Director John Brennan stated that “the cause-and-effect relationship between the use of EITs and useful information subsequently provided by the detainee is, in my view, unknowable.”
As evidence of these distinct perspectives (and the surprisingly limited range of debate), consider the treatment of the Karachi plot in both the SSCI report and the CIA response. The Karachi plot entailed a plan to attack the U.S. Consulate in Karachi with suicide attack vehicles including cars and motorcycles. From 2003 to 2009, the CIA repeatedly represented the thwarting of this plot as evidence of the efficacy of EITs. President Bush also cited the Karachi plot in a 2006 speech as evidence that EITs “helped stop a planned attack on the U.S. embassy in Karachi.” The SSCI report argues that this represents a clear misrepresentation, since the capture of the operatives, the seizure of the explosives, as well as the details of the planned attack were all accomplished prior to the use of EITs on the plotters. Indeed, the Senate report contends that these plotters were apprehended by the unilateral action of Pakistan, rather than the CIA. In its response, the Agency “acknowledges that… we mischaracterized the impact of the reporting we acquired from detainees on the Karachi plots.” However, the CIA contends that it acquired new information on potential future plots through the use EITs on the detained plotters that it used to recommend added security measures. But even with this new intelligence, the CIA admits that “we have no information specifically indicating whether the additional Karachi plotting was disrupted by those measures, by Pakistan’s detention of Ammar [al-Baluchi], Khallad [bin Attash], and other extremists, or by other unknown factors.” In short, the CIA could not defend its previous claim that EITs helped stop a planned attack. It could only claim that EITs may have produced some unique, actionable information.
Hence, the range of debate among those with access to the classified records is that either torture completely failed on its own terms or that it helped on the margins. The effectiveness of torture, of course, has also been analyzed by scholars, who identify myriad reasons why torture is unlikely to produce accurate, actionable intelligence. In the end, however, the CIA’s own bottom line is perhaps the most telling conclusion to be drawn on the tactical effectiveness of the use of torture in the war on terror. Although the report defends the use of torture as important for national security, it also states that the use of enhanced interrogation techniques has ended and should not resume. CIA Director John Brennan writes, “while I am the Director of the CIA, this program will not under any circumstances be reinitiated…. enhanced interrogation techniques are not an appropriate method to obtain intelligence and their use impairs our ability to continue to play a leadership role in the world.” In other words, according to the CIA reply, the value of the evidence produced by torture is not so great that the United States should continue the practice even though it faces an ongoing terrorist threat.
This short analysis suggests that the tactical benefits of torture are exceedingly limited. Furthermore, these tactical benefits must then be weighed against the strategic costs of torture practices. Since the tactical benefits of torture are so low, even moderate strategic costs are likely to off-set them. The remainder of this paper will demonstrate that these strategic costs not only exist, but are often quite heavy. Any sound assessment of torture must take this factor into account.
Assessing the overall value of torture by looking only at tactical effectiveness – i.e., the relative worth of different ways to interrogate a prisoner – is not enough. As the CIA’s own reply suggests, the use of torture can have strategic costs that reduce the overall national security of the United States. These costs should loom particularly large, if, as all evidence suggest, the tactical benefits are slight.
How does torture produce strategic costs? Any brutality in war and other settings that harm individuals can have strategic costs. Indiscriminate punishment of civilians is known to produce spirals of anger and hate among attacked populations. As a result, professional militaries since World War II have sought to impose harm on civilians selectively, indirectly, or not at all. What is special about torture, however, is that it is a form of selective punishment that produces intense strategic costs which are similar in kind to indiscriminate punishment. The reason is that torture easily triggers a sense of humiliation, more directly assaulting the human dignity of individuals than many other forms of brutality, and so can create wider reactions in both potentially hostile and even friendly publics.
Specifically, torture causes blowback from two audiences: the terrorists and the communities in which they operate, and the publics of the torturing state. Humiliation extends well beyond those who are tortured to include the victim’s community and fosters radicalizations, increasing mobilization by militant groups and raising the level of violence. At the same time, the humiliation of torture divides the public of the torturing state about the hypocrisy of their actions. This reaction is especially salient when that state is a liberal democracy like the United States where military strategy is vulnerable to shifts in popular support.
Social psychologists have long explained the logic of what to expect from humiliation. Individuals subjected to this form of controlled abuse are exposed to more than physical and psychological pain; they are dominated, in a position of helplessness, and treated with contempt. Put simply, torture means the person “does not count.”At its core, humiliation is about the powerful denying equal status to the weak to have control over the most basic elements of their lives. As such, the sense of humiliation is associated with feelings of being unjustly degraded, ridiculed to the point that one’s identity has been demeaned or devalued. Since loss of status requires witnesses, humiliation is a function of publicity, and the more publicity leads to greater humiliation:
The status to claim the enjoyment of basic human rights is, by its nature, a core, fundamental status. A humiliation that involves public demonstration of an individual’s inability to claim such rights has a devastating effect. For example, individuals in wartime who are forced by their captors, against all of their cultural and religious beliefs, to strip naked, to be led around on all fours on a leash, or to witness the rape of their own wives and daughters, would be examples of persons humiliated in this way. These situations constitute de facto “degradation ceremonies” in which the message to the degraded individual is: “If there is something that you most want to resist, it is this. Yet, look at you. Look how utterly powerless you are to claim what is most essential to you.”
Further, humiliation can spread beyond those abused to others in two ways. First, acts of even regulated humiliation can spread into unregulated humiliation involving new perpetrators not associated with the original acts because they come to believe that these acts of humiliation are permissible. In a famous experiment by Stanley Milgram:
Stark authority was pitted against the subjects’ strongest moral imperatives against hurting others, and, with the subjects’ ears ringing with the screams of the victims, authority won more often than not. The extreme willingness of adults to go to almost any lengths on the command of an authority constitutes the chief finding of the study and the fact most urgently demanding explanation.
Second, the sense of humiliation can spread among other, particularly kindred, populations, who think that one society that subjects members of another society to torture is making a statement about the worth of all in that kindred society. The very fact that a society chooses to torture members of another society but not its own is often seen as decisive evidence of unjust degradation and contempt.
The response to humiliation varies, but is often intense. Whether experienced by individuals or groups, humiliation can compel risk-taking even to the point of death, either for revenge, to reassert lost status, or simply to escape the situation. Not surprisingly, humiliation plays an important role in the escalation of inter-individual and inter-group conflict and can also lead those not immediately involved to distance themselves from the situation.
In fact, the core logic of the program of enhanced interrogation techniques was built on the idea of humiliation, or “learned helplessness,” as the two architects of the CIA’s torture program, the contractor psychologists Bruce Jessen and James Mitchell, called it. Learned helplessness means to break down an individual’s self-control, until he or she is emotionally and psychologically unequipped to disobey.”
Based on this logic of humiliation, we can identify three specific strategic costs of torture in the war on terrorism.
First, a policy of enhanced interrogation creates a systematic culture that is permissive of abuse. The debate about the efficacy of torture has focused narrowly on 39 detainees who were subject to EITs. It is no surprise then that many have found little support for strategic consequences stemming from the use of torture in just this handful of cases. However, the officially sanctioned policy of torture directly resulted in systematic abuse of prisoners on a much larger scale and these strategic costs must be taken into account. While American policy makers have blamed the highly-publicized prisoner abuse scandal at Abu Ghraib on the actions of a few ‘bad apples,’ such actions were both systematic and sanctioned by higher-ups, in the case of Abu Ghraib and in detention facilities elsewhere. The explicit approval for torture by high-level figures in the administration, and the subsequent adoption of these measures by subordinates, is reminiscent of the findings of the infamous Milgram experiment. Even tacit approval by authority figures significantly increases the willingness of subordinates to impose life-threatening pain on other.
Even though some would like to isolate the effect of enhanced interrogation techniques from those of the abuses at Abu Ghraib, this would be a mistake. The two are intimately linked, both in the sense that the formal regulated program encouraged the informal, unregulated subsequent abuse, and in the minds of majorities in both the United States and key Middle Eastern countries. These audiences viewed the abuse at Abu Ghraib and other parts of the U.S. program on torture as being essentially the same. In summer 2004, 51% of the U.S. public believed that Abu Ghraib abuses had been authorized by higher authorities (a sizeable 12% refused to answer), while an early opinion poll of over 1,000 individuals from six large cities in Iraq found that 54% of Iraqis believed “all Americans” were responsible.
Second, a systematic program of torture can harm national security by mobilizing more recruits for terrorist groups and inspiring more attacks. Americans may conclude from watching the thousands who have turned out to shout ‘Death to America’ in protests across the Muslim world, and the hundreds who have committed acts of terrorism against the United States and its allies, that all Islamic fundamentalists already hate the United States so much that it could not provoke them further no matter what it did. This would be a tragic miscalculation. While it is true that there are millions who hold strong anti-American sentiments, so far only a tiny number of them have become terrorists and fewer still have carried out attacks. If American action appears to confirm the worst accusations of American arrogance, the United States will help extremists recruit a new generation of willing terrorists, one that will be far larger than the last. Leaders of anti-American terrorist groups already base their recruitment appeals on claims of oppression, injustice, and humiliation. A program subjecting dozens, much less hundreds to thousands, to torture is likely to become powerful evidence that the terrorist leaders’ claims are valid, widening the appeal of these anti-American groups beyond what would otherwise occur.
How exactly does U.S. torture facilitate this shift in relative support in favor of terrorist organizations? When incidents of U.S. torture become publically known, terrorist organizations have made great efforts to accentuate the repulsive nature of these events in their discourse to followers and potential sympathizers. Indeed, there is voluminous evidence that U.S. torture plays a central role in videos, statements, and discourse by Al Qaeda Central and other terrorist groups (in Iraq and beyond) who have the intent of recruiting individuals to carry out attacks against Americans or their allies, and also nurture a general hostility towards the United States.
Torture as a strategy is designed to humiliate the victim including fostering a sense of helplessness and a total loss of self-worth. Furthermore, torture is a situation in which the weak actor is dominated by the strong actor, with this strong actor denying both equality and dignity to the weak actor. When the designation of torturer and tortured are drawn along identity boundaries, such as ethno-religious distinctions, the public revelation of torture practices will lead the kindred population of the victims of torture to presume that the torturing state’s society views that entire population with contempt and seeks their degradation. Put more simply, torture reinforces beliefs among a victim’s community that the torturing state presents an existential threat not just to detainees but to the entire community.
The heightened multimedia capabilities of terrorist organizations, combined with the rocketing consumption of online audio-visual material, has enabled the enemies of the United States to propagate humiliation and fear in order to mobilize recruits and material support within and far outside Iraq and the Middle East. The internet has allowed terrorist organizations to broadcast their discourse to a far wider audience. Following the abuses in Abu Ghraib and other incidents of torture, this discourse has demonized the United States as a violator of human rights, particularly against Muslims. Once information regarding – and visual evidence of – U.S. torture enters the public domain (via leaks or official disclosures), terrorist organizations rapidly and repeatedly allude to these practices to discredit and cultivate resentment towards the United States. It is these repeated invocations that perpetuate the emotional effects of the original incident – allowing the terrorists to strategically benefit from revelations of U.S. torture practices for an extended duration.
A systematic program of torture is particularly likely to increase future attacks. While the motivation for, and principle justification of, the U.S. policy of torture was to save American lives, it likely has had the opposite effect.
U.S. torture can lead to an increase in insurgent violence as a direct consequence of the policy’s effect on recruitment. Leaders of anti-American terrorist groups already base their recruitment appeals on claims of oppression, injustice, and humiliation. Torture is particularly likely to increase future attacks because terrorist leaders can easily use them to demonize American goals and behavior as barbaric violations of human dignity and also because revenge for past wrongs to family and friends is a potent source of violence in the traditional societies in the Middle East most associated with terrorist groups. Evidence from martyr videos, the last wills and testaments of suicide bombers, confirms the centrality of humiliation and revenge to the insurgents’ strategy for legitimating their campaign of violence as well as justifying specific attacks. To the extent that outrage over U.S. torture swells the enemy’s ranks with additional fighters and their coffers with additional resources, it improves their ability to mount larger and more frequent attacks.
Torture can increase the acceptability of violence by the insurgent group against the torturing state, its soldiers, and society. Insurgent groups depend on local support for their ability to mount and sustain their campaigns against the occupier and its allies. Because torture is universally condemned as barbarous, a policy of torture allows insurgent groups to draw a moral equivalence between their violent means and those of their enemy, an equivalence that will be readily interpretable by the group’s constituents and reinforced by the group’s propaganda. Groups draw on this logic to add moral justifications to attacks against civilians and the use of suicide attacks and beheadings, attacks which might otherwise be more likely to alienate their supporters.
Torture can also lead to increased attacks by lowering the ability of the United States to prevent attacks. A key goal of any counterinsurgency (COIN) strategy is obtaining intelligence on insurgent capabilities and activities from the local population critical to preventing attacks and defeating the insurgent group. The pool of potential voluntary informants in the local population is vastly greater than the number of insurgents or potential insurgents the U.S. and its allies can capture and interrogate. Convincing an occupied public to share information with the U.S. is a significant challenge even without engaging in the kinds of policies, like torture, that undermine the credibility of strategies premised on winning the ‘hearts and minds’ of the occupied public. The humiliation and ill will generated by a systematic policy of torture targeting their countrymen and coreligionists undermines efforts by the U.S. to collect valuable intelligence about the insurgency and hinders its ability to prevent attacks, thereby putting U.S. servicemen fighting in Afghanistan and Iraq at risk.
Third, the use of torture can reduce support for the war on terror from liberal segments of otherwise sympathetic publics. Indeed, it can do so even when there is evidence that it contributed to tactical successes. The most famous tactical success of torture is probably French army victory in Battle of Algiers in 1957. In this case, historians widely credit the torture of militants as playing a key role in defeating the terrorist network in the city. More important, however, is that this victory did not stop the resistance from continuing the fight in other areas of the country. Further, the use of torture shocked the French public. Large segments quickly distanced themselves from the war, and, by 1962, 90% voted to abandon the war altogether. The most famous tactical success of torture in history led directly to strategic failure. One can win a battle and lose the war.
Similarly, the U.S. policy of torture, resulting in the prisoner abuse scandal at Abu Ghraib, shocked American voters, especially ‘liberal hawks’ who supported the war for its ideals of democratization. Human rights activists, liberal public intellectuals, academics, and at least one Nobel Peace Prize winner – publicly supported the war on terror on humanitarian grounds. Of course, if the moral case matters to gain public support, then we should expect that policies that directly contradict the moral case would undermine public support. Torture flies squarely in the face of moral justifications for America’s war on terror. No surprise that in early 2004, a majority of the American public, reflecting a large portions of the Democratic Party and ‘liberal’ America who were likely influenced by this moral discourse, withdrew their support for the war.
One might object that, had the prisoner abuse at Abu Ghraib not been exposed with lurid photos, the strategic impact of the U.S. policy on torture might have been significantly less. After all, the program was shrouded in secrecy precisely to prevent the kinds of blowback that would have undermined the broader goals of the war on terror. However, it is highly unlikely that the strategic costs can be avoided through secrecy, mainly because, as a democracy, the United States is highly unlikely to prevent information about its torture program from becoming public. And, as the Abu Ghraib scandal shows, once public, the nature of the acts is likely to ensure widespread media coverage. Thus while the sections in the essay below catalog strategic costs in light of the scandal, any evaluation of the value of torture must take the anticipation of such revelation and the resulting costs into account.
In sum, there is evidence that U.S. torture: 1) created a pervasive culture of abuse, the condition of possibility for scandals like Abu Ghraib; 2) contributed to Al Qaida propaganda and recruitment, and contributed to further attacks; and 3) weakened American support for the war on terror, particularly the Iraq war. Although a full evaluation of the strategic effectiveness of torture in the war on terror may not be possible without access to U.S. classified records, and a precise determination of the level of the strategic costs cannot be made, the evidence shows a consistent set of negative consequences related to future attacks. Given that the tactical benefits of torture are low, this evidence of the strategic costs suggests that torture in the war on terror has done more harm than benefit to U.S. national security.
The first cost of torture is that it created a pervasive culture of abuse. The policy of enhanced interrogation encouraged the systematic abuse of detainees in order to ‘soften them up.’ Declassified military reports explicitly indicate that the abuse at Abu Ghraib was encouraged in order to support interrogations. The report found that:
“twenty-seven 205 MI BDE [205th Military Intelligence Brigade] Personnel allegedly requested, encouraged, condoned or solicited Military Police (MP) personnel to abuse detainees and/or participated in detainee abuse and/or violated established interrogation procedures and applicable laws and regulations during interrogation operations at Abu Ghraib”
In many cases, this abuse amounted to torture. The function of the infamous ‘torture memos’ was to justify that the Geneva convention did not apply to any of the detainees from the Afghanistan war, removing criminal liability for any abuse suffered while they were in American prisons. Soldiers were routinely assured that the actions that they were engaging in were officially sanctioned and legal.
Permissiveness for torture came from the highest levels and there is widespread evidence that decision makers and commanders either encouraged, or at least ignored, widespread abuse of prisoners. Secretary of Defense Donald Rumsfeld specifically authorized ‘counter-resistance techniques’ and even suggested harsher treatment, adding the note “I stand for 8-10 hours a day. Why is standing limited to four hours?” The Senate report documents that CIA Headquarters pushed to continue torturing detainees even after interrogators determined that the detainees were cooperating. In the case of Abd al-Rahim al-Nashiri, CIA headquarters went so far as to remove interrogators who defended the compliance of the subject and to replace them with unqualified personnel who went on to carry out interrogation techniques that were explicitly prohibited, with the knowledge and consent of the Chief of Base. These actions were justified on the basis of an “imminent…9/11-style attacks,” for which no support has been provided and for which al-Nashiri did not provide any information. Given the actions of those in charge, it is no surprise that prisoner abuse was widespread and systematic. The fact that the soldiers at Abu Ghraib took photographs with the detainees demonstrates that they felt that the command structure condoned, if not encouraged, such behavior. These photographs were openly shared and discussed at Abu Ghraib, in full view of commanding officers, and reports even indicate that those horrific photographs were used as screen savers on military computers.
The prisoner abuse that has been blamed on just a few soldiers has been a regular part of detention in American facilities. Abuses similar to those at Abu Ghraib were reported in other American detention facilities as early as 2002. Human Rights Watch documents detainees being beaten, subject to prolonged sleep and sensory deprivation, and being held naked at American facilities at Bagram Airbase in Afghanistan, cases not part of the formal detention and interrogation program and therefore not included in the 39 incidents analyzed in the SSCI report. Amnesty International and the International Committee for the Red Cross (ICRC) report similar systematic prisoner abuse at Guantanamo Bay as well as other facilities in Afghanistan and Iraq, and the extradition of hundreds of suspects to countries known to perform torture is well-known. As previously mentioned, the death of 86 detainees in U.S. custody since 2002 illustrates the extent to which the harsh treatment of prisoners goes far beyond the 39 cases admitted by the Senate report. In the end, Human Rights Watch concludes that many hundreds to thousands of detainees were subject to torture in American detention facilities.Abu Ghraib was unique not because of the abuse but because photographs of the abuse became public.
The second cost of torture is increased recruitment to terrorist groups and attacks against the United States and allies.
While there are no precise measures of the degree to which Abu Ghraib and Guantanamo feature in jihadist recruitment propaganda, it is clear that in the most salient discourses – including calls to Jihad, martyr videos, and beheadings – invocations of torture feature prominently. By supplying a ready-made symbol for jihadists, the U.S. policy of torture hurts the U.S. more than it helps. Moreover, the effects persist long after the scandal. As General David Petraeus argues: “whenever we have, perhaps, taken expedient measures, they have turned around and bitten us in the backside…Abu Ghraib and other situations like that are nonbiodegradables. They don’t go away. The enemy continues to beat you with them like a stick.”
Thérèse Postal, writing for The Atlantic, traces how jihadists from al Qaeda Central (AQC) to the Afghan Taliban have used U.S. detention and interrogation in their propaganda, focusing specifically on Guantanamo. She demonstrates that Guantanamo is a potent symbol in jihadist rhetoric, largely inseparable from U.S. torture policy and the abuses at Abu Ghraib. Osama bin Laden, writing in the first issue of al Qaeda in the Arabian Peninsula’s (AQAP’s) Inspire magazine (2010), listed “the crimes at Abu Ghraib and Guantanamo, those ugly crimes which shook the conscience of humanity” among the actions by the U.S. and its allies that justified jihad. Subsequent issues repeated this theme. Issue 2 includes essays by two former detainees calling for new recruits to Jihad. In Issue 7, an essay explains that Guantanamo Bay “exposed the West for what it really is” and “showed the world the American understanding of human rights.” A 2013 issue includes an essay on Guantanamo detainees, many of whom the author claims are innocent. As Postel notes, given the large number Yemenis at Guantanamo, “[t]his is a very salient message for AQAP’s base.”
Images of torture at Abu Ghraib have been used by organization such as AQC in their propaganda videos, designed to mobilize support for the group and create a culture of martyrdom legitimizing the tactic of suicide attack. For instance, one video released in 2004 “flashes a picture of an Iraqi detainee, shrouded in a black hood, with electrodes attached to his naked body as a voice drones: ‘Abu Ghraib prison. Whenever I remember our free brothers kept inside the crusaders’ prisons … I feel the earth turning and I promise God to seek vengeance on every hand which participated in performing the chapters of this conspiracy.’” In another example, Abu Musab Zarqawi, the leader of al Qaida in Iraq (AQI), uses the video beheading of American Nicholas Berg as an opportunity to call on Muslims everywhere to avenge the humiliation of Abu Ghraib:
“Is there any excuse left to sit idly by? How can a free Muslim sleep soundly while Islam is being slaughtered, its honor bleeding and the images of shame in the news of the satanic abuse of the Muslim men and women in the prison of Abu-Ghraib. Where is your zeal and where is the anger for the religion of Allah? And where is the jealousy over the honor of the Muslims and where is the revenge for the honor of the Muslim men and women in the prisons of the Crusaders?” 
The torture and consequent humiliation of Muslim prisoners by their non-Muslim dominators is presented as an assault on the right to dignity of Muslim society requiring Muslims to seek empowerment through joining or aiding terrorist organizations.
Torture also serves as a central motivation driving recruitment to terrorist groups. Although we cannot measure the precise impact of the U.S. policy on torture on recruitment to insurgent groups in Iraq and elsewhere, we do know that the increased flow of foreigners to fight against the U.S. and its allies in Iraq was substantially influenced by Guantanamo Bay and Abu Ghraib prisons. Foreign fighters matter because they augment the military capabilities of the insurgent groups they join and pose a threat to their home states when they return. While foreign fighters represent only a fraction of the overall number of militants filling the ranks of anti-American insurgent groups, the success with which groups are able to recruit them serves as a lower bound on torture’s effect on recruitment. Its effect on recruitment of local fighters is likely to be even greater.
There are no precise figures on the number of foreign fighters coming into Iraq in the first years of the insurgency, though estimates by the U.S. military show an increase from 500 in January 2004 to 3,000 by November, though the November total was later lowered to 1,000. The number of foreigners held in detention also increased, from 90 in July to 140 by September 2004. The best estimates of the flow of foreign fighters into Iraq come from AQI records captured by the U.S. military in Sinjar, Iraq, in October 2007. Analysis of those documents, which only capture a part of AQI’s overall operations in Iraq, revealed that as many as 40 fighters were arriving every month. Other estimates put the total as high as 90 foreigners entering Iraq illegally per month in 2007.
It is widely accepted that the abuse of prisoners at Abu Ghraib and Guantanamo contributed significantly to the motivations of these fighters. In a 2013 interview with the journal Foreign Affairs, Stanley McChrystal, who served as the commanding General of U.S. Joint Special Operations Command in Iraq from 2003 to 2006, stated that “there were hundreds of foreign fighters that came in [to Iraq] because they were responding to Abu Ghraib.” U.S. interrogators, those with direct access and knowledge about detainee motivations, confirm the centrality of Abu Ghraib as motivation for joining the Iraqi insurgency. According Matthew Alexander (pseudonym), the lead interrogator on the team that uncovered Zarqawi’s location in Iraq in 2006: “I learned in Iraq that the No. 1 reason foreign fighters flocked there to fight were the abuses carried out at Abu Ghraib and Guantanamo. Our policy of torture was directly and swiftly recruiting fighters for al-Qaeda in Iraq.”
The Saudi Arabian government estimated that, as of mid-200, 352 Saudis had crossed into Iraq to join the insurgency. In addition, 63 would-be recruits were captured by Saudi security forces and interrogated. When asked about their motive for joining, “[t]he catalyst most often cited was Abu Ghraib, though images from Guantanamo Bay were mentioned.” Importantly, less than 85% of those questioned had previous associations with militancy, suggesting that the occupation of Iraq and the torture and humiliation of Arabs associated with it after Abu Ghraib was central to their willingness to fight. This recruitment effect was not reserved to Iraq. It was also found in Afghanistan. A leaked cable detailing discussion between U.S. Ambassador Richard Holbrook and officials in Saudi Arabia reveals that the Saudis arrested 250 would-be jihadists in the immediate aftermath of the Abu Ghraib torture revelations. The Saudis specifically claimed that the individuals were inspired to fight in Afghanistan by images of torture from Abu Ghraib.
Finally, given the permissive culture of torture that gave rise to prisoner abuse in Iraq, imprisonment is even more likely to have radicalized Iraqis as well. American incarceration policies in the early years of the post 9/11 war in Iraq created the perfect conditions for this process of radicalization via incarceration. As the Red Cross reported, prisoner abuses – most of which are easily defined as torture – were regular practices throughout the U.S. militarily run prison system in Iraq, not just Abu Ghraib. Those detained in Iraq by the U.S. during this period were thus likely to have been subjected to mistreatment. According to a Human Rights Watch report released in June 2004, in the fourteen months since the invasion “more than 12,000 Iraqis have been taken into custody by U.S. forces and detained for weeks or months.” This becomes very disconcerting when combined with another message in the report: an estimated “70 percent to 90 percent of Iraqi detainees were arrested by mistake…” In other words, a majority of those detained were “arrested by mistake” but were still either exposed to U.S. torture themselves or likely to witness the humiliation and torture of others. It is hard to imagine that some fraction of these did not join the insurgents upon their release.
One of the most notable examples of radicalization via incarceration cited by many is Abu Bakr al-Baghdadi, the leader of the Islamic State of Iraq and Syria (ISIS). While Baghdadi had some involvement in terrorist organizations prior to his incarceration at Camp Bucca, his time in the camp is acknowledged as a pivotal moment that deepened his involvement and steeled his conviction. It is doubtful that Baghdadi would have climbed to the leadership of ISIS without this incarceration experience. In short, the practice of largely indiscriminate incarceration, combined with abusive treatment while detained, has created additional terrorists, including some within the upper echelons of leadership.
The U.S. use of torture also increased terrorist attacks in Iraq and elsewhere. Demonstrating the independent effect of the U.S. policy of torture or the revelations of prisoner abuse at Abu Ghraib on insurgent violence is difficult for a number of reasons. In Iraq, where we would expect the effect to be most visible, insurgent violence was driven by multiple factors and grievances stemming from the fact of foreign military occupation and, for the Sunnis, the inevitability of a Shia-dominated government made possible by the U.S. overthrow of Iraqi President Saddam Hussein. Moreover, the Abu Ghraib scandal broke relatively early in the history of the occupation, at a time of escalating U.S. military engagement with an already growing insurgency. Finally, the benefits to recruitment are unlikely to be immediate. As a result, the extent to which Abu Ghraib contributed to the violence in Iraq in the months following the revelations, or over the course of the insurgency, will likely remain unclear. What we do know is that the insurgency in Iraq intensified significantly in the second half of 2004. And while there are likely many overlapping causes, torture was one.
Specifically there is strong circumstantial evidence – direct testimony from both attackers and U.S. interrogators – that link U.S. torture to additional attacks. Although we cannot say the precise number of attacks that were caused independently by U.S. torture, it is beyond a reasonable doubt that at least some additional attacks occurred, most likely many.
The Chicago Project on Security and Terrorism (CPOST) maintains a public database of all suicide attacks in Iraq around the world since the early 1980s. The database has full text documents used for verifying the attacks that are computer searchable. Although covering only a portion of all terrorist events, the data does cover the most deadly forms of terrorism and allows for a consistent methodology in tracking attacks and motives over time. Overall, the trajectory of suicide attacks in Iraq escalated rapidly following the Abu Ghraib scandal in spring 2004, from a yearly total of 34 in 2003 to 121 in 2004 and 293 in 2005 – all before the February 22, 2006 attack on the al-Askari Mosque in Samarra, commonly noted as the trigger for subsequent escalation.
The data reveals that the escalation begins in June 2004, one month after the publication of the Abu Ghraib photos. From January to May there were a total of 29 suicide attacks in Iraq, while from June to December there were 107, an increase from an average of 6 to 15 attacks per month. When only attacks on U.S. military and diplomatic targets in Iraq are included, the increase is more dramatic: from an average of 1.5 attacks per month through May to 6 attacks per month thereafter. In September 2014, insurgents attacked Abu Ghraib prison for the first time using suicide attacks, attacking it twice that same month.
|Figure 1. Monthly Suicide Attacks in Iraq, 2004-2005|
|Source: CPOST Suicide Attack Database, January 21, 2015 Release|
What evidence links the increase in violence in Iraq to Abu Ghraib and U.S. torture? First, Abu Ghraib features in the explicit justification of attacks. In a martyr video released by Zarqawi’s group in July of 2004, a Saudi suicide bomber proclaims his attack to be avenging the torture, rape and death of a woman named “Fatima” held at Abu Ghraib prison. In the video, Fatima relates the horrors and humiliations visited upon Iraqi women at the prison in a letter. These horrors are so that that she calls on her rescuers to “Kill us with them so that we might be at peace.” Responding to her call, the bomber points to the explosives in his vehicle and refers to them as “Fatima’s dowry” before driving off presumably to his target. The video, dated December 14, 2004, corresponds to a suicide attack on the Green Zone in Baghdad, headquarters for the U.S. occupation as well as the transitional Iraqi government.
On September 15, 2006, Abi-Abdallah al-Yemeni carried out a suicide attack with three other attackers against and Oil Terminal in Yemen, stating in martyr video:
“Do we really consider ourselves men when our sisters in Abu Ghraib prison are being raped? Don’t we have any sense of gallantry – even when there are many more sisters who are imprisoned in Arab countries which pretend to be Islamic? By God, their freedom is our responsibility and if you want to free them you have to act like Khawla [famous Muslim warrior].”
On May 5, 2007, Abu Abdul Rahman carried out a suicide attack against a U.S. Army Center at Abu Ghraib, killing 15 and declaring in his martyr video:
“We are coming to slaughter you, because we have never forgotten that you exchanged the law of God for the law of the dissolute slaves. We have not forgotten that you oppressed our brethren, sent them to prisons, and killed them.”
Overall, a search of the CPOST database records for “torture” and “Abu Ghraib” found 26 suicide attacks with direct evidence that the attacker was motivated by U.S. torture at Abu Ghraib.
Second, U.S. interrogators at Abu Ghraib prison – who had direct knowledge of the motivations of terrorists on the ground in Iraq – contend that U.S. torture was a top motive inspiring those engaged in future terrorist attacks in Iraq. As an Air Force senior interrogator who was in Iraq in 2006 explained:
“It’s no exaggeration to say that at least half of our losses and casualties in that country have come at the hands of foreigners who joined the fray because of our program of detainee abuse. …And this is not my opinion – these statistics are tracked and briefed to every interrogator who arrives there that torture and abuse was Al-Qaeda’s number one recruiting tool.”
Third, not only did suicide attacks increase dramatically after Abu Ghraib, but the months after the scandal also saw the first beheadings of captured foreigners by Iraqi jihadist groups. Televised beheadings, extremely barbaric acts, were a tactic used to grim effect most famously by Zarqawi’s AQI and continue to be used by ISIS, AQI’s successor organization in Iraq and Syria. In the speech Zarqawi gave before beheading American Nick Berg in May 2004, he justified the act as revenge for Abu Ghraib and calls for more attacks:
We say to you, the dignity of the Muslim men and women in the prison of Abu Ghraib and others will be redeemed by blood and souls. You will see nothing from us except corpse after corpse and casket after casket of those slaughtered in this fashion….So kill the infidels wherever you see them, take them, sanction them, and await them in every place.”
But the speech accompanying the brutal murder of Nick Berg was not the only element of the video that linked back to Abu Ghraib. The insurgents had dressed Mr. Berg in a bright orange jumpsuit that was instantly recognizable as the same kind of clothing worn by detainees in the photos from Abu Ghraib and the Guantanamo prison and that had become an iconic symbol of the U.S. program of torture. AQI and its successor organizations clothed their victims the iconic orange jumpsuits in nearly all their subsequent high-profile beheadings. According to one former detainee: “[ISIS] were born out of the dungeons of Abu Ghraib, where these torture methods were being implemented. And that’s why we’re here watching people being beheaded in orange suits – it’s a direct response to the brutal torture that took place in Guantanamo.”
Fourth, Abu Ghraib as an inspiration for attacks is not limited to Iraq. As described above, Saudi Arabia arrested fighters leaving for Afghanistan who claimed to have been radicalized by the Abu Ghraib scandal. According to CPOST data, at least some fraction of known suicide bombers who perpetrated suicide attacks in Afghanistan consists of foreigners, including Saudis. While we do not have records of these bombers’ explicit rationales, it is not a stretch to imagine humiliation of Muslims at Abu Ghraib as a source of radicalization and mobilization in some if not many of these cases.
In a video, Anwar Anwar al-Awlaki explains the motivations for Umar Farouk, who failed in his attempt to down an American airliner on Christmas Day in 2009, and Hassan Nidal, the U.S. Army psychiatrist who killed 16 U.S. soldiers at Fort Hood. He identifies “Abu Ghraib, and Bagram, and Guantanamo Bay” – along with the invasion of Afghanistan and Iraq and American drone strikes in Yemen – as reasons for these attacks. And at least one of the perpetrators of the 2015 Charlie Hebdo Paris attacks was radicalized by U.S. torture in Iraq. Sharif Kouashi, one of the two brothers carrying out the attack, was originally prosecuted by French authorities in 2007 for seeking to go to Iraq in 2005 to fight Americans. In his court statement, Sharif said, “It was everything I saw on the television, the torture at Abu Ghraib prison, all that, which motivated me.” Kouashi was only one of group of French Muslims who sought to go to Iraq at this time and most, if not all, would presumably have had similar motivations.
Finally, American leaders appear convinced of the link between torture and the risk of increased violence against Americans and American interests. American political leaders – including heads of intelligence and President Barack Obama – have been sufficiently worried about knowledge of U.S. torture creating a larger wave of anti-American terrorism and other negative consequences that they have suppressed the release of photographs and other information so as not to make the strategic costs worse. This is in spite of strong American public support for releasing them. It is a telling example of how the link between torture and risk of attacks remains salient even many years after Abu Ghraib became synonymous with the abuses of the U.S. detention program.
The third cost of torture is the loss of domestic support for military operations, hampering the ability of the United States to confront critical threats to its national security. American military presence around the globe – including in Iraq and Afghanistan – is partially justified in humanitarian terms. In 2002, American leaders bolstered the case for the Iraq war, and also expanding operations in Afghanistan, by painting military policies in glowing moral terms. Alongside national security concerns over potential weapons of mass destruction, the conquest of Iraq was presented as important for spreading American values of democracy and freedom to an unfree nation dominated by an authoritarian regime. So-called ‘liberal hawks’ were an important constituency, bolstering support for both the invasion and the continued presence of American forces after the end of combat operations. Given today’s intense partisan divide, the support of this constituency was crucial to building a broad consensus for the war in Iraq, but its continued support was predicated on the policies implemented on the ground.
Regulated and unregulated policies of torture drove away support for the Iraq War among liberal hawks. The resulting partisan divide eventually resulted in the premature withdrawal of American forces, creating the security situation that we face today. Gallup polls consistently asked the same question from March 2003 to June 2014 about whether respondents thought the war in Iraq was the right decision (Figure 2). The first 6 months of the war show high levels of support, which consistently fell over time until 37% of the public thought that the war was the right decision in June 2008, the lowest proportion recorded. The largest drop after fall 2003 is around the release of the pictures of U.S. torture at Abu Ghraib, which aired on the television program 60 Minutes on April 30, 2004. Between September 2003 and July 2004, public support for the war dropped by nearly twenty-percentage points, from a solid 57% majority to a clear 40% minority. Although poll numbers were modestly volatile over the subsequent two months, the mid-April to mid-June 2004 period marks the clear turning point when a majority of Americans first turned against the Iraq war, and the public did not consistently turn back to supporting the war from this point on.
Not only does this significant drop in public support correspond with the timing of the Abu Ghraib scandal, but the difference in public support can be accounted for by the drop in public support among respondents who felt that torture was rarely justified. Where 44% of the public, a plurality, thought that torture was rarely justified but supported the war in the September 2003 poll, only 27% of the public held such views in a July 2004 poll (Figure 1). What changed was not the public’s opinion on torture – 80% believed that torture was rarely justified in both polls – but rather support for the war among those who opposed torture. The 17 percentage point drop in support for the war among respondents who opposed torture accounts for all of the change in public opinion. While important events continued to tip public opinion in one direction or the other over the course of the war, this dip following the Abu Ghraib scandal was sustained, and public support never returned to pre-scandal levels.
Despite its steep strategic costs, it is conceivable that, in rare and exceptional circumstances, torture would produce immediate tactical benefits in lives saved, thereby outweighing the strategic costs of lives lost. In the war on terror, the torture of Khalid Sheik Mohammad (KSM) comes the closest to fitting this description, since he was directly involved in the planning of 9/11 and other active al Qaeda plots to kill numerous people and there is no way to know with certainty that torture would not have revealed crucial information that other methods could not have. However, even in this case, torture produced at most marginal benefits rather than ‘smoking gun’ intelligence that clearly saved lives. Most importantly, the use of torture was not limited to KSM, but was extended as a formal and regulated program to many dozens of individuals and as an informal and unregulated program to many more. In the end, the formal and regulated program was the root cause of all the major strategic costs that ultimately harmed U.S. national security.
How can the United States secure the tactical benefits of torture in exceptional circumstances, while avoiding the heavy strategic costs that come with producing cultures of permissiveness that facilitate the widespread, unregulated use of torture? Neither secrecy nor legislative oversight is likely to prove viable solutions. Instead, this paper contends that decisions regarding the use of torture should be authorized on an individual basis by the President of the United States. Furthermore, the President ought to conduct an assessment of the balance between tactical benefits (of torturing that specific individual) against the likely strategic costs should that torture be revealed to the public. Such an approach would maintain the exceptional nature of torture, increase the gravity of each individual choice to torture, and provide the calculus upon which this decision should be based.
Maintaining a torture program in secrecy would only temporarily delay the onset of strategic costs. This is because strategic costs set in when torture practices are revealed to the publics of both the state conducting torture and the societies containing the terrorist organizations. But sustaining the secrecy of a torture program is ultimately impossible. First, individuals within these covert organizations may leak details or documents regarding the program. Arguably, the more upsetting the practices of the program, the more likely someone will leak information. Second, even with secrecy, the practice of torture can spread through a culture of permissiveness. The proliferation of torture, including to less clandestine operations, increases the likelihood that the practice will come to public attention. The fact the United States is a democracy with weak limits on censorship of the media only increases the prospects that secret torture programs will become public.
Legislative regulations, either banning torture or regulating its use through written law, are also likely to prove insufficient. During periods of crisis, these legislative regulations are likely to be changed, if not simply ignored or superseded by the executive. Furthermore, loopholes and skewed interpretations could be used to avoid conviction. In a national emergency, laws against torture are unlikely to provide reliable shelter, and the U.S. Supreme Court routinely delays reviewing potential violations of laws limiting the executive authority in wartime.
Instead, this article recommends that the authority to approve torture of an individual ought to reside in – and remain with – the President of the United States. The right to authorize the torture of individuals should not be delegated to the CIA or any other agency in a blanket fashion. That is, the President’s approval would have to be granted for each specific individual under consideration for torture. This would break with previous practices where the CIA was given broad authorization for using the practice of torture. By relying on the President of the United States to provide authorization on a case-by-case basis, torture would remain the exception and each decision would be treated with greater gravity than was previously practiced. Part of this gravity would come from the reduction of plausible deniability on the part of the President.
In assessing each individual case, the President should weigh the estimated tactical benefits of torturing any specific individual against the estimated strategic costs should that torture become public. The tactical benefits will flow from the likelihood that the individual holds critical information that can be utilized to prevent future attacks and greatly accelerate the dismantling of an anti-American terrorist organization. In contrast, the strategic costs that are likely to result from any acts of torture entail the risk of unregulated torture, the loss of public support, and perhaps most critically, future attacks. While the costs are impossible to estimate precisely, this article has found evidence beyond a reasonable doubt that U.S. torture policies are directly linked numerous subsequent attacks. As benchmark and a lower bound, the President should consider authorizing an act of torture only if the expected benefits are worth incurring the risk of additional attacks against the United States and its allies.
Much of the debate over torture and counterterrorism revolves around the question of its tactical usefulness. The Senate Select Committee on Intelligence recently concluded that the Central Intelligence Agency gained nothing from torturing al Qaeda prisoners. According to the committee, the CIA’s program of “enhanced interrogation techniques” failed to generate useful information that wasn’t already available from other sources. The CIA disputes this conclusion. While the Agency acknowledges that the program was problematic, it argues that enhanced interrogation yielded important information that helped lead to the arrest and killing of high-level al Qaeda officials.
The tactical utility of torture is obviously an important question, because without it there can be no defense of the CIA’s actions. The focus on immediate results, however, obscures deeper questions about the long-term costs. It may be that these costs are so high that torture is never justified, even if it succeeds at eliciting reliable intelligence. So, for the purpose of argument, assume torture works. Assume that it provides intelligence that helps locate individual terrorists, reveal their networks, and provide warning about new attacks. And assume that no other methods are available to extract the same information in the same amount of time. Is it still worthwhile? Was the CIA justified in using torture in the war on al Qaeda?
Answering these questions requires thinking about strategy, which in turn requires thinking about what victory means in this kind of conflict. In general there are two ways to win a war: coercion and brute force. Coercive victories involve efforts to compel the enemy to do your will. Brute force victories are simply about destroying the enemy. The war on al Qaeda is peculiar because it does not fit neatly in either category. Al Qaeda has never been interested in a political settlement and seems insensitive to coercion. But while this makes it a candidate for brute force, it is not clear how much violence is enough to achieve victory. Measuring brute force gains against states is relatively straightforward: we can assess their bureaucratic cohesion, control over territory, and military power. This is not the case for a group that does not rely on territorial control or conventional capabilities. And even though most of the original al Qaeda leaders are dead or incarcerated, others have taken up the mantle. Far-flung terrorists call themselves “al Qaeda” to gain prestige, regardless of whether they are descendants of the original group, and they may claim responsibility for attacks whether or not they actually participated.
For these reasons, victory in the war on al Qaeda is neither a matter of coercion or destruction. It is a decision about risk acceptance. The United States can declare victory if it decides that dismantling the original al Qaeda eliminated the risk of large-scale violence against Americans. But if it is unwilling to live with the risk of smaller attacks by individuals or groups claiming to represent al Qaeda, then it will fight on and victory will remain elusive.
The best case for the strategic effectiveness of torture was that it helped unwind a particularly menacing terrorist group. Al Qaeda was unique and dangerous. It was rich, well organized, and run by charismatic leaders. It was able to operate across borders with relative impunity and it was dedicated to killing large numbers of civilians. And while the intelligence community understood the broad outlines of the group, it had little information that could be used to support anything more than occasional strikes on al Qaeda targets. It certainly did not have the ability to support a sustained counterterrorism campaign. Torture might yield the kind of information that would allow the intelligence community to disrupt future attacks and ultimately dismantle al Qaeda. Remnants of the groups would remain, as would other individuals who flew the flag without having any connection to the original group, but they would lack the organizational resources to present a real threat. At that point U.S. leaders could quietly declare victory.
But the decision to use torture was problematic, because torture creates risks of its own. News of torture might generate popular support for the victims and their cause, however odious. Terrorists may become sympathetic figures; murderous zealots may become martyrs. In addition, news of torture might stimulate a broad backlash against cooperation with the United States. Even longtime allies might hesitate to join Washington in new policy ventures, and U.S. moral suasion will be less effective if the United States loses its moral standing. Finally, news of torture might do damage to U.S. allies that cooperated in the past, even if they had nothing to do with coercive interrogation. Close allies might suffer guilt by association, and domestic opposition might constrain their policy options.
Torture also creates the risk of a domestic backlash that may ultimately limit U.S. options. Public revulsion about the nature of the fight may cause political leaders to stop fighting altogether, even if ongoing operations are tactically productive. It might also create deep legislative fissures that lead to strategic sclerosis. Disturbing stories about torture inspire both fury and intense national introspection, causing political elites to publicly question whether the war is worth fighting if it requires such methods.
Intelligence agencies usually bear the responsibility for torture, even if they are following policy direction. They may become very risk-averse in the wake of public controversies, which will make it harder to collect information on hard targets. Espionage often requires difficult moral compromises, which intelligence officials may simply avoid if they believe they will face domestic punishment later. Torture scandals might also make it harder for intelligence agencies to recruit case officers and analysts who have strong moral objections. Intelligence-policy relations may suffer if intelligence professionals come to believe that policymakers who tacitly urged them on are now hanging them out to dry. Finally, it might lead to a serious breakdown in the relationship between intelligence agencies and congressional overseers.
We do not know how many of these risks have become reality, and it will never be easy to make definitive judgments about the strategic effects of the CIA’s program. One reason is that issues like foreign opposition to U.S. foreign policy are over-determined. It is hardly surprising that the United States faced heavy criticism after reports of torture began circulating, but anti-Americanism was on the rise for other reasons, most significantly the war in Iraq and the revelations of abuse at the Abu Ghraib prison. Torture may have had a significant impact on foreign opinion and may have damaged U.S. diplomacy, but it is hard to tell, given all the other controversies that arose after 9/11.
Comparative analyses of past cases are not terribly useful given the peculiar nature of the war on al Qaeda. While other countries have suffered serious strategic consequences from their decisions to use torture, the best-known examples are not really analogous to the U.S. case. At first glance the French-Algeria War looks like an obvious cautionary tale, for instance, and the parallels are striking. In both cases interrogators believed they were getting useful information from torture, and in both cases they believed they had implicit top-cover from political leaders. In both cases politicians and operators accused each other of bad faith in the aftermath. And both cases led to a domestic controversy despite the fact that the public supported (or at least didn’t oppose) the overall war effort. Fighting al Qaeda was never controversial in the United States, but grim details of CIA techniques in the Senate report produced a burst of criticism about how the U.S. had fought. This is somewhat akin to the Algeria case, as French public opinion was surprisingly apathetic during the war until reports of torture began circulating.
But there are vast differences as well. According to the CIA, interrogators used enhanced methods on thirty-nine prisoners. Three were waterboarded, and one died in custody of hypothermia. The French, on the other hand, engaged in routine torture during the most violent period of the war, and thousands of prisoners were subject to some combination of beating, electric shocks, hanging by the feet or wrists, and waterboarding. French officials looked the other way as all this was occurring, and made no effort to control detention and interrogation procedures. This was in the context of a brutal war in which the French responded to terrorist attacks with mass arrests and the forced resettlement of thousands of Algerian Muslims. The scope of violence and the staggering amount of abuse almost certainly contributed to the domestic political upheaval that followed.
The current controversy has been far less intense, and there is no chance it will produce the kind of domestic political crisis that helped bring down the French Fourth Republic. There may be lasting effects on intelligence tradecraft and intelligence oversight, however. Intelligence officials are likely to err on the side of caution when it comes to espionage and interrogation, sacrificing opportunities to collect new information or vet existing sources in order to avoid the risk of scandal. It has obviously deepened the rift between the CIA and key lawmakers. Even longtime champions of the intelligence community, especially Democratic Senator and former Chair of the Senate Intelligence Committee Dianne Feinstein, have been outspoken in their criticism of the Agency. The collective memory of this episode may poison how intelligence officials view their overseers, and vice versa, for many years. But the long-term fallout is uncertain. It is usually easier to evaluate tactical results than strategic consequences, and it does not help that there are no obvious historical analogies to guide our judgment in this case.
If we cannot be sure about the long-term effects, what can we say about the recent past? The CIA argues that the program helped “disrupt plots, save lives, and remove senior al-Qa’ida leaders from the battlefield.” It contends that information extracted during harsh interrogations was crucial to finding and killing Osama bin Laden, the founder of al Qaeda, in May 2011. While this was an important symbolic moment in the war on al Qaeda, the strategic effects were probably quite small. Captured documents from bin Laden’s compound may have enabled the United States to strike some of his lieutenants, but his cohort was already in steep decline. It is unclear whether his death had much effect on the operations or recruiting efforts of al Qaeda franchises in other countries, which by that time were operating largely on their own, and, in some cases, were even in conflict with the surviving leaders of the original group. To be sure, his death was an opportunity to shift domestic attention away from terrorism and on to other priorities. But given that many Americans had already soured on military action in Afghanistan and the Middle East, they were unlikely to protest such a shift anyway.
More importantly, the CIA contends that the program was pivotal in locating and capturing senior leaders in the early years of the war. This may have obstructed al Qaeda’s plans because it left the group short on individuals with the wherewithal to orchestrate large-scale violence. If this is true, then torture helped elicit information that may have prevented al Qaeda from launching more spectacular attacks, including a “second wave” of 9/11-style attacks using airplanes against targets on the west coast.
Now consider the counterfactual. Suppose the CIA was unable to capture key al Qaeda figures because it did not use torture, and suppose the group was able to pull off some kind of second wave. How would this have affected U.S. strategy? In one sense the outcome would have been the same, the Bush administration would have redoubled the effort to destroy bin Laden’s group. On the other hand, more spectacular attacks after 9/11 might have had profound effects on risk assessment. U.S. officials would have been less willing to tolerate the existence of any group claiming some connection to al Qaeda, no matter its size or location. They might have leaned toward permanent vigilance. They might not have raised the idea of leaving Iraq or Afghanistan. And they would have been wary of recognizing their own successes out of fear of letting their guard down. The upshot would have been a more intense and open-ended war on al Qaeda and any associated movements, but with less chance of ever being satisfied with the result.
This counterfactual may explain the strange contradiction in the CIA’s position on torture. The Agency has gone to great lengths to defend the value of the program, but Agency leaders promise never to do it again. If it worked, and if terrorism remains a problem, then why forego the option? Domestic politics is surely part of the answer, but I suspect that changing risk assessments are more important. In the immediate aftermath of 9/11 U.S. officials faced an enemy unlike others they had faced before, including other terrorist groups. Al Qaeda’s unusual combination of resources and organization, and its commitment to killing large numbers of Americans, drove the CIA towards extreme measures. In those circumstances, officials were willing to take the risks associated with torture in order to reduce the risk of another devastating attack. But after the U.S. dismantled bin Laden’s organization, terrorism went back to being a manageable problem, and policymakers seem to have concluded that lesser terrorist groups should not dominate American long-term attention. Although critics blamed President George W. Bush for taking his eye off the ball, and President Barack Obama for what they saw as premature declarations of victory, their decisions were necessary for preventing overstretch, which is an enduring problem in American grand strategy. Such a shift might not have been possible if policymakers had rejected torture at the outset, and if al Qaeda leaders, still at large, had been able to execute another 9/11 in the early years of the war.
All of this, of course, rests on the controversial assumption that the CIA’s interrogation program worked. If the assumption is wrong then the counterfactual is irrelevant. And we should take counterfactuals with a grain of salt in any case, because we cannot know exactly how leaders would have behaved in changed circumstances and what other events might have affected U.S. strategy. But if the scenario I describe is plausible, the implications are deeply unsettling. The argument is that if torture prevented additional atrocities, then it also prevented Washington from going too far in response. This suggests that while U.S. officials sacrificed core ideological principles to protect the United States from more attacks, the main strategic benefit was protection against their own worst instincts.
 For the SSCI Report, Senate Select Committee on Intelligence, “Committee Study of the Central Intelligence Agency’s Detention and Interrogation Program, executive summary, December 2012, declassified-2014, see: http://www.intelligence.senate.gov/study2014/executive-summary.pdf.
 John Brennan to the Honorable Diane Feinstein and the Honorable Saxby Chambliss, Comments on the Senate Select Committee on Intelligence Report on the Rendition, Detention, and Interrogation Program, June 27, 2013, approved for release on December 8. Quotations are on pages 1 and 3. See Central Intelligence Agency, “Comments on the Senate Select Committee on Intelligence’s study of the Central Intelligence Agency’s Former Detention and Interrogation Program,” June 2013, declassified-2014., available online at: https://www.cia.gov/library/reports/CIAs_June2013_Response_to_the_SSCI_Study_on_the_Former_Detention_and_Interrogation_Program.pdf. Hereafter cited as CIA response.
 Moreover, the meaning and consequences of these identities can change over time. An eighteenth century slaveholder from Virginia might aspire to be President of the United States; in other settings such behavior might be frowned upon, in still others, criminalized. In contemporary society, most would dare not even consort with a reputed slave-trader.
 I have my doubts that this would have won the war, but imagine it did for the purposes of discussion.
 A quick Rorschach test: how might the U.S. label another country that announced this as their policy?
 Tony Judt (with Timothy Snyder), Thinking the Twentieth Century (New York: Penguin Press, 2012), 310.
 A $10 million settlement for the victims and their families was reached with the U.S. government in 1973; President Clinton apologized on behalf of the nation in 1997. http://www.cdc.gov/tuskegee/timeline.htm
 Hemmingway, letter to Maxwell Perkins, May 30, 1942, quoted in Sean Hemmingway (ed.), Hemmingway on War (New York: Scribner, 2003), p.299.
 This unhelpful trope is so commonly trotted out that I will play one round. So, if I were in a room full of torture implements with Steve the confessed terrorist, who had buried a nuclear weapon somewhere in New York City and with just a few precious moments to go, I might give the torture a try. Sometimes you break the law. But I would be breaking the law.
 Robert Jervis, Perception and Misperception in International Politics (Princeton, NJ: Princeton University Press, 1976) pp. 128-129.
 Senate Select Committee on Intelligence, “Committee Study of the Central Intelligence Agency’s Detention and Interrogation Program, executive summary, December 2012, declassified-2014, available online at: http://www.intelligence.senate.gov/study2014/executive-summary.pdf. Hereafter cited as SSCI.
 For the CIA’s response, see Central Intelligence Agency, “Comments on the Senate Select Committee on Intelligence’s study of the Central Intelligence Agency’s Former Detention and Interrogation Program,” June 2013, declassified-2014., available online at: https://www.cia.gov/library/reports/CIAs_June2013_Response_to_the_SSCI_Study_on_the_Former_Detention_and_Interrogation_Program.pdf. Hereafter cited as CIA response.
 Raymond Tolliver, The Interrogator: The Story of Hanns Joachim Scharff, Master Interrogator of the Luftwaffe, rev. ed. (Lancaster, PA: Schiffer, 1997).
 Chris Mackey and Greg Miller, The Interrogators: Inside the Secret War Against al Qaeda (New York: Little, Brown, 2004).
 Paul Aussaresses, The Battle of the Casbah: Terrorism and Counter-Terrorism in Algeria, 1955-1957 (New York: Enigma Books, 2010).
 Raphaelle Branche, “Torture of terrorists? Use of torture in a ‘‘war against terrorism’’: justifications, methods and effects: the case of France in Algeria, 1954–1962,” International Review of the Red Cross v. 89 n.687 (September 2007), p.554.
 Intelligence Science Board, Intelligence Interviewing: Teaching Papers and Case Studies (April 2009), p. 4.
 This interviewing was not interrogation and was conducted under an interview protocol in keeping with U.S. social science standards for human subjects’ protection.
 Stathis Kalyvas, The Logic of Violence in Civil War (Cambridge: Cambridge University Press, 2006).
 SSCI 254.
 SSCI 89-90. See also examples on 94-96.
 CIA response cover note 3
 SSCI 89
 SSCI 308, fn 1725.
 SSCI 308-310 and CIA response list of examples 19.
 SSCI, 308-309.
 SSCI, 66-70.
 SSCI, 78.
 SSCI, 79.
 SSCI 76.
 Senate Select Committee on Intelligence, Committee Study of the CIA’s Detention and Interrogation Program, April 3, 2014. Declassified December 3 2014. Hereafter SSCI report.
 M. Cherif Bassiouni, The Institutionalization of Torture by the Bush Administration: Is Anyone Responsible? (Portland, OR: Intersentia, 2010), 390.
 On why the criteria for the ticking time-bomb scenario are unlikely to be met in practice, see Matthew Hannah, “Torture and the Ticking Bomb: The War on Terrorism as a Geographical Imagination of Power/Knowledge,” Annals of the Association of American Geographers 96, no. 3 (September 1, 2006): 622–640.
 The memo also defines “imminent,” arguing that “Common law cases and legislation generally define imminence as requiring that the threat be almost immediately forthcoming…threats referring vaguely to things that might happen in the future do not satisfy this immediacy requirement …because there is a lack of certainty that it will occur.” Jay S. Bybee, Assistant Attorney General, Memorandum for Alberto R. Gonzales, Counsel to the President RE Standards of Conduct for Interrogation (Washington, DC: U.S. Justice Department, Office of Legal Counsel, August 1, 2002).
 For exceptions that recognize strategic costs associated with torture, see: Joe Santucci, “A Question of Identity: The Use of Torture in Asymmetric War,” Journal of Military Ethics 7, no. 1 (March 1, 2008): 23–40; Mark A. Costanzo and Ellen Gerrity, “The Effects and Effectiveness of Using Torture as an Interrogation Device: Using Research to Inform the Policy Debate,” Social Issues and Policy Review 3, no. 1 (2009): 179–210; Emilie M. Hafner-Burton and Jacob N. Shapiro, “Tortured Relations: Human Rights Abuses and Counterterrorism Cooperation,” PS: Political Science & Politics 43, no. 03 (July 2010): 415–419.
 Regarding the Senate report, it is important to note the highly divergent views of the majority and minority views. While the (Democrat) majority report views the detention and interrogation program as wholly unnecessary and largely ineffective, the (Republican) minority report seeks to defend the CIA claim that this program did produce some unique and crucial information. Recognizing this divide, we consider the Senate report to be the majority view, unless otherwise noted in the text.
 SSCI: Findings and Conclusions, 2.
 Director, Central Intelligence Agency, C.I.A. Comments on the Senate Select Committee on Intelligence Report on Rendition, Detention, and Interrogation Program, June 27, 2013, 2, declassified on December 8, 2014.
 Mark Hosenball, “CIA Chief Admits Agency Used ‘Abhorrent’ Methods on Detainees,” Reuters (LANGLEY, Va., December 11, 2014), accessed January 15, 2015, http://www.reuters.com/article/2014/12/11/us-usa-cia-torture-brennan-idUSKBN0JP29F20141211.
 Senate Select Committee on Intelligence, Committee Study of the CIA’s Detention and Interrogation Program: Findings and Conclusions, April 3, 2014, 242.
 “President Bush Delivers Remarks on Terrorism,” The Washington Post, September 6, 2006, sec. Politics, accessed January 30, 2015, http://www.washingtonpost.com/wp-dyn/content/article/2006/09/06/AR2006090601425.html.
 Director, Central Intelligence Agency, C.I.A. Comments, 6.
 For a review, see Costanzo and Gerrity, “The Effects and Effectiveness of Using Torture as an Interrogation Device”; Alex J. Bellamy, “No Pain, No Gain? Torture and Ethics in the War on Terror,” International Affairs 82, no. 1 (2006): 121–148; Ruth Blakeley, “Why Torture?,” Review of International Studies 33, no. 03 (July 2007): 373–394; Ronnie Janoff-Bulman, “Erroneous Assumptions: Popular Belief in the Effectiveness of Torture Interrogation,” Peace and Conflict: Journal of Peace Psychology 13, no. 4 (November 6, 2007): 429–435.
 Director, Central Intelligence Agency, C.I.A. Comments, 1.
 See Robert A. Pape, Bombing to Win: Airpower and Coercion in War (Ithaca, NY: Cornell University Press, 1996); Alexander B. Downes, Targeting Civilians in War, Cornell studies in security affairs (Ithaca: Cornell University Press, 2008).
 M Vorbrüggen and HU Baer, “Humiliation: The Lasting Effect of Torture.,” Military medicine 172, no. 12 Suppl (December 2007): 29–33.
 Walter J. Torres and Raymond M. Bergner, “Humiliation: Its Nature and Consequences,” Journal of the American Academy of Psychiatry and the Law Online 38, no. 2 (June 1, 2010): 201.
 Stanley Milgram, Obedience to Authority: An Experimental View, 1st edition. (New York: Harper & Row, 1974).
 Donald C. Klein, “The Humiliation Dynamic: An Overview,” Journal of Primary Prevention 12, no. 2 (December 1, 1991): 93–121.
 Spencer Ackerman, “Torture Victims Will Bear Psychological Scars Long after C.I.A. Report Scandal Fades,” The Guardian, December 13, 2014, accessed January 30, 2015, http://www.theguardian.com/law/2014/dec/13/learned-helplessness-enduring-effects-torture-haunt-victims.
 President Bush referred to the abuse at Abu Ghraib prison as “disgraceful conduct by a few American troops” (Dana Milbank, “Bush Seeks to Reassure Nation on Iraq,” Washington Post, May 25, 2004, accessed January 30, 2015, http://www.washingtonpost.com/wp-dyn/articles/A52711-2004May24.html); and Secretary of Defense Rumsfeld called it “an exceptional, isolated” case (Donald Rumsfeld, “Defense Department Operational Update Briefing,” May 4, 2004). ). Only a few low ranking soldiers have been sentenced for their actions, see Human Rights Watch, Getting Away with Torture? Command Responsibility for the U.S. Abuse of Detainees, April 2005.
 Milgram, Obedience to Authority.
 Robert Jervis, “Tortured Interrogations,” forthcoming in Foreign Affairs, 2015
 Steven Kull, Americans on Detention, Torture, and the War on Terrorism (PIPA/Knowledge Networks, July 22, 2004); Global Policy Forum, Public Opinion in Iraq: First Poll Following Abu Ghraib Revelations, May 14, 2004.
 Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (Random House, 2005); Robert A. Pape and James K. Feldman, Cutting the Fuse: The Explosion of Global Suicide Terrorism and How to Stop It (University of Chicago Press, 2010).
 See, for example, Manuel R. Torres, Javier Jordán, and Nicola Horsburgh, “Analysis and Evolution of the Global Jihadist Movement Propaganda,” Terrorism and Political Violence 18, no. 3 (September 1, 2006): 399–421.
 Pape, Dying to Win; Pape and Feldman, Cutting the Fuse.
 See, for example, Mohammed M. Hafez, “Martyrdom Mythology in Iraq: How Jihadists Frame Suicide Terrorism in Videos and Biographies,” Terrorism and Political Violence 19, no. 1 (2007): 95–115.
 Mark Danner, “Abu Ghraib: The Hidden Story,” The New York Review of Books, October 7, 2004, accessed January 27, 2015, http://www.nybooks.com/articles/archives/2004/oct/07/abu-ghraib-the-hidden-story/.
 Final outcome of the French referendum on Algerian independence (Evian Accords), April 8 1962. Henry Ginger, “Algeria Truce Is Ratified by 90% of Vote in France in Setback for Rightists,” New York Times, April 9, 1962.
 Lou DiMarco, “Losing the Moral Compass: Torture and ‘Guerre Revolutionnaire’ in the Algerian War,” Parameters 36, no. 2 (2006): 63–76; Steven Fallon, “French Failure in Algeria: Public Relations Disaster,” Small Wars Journal (October 30, 2013), http://smallwarsjournal.com/jrnl/art/french-failure-in-algeria-a-publicrelations-disaster.
 Thomas Cushman, ed., A Matter of Principle: Humanitarian Arguments for War in Iraq (Berkeley, CA: University of California Press, 2005).
 Anthony R. Jones, George R. Fay, and United States Department of the Army Office of the Chief of Public Affairs, Executive Summary: Investigation of Intelligence Activities at Abu Ghraib. AR 15-6 Investigation of the Abu Ghraib Prison and 205th Military Intelligence Brigade (U.S. Army Public Affairs, 2004), 4.
 William Haynes II, Action Memo: Counter-Resistance Techniques (General Counsel of the Department of Defense, November 27, 2002), 1, National Security Archive at George Washington University.
 Senate Select Committee on Intelligence, SSCI, 66–70.
 Ibid., 69–70.
 Kate Zernike, “Only a Few Spoke Up on Abuse As Many Soldiers Stayed Silent,” The New York Times, May 22, 2004, sec. World, accessed January 30, 2015, http://www.nytimes.com/2004/05/22/world/reach-war-witnesses-only-few-spoke-up-abuse-many-soldiers-stayed-silent.html.
 Human Rights Watch, The Road to Abu Ghraib, June 2004, 1.
 Human Rights Watch, Getting Away with Torture?; International Committee of the Red Cross, Report on the Treatment by the Coalition Forces of Prisoners of War and Other Protected Persons by the Geneva Conventions in Iraq during Arrest, Internment, and Interrogation (ICRC, February 2004), http://www.informationclearinghouse.info/article6170.htm; Amnesty International, Iraq: Continuing Failure to Uphold Human Rights, July 23, 2003.
 Bassiouni, The Institutionalization of Torture by the Bush Administration, 390.
 Thérèse Postel, “How Guantanamo Bay’s Existence Helps Al-Qaeda Recruit More Terrorists,” The Atlantic, April 12, 2013, accessed January 30, 2015, http://www.theatlantic.com/international/archive/2013/04/how-guantanamo-bays-existence-helps-al-qaeda-recruit-more-terrorists/274956/.
 Talek Harris, “Haunting Footage of Suicide Attacks in Militant Propaganda Video,” Agence France Presse, July 6, 2006, sec. International News.
 Soj, “The 5 Minute Hate: Nick Berg,” Daily Kos, May 15, 2004, accessed January 30, 2015, http://www.dailykos.com/story/2004/05/15/29241/-The-5-Minute-Hate-Nick-Berg.
 Heghammer estimates there were between 4,000 and 5,000 foreign fighters in Iraq between 2003 and 2010, and between 1,000 and 1,500 I Afghanistan over the same period. See Table 1 in Thomas Hegghammer, “The Rise of Muslim Foreign Fighters: Islam and the Globalization of Jihad,” International Security 35, no. 3 (2010): 61.
 Global Policy Forum, Public Opinion in Iraq: First Poll Following Abu Ghraib Revelations.
 Michael E. O’Hanlon and Ian Livingston, Iraq Index: Tracking Variables of Reconstruction & Security in Post-Saddam Iraq, September 30, 2010.
 Brian Fishman and Joseph Felter, Al-Qa’ida’s Foreign Fighters in Iraq (Combating Terrorism Center, January 2, 2007), 22.
 Michael E. O’Hanlon and Adriana Lins de Albuquerque, Iraq Index: Tracking Variables of Reconstruction & Security in Post-Saddam Iraq, January 19, 2005.
 “Generation Kill,” Foreign Affairs, April 2013, accessed January 30, 2015, http://www.foreignaffairs.com/discussions/interviews/generation-kill.
 Matthew Alexander, “I’m Still Tortured by What I Saw in Iraq,” The Washington Post, November 30, 2008, sec. Opinions, accessed January 30, 2015, http://www.washingtonpost.com/wp-dyn/content/article/2008/11/28/AR2008112802242.html.
 Anthony H. Cordesman, Iraq and Foreign Volunteers (Washington D.C.: Center for Strategic and International Studies, 2005), 4–5, http://csis.org/files/media/csis/pubs/051117_iraqforeignvol.pdf.
 Stephen Webster, “Cable: Abu Ghraib Torture Inspired Hundreds of Saudis to Jihad,” December 2, 2010, accessed January 30, 2015, http://www.rawstory.com/rs/2010/12/cable-shows-torture-photos-inspired-hundreds/.
 Human Rights Watch, The Road to Abu Ghraib, 24.
 “Red Cross: Iraq Abuse ‘tantamount to Torture,’” Msnbc.com, accessed January 28, 2015, http://www.nbcnews.com/id/4944094/ns/world_news-mideast_n_africa/t/red-cross-iraq-abuse-tantamount-torture/.
 “Who Is ISIS Leader Abu Bakr Al-Baghdadi?,” Newsweek, last modified December 8, 2014, accessed January 28, 2015, http://www.newsweek.com/2014/12/19/who-isis-leader-abu-bakr-al-baghdadi-290081.html.
 Riaz Hassan, Life as a Weapon: The Global Rise of Suicide Bombings (Routledge, 2014), 56.
 Alexander, “I’m Still Tortured by What I Saw in Iraq.”
 Soj, “The 5 Minute Hate.”
 See, for example, Adam Withnall, “After the Torture Revelations, the Backlash Begins,” The Independent, December 10, 2014, accessed February 2, 2015, http://www.independent.co.uk/news/world/americas/cia-torture-report-extremist-backlash-to-controversial-report-begins-with-calls-to-take-revenge-on-the-despicable-kuffar-9914490.html.
 Angelique Chrisafis, “Charlie Hebdo Attackers: Born, Raised and Radicalised in Paris,” The Guardian, January 12, 2015, sec. World, accessed January 30, 2015, http://www.theguardian.com/world/2015/jan/12/-sp-charlie-hebdo-attackers-kids-france-radicalised-paris.
 Scott Wilson, “Obama Reverses Pledge to Release Photos of Detainee Abuse,” The Washington Post, May 14, 2009, sec. Politics, accessed January 30, 2015, http://www.washingtonpost.com/wp-dyn/content/article/2009/05/13/AR2009051301751.html.
 This broad consensus included 77 Senators and approximately 70% of the public. For an analysis of the partisan divide in Iraq, see: Gary C. Jacobson, “A Tale of Two Wars: Public Opinion on the U.S. Military Interventions in Afghanistan and Iraq,” Presidential Studies Quarterly 40, no. 4 (2010): 585–610.
 These were the most proximate polls that asked respondents about both their support for the Iraq war and their opinion on the use of torture. Data accessed through the Roper Center.
 For analysis of the causes of other major drops in public support for the war in Iraq see: John Mueller, “The Iraq Syndrome,” Foreign Affairs 84, no. 6 (November 1, 2005): 44–54; Erik Voeten and Paul R. Brewer, “Public Opinion, the War in Iraq, and Presidential Accountability,” The Journal of Conflict Resolution 50, no. 6 (December 1, 2006): 809–830.
 The false presumption that the majority of Americans supports torture is pervasive in both the public and academic discourse, for analysis of public support for torture over time, and reasons why this false belief is so common, see: Paul Gronke et al., “U.S. Public Opinion on Torture, 2001–2009,” PS: Political Science & Politics 43, no. 03 (July 2010): 437–444; Jeremy D. Mayer and David J. Armor, “Support for Torture over Time: Interrogating the American Public about Coercive Tactics,” The Social Science Journal 49, no. 4, Special Issue: National and state politics: A current assessment (December 2012): 439–446.
 Geoffrey Stone. Perilous Times: Free Speech in Wartime: From the Sedition Act of 1798 to the War on Terrorism. New York: W. W. Norton.
 “The White House Has Been Covering Up the Presidency’s Role in Torture for Years,” The Intercept, n.d., accessed January 30, 2015, https://firstlook.org/theintercept/2014/03/13/president-obama-covering-presidencys-role-torture-4-years/.
 Senate Select Committee on Intelligence, Committee Study of the Central Intelligence Agency’s Detention and Interrogation Program, December 3, 2014; http://www.intelligence.senate.gov/study2014/executive-summary.pdf
 Central Intelligence Agency, Comments on the Senate Select Committee on Intelligence’s study of the Central Intelligence Agency’s Former Detention and Interrogation Program, June 27, 2013,, approved for release on December 8, 2014; https://www.cia.gov/library/reports/CIAs_June2013_Response_to_the_SSCI_Study_on_the_Former_Detention_and_Interrogation_Program.pdf
 As one French general put it, “it was tolerated if not actually recommended.” Paul Aussaresses, The Battle of the Casbah, trans. by Robert L. Miller (New York, NY: Enigma Books, 2006), 129. U.S. interrogators may have felt the same in the aftermath of the 9/11 attacks, when Vice President Dick Cheney spoke publicly about working on “the dark side” and said it “vital for us to use any means at our disposal, basically, to achieve our objective.” Interview with Tim Russert on Meet The Press, September 16, 2001; http://georgewbush-whitehouse.archives.gov/vicepresident/news-speeches/speeches/vp20010916.html
 Alistair Horne, A Savage War of Peace: Algeria 1954-1962 (New York: NYRB Classics 2006 ), 232-234, and Benjamin Stora, Algeria 1830-2000: A Short History, trans. Jill Marie Todd (Ithaca, NY: Cornell University Press, 2001 ), 87-93
 The CIA’s declassified response to the SSCI report includes hundreds of supporting footnotes, but many of them are redacted. What remains the Agency’s summary of what it said it learned through interrogations, along with an analysis of how it used that information.” Central Intelligence Agency, Comments on the Senate Select Committee on Intelligence’s study of the Central Intelligence Agency’s Former Detention and Interrogation Program, June 27, 2013, 29.
 CIA, Comments, 8-9.