Testimony on American Interrogation Policy in the War Against Terrorism

July 17, 2008

In testimony before the House Committee on the Judiciary, Benjamin Wittes examined America’s interrogation policy in the war against terrorism and offered steps towards a healthier statutory environment.

Thank you, Mr. Chairman and members of the committee, for inviting me to testify concerning American interrogation policy in the war against terrorism. I am a Fellow in Governance Studies and Research Director in Public Law at the Brookings Institution. I am the author of the book, Law and the Long War: The Future of Justice in the Age of Terror, from which this testimony is adapted. I have written extensively on the challenges to the legal system posed by the September 11 attacks. I also serve on the Hoover Institution Task Force on National Security and Law. The views I am expressing here are my own.

The histories of the interrogation programs of both the military and the intelligence community have been debated at great length. The scope of the administration’s errors and excesses are well known, as are its claims that the CIA’s high-value detainee program yielded critical intelligence unobtainable by traditional means. I do not today intend to focus on the past, but on the future, that is, on the contours of the interrogation laws we need prospectively in order to prosecute the war on terrorism in a manner at once effective and consistent with American values.

Specifically, I intend to make three points: first, that Congress in the McCain Amendment addressed the problem of military interrogations and did so successfully; second, that the policy that Congress adopted in that statute can be adapted relatively easily to address interrogations by the CIA, though not by simply applying the McCain Amendment itself to the agency; and third, that for a tiny subset of detainees, the executive is likely in the future to face overwhelming pressure to breach the rules and, consequently, that some provision to govern the circumstances of such breaches is necessary too.

First, the currently contested terrain in the battles over interrogation policy is actually much narrower than most people imagine. For one thing, it no longer involves torture, for the administration no longer asserts the legal propriety of engaging in conduct that breaches the federal ban on torture—as its earlier Office of Legal Counsel memoranda appeared to do. In practice, the contested legal ground today involves a lesser category of illegal abuse: “cruel, inhuman, and degrading treatment.” More importantly, between the McCain Amendment and the new Army Field Manual on intelligence gathering, Congress and the administration have largely solved the problem of military interrogations, meaning that the dispute today does not involve tens of thousands of military detainees.

The McCain Amendment offered an elegantly simple fix for the military’s post-9/11 interrogation woes. The law gave the military, within the parameters of more general requirements of humane treatment, great latitude to set its own rules, requiring only that it publish and follow them. This the military then did in the revised Army Field Manual released after an exhaustive internal discussion in September 2006. The new field manual offers a limited degree of additional flexibility over the old one. It also contains a great deal more specificity about all of the interrogation tactics it authorizes than did the prior version, so that interrogators have a better sense of what exactly it allows and disallows. And it spells out in considerable detail what tactics are not permitted. It complies with the Geneva Conventions. And as the military can amend it at any time, interrogation policy can remain fluid, even as it remains accountable under the law. It is a mark of how successful—if largely unnoticed—the policy changes have been that not even human rights groups today complain much about contemporary military interrogation policies. The complaints about coercive interrogations by the military, however justified, are all retrospective.

Second, the residual dispute is limited to the rules governing the interrogations of the comparatively tiny numbers of detainees held by the CIA. Though narrow, this dispute is important both because these are, generally speaking, the highest-stakes interrogations and because the rules in the CIA’s program define the outer parameters of what the United States will do in interrogations.

The current state of the law is simply inadequate. Current law articulates flat bans on vaguely defined categories of abuse: torture and cruel, inhuman, and degrading treatment. The absoluteness of the bans allows Congress to take the institutional position that it has authorized nothing untoward in any questioning of anyone. The vagueness, meanwhile, gives the administration enough interpretive latitude to authorize some pretty extreme tactics—waterboarding being only the furthest out on the limb. On the books, to put it simply, is an absolute injunction not to do anything too mean—an injunction that leaves far too open the question of what meanness is and how much of it is too much.
The result is a terrible conundrum for interrogators in the field: We want them to be aggressive, to walk up to the line of legality in an effort to get information that will stop the next attack. If any space remains between their conduct of interrogations and that line, the next 9/11 Commission will devote a chapter to the “missed opportunity” the interrogation represented; the interrogators’ timidity will become an “intelligence failure” worthy of study. Yet on the other side of that line lies illegality. And we have refused as a society to draw the line clearly or to promise that it won’t move. We are, in short, asking men and women in the service of their country to live their professional lives standing on and leaning over the border of unlawful conduct we lack the courage to define precisely. It is an abdication Congress needs to redress.

With relatively simple statutory changes, Congress can use the McCain Amendment as a model for the CIA. Within the confines of the existing legal strictures on interrogation, Congress should permit the agency the use of any technique to which it willingly attaches its name. That is, the CIA should have its own list of approved techniques, amendable at any time, to which the law binds its compliance. This list, in fact, already exists, but as in the military prior to the McCain Amendment, it exists purely as policy, not as law. The idea is to bind the agency to it and to harness the political heat associated with sunshine and congressional oversight to smooth off its roughest edges.

The CIA ought not be bound by the Army Field Manual itself. Even a completely responsible palette of procedures for the CIA would probably differ in some respects from the list in the Army Field Manual, which has to create rules not merely for the unlawful combatant but for routine interrogations of protected persons like prisoners of war and, therefore, should not even approach the legal limits of interrogation roughness. As the agency will never have custody over such privileged belligerents and will generally, in fact, hold only those unlawful combatants deemed (correctly or erroneously) the most dangerous in the world, a CIA field manual might properly permit more than the Army Field Manual does. The Army document, after all, is used in zones of active combat by thousands of young soldiers with limited training; by contrast the CIA has many fewer interrogators whom it can train much more rigorously if it chooses. The agency can probably allow more without letting things spin out of control, and it therefore might reasonably contemplate the judicious use of some of the enhanced techniques the military ultimately rejected in its rewrite of the field manual.

Whatever one might think of such tactics as yelling at detainees, removing comfort items, or denying hot rations, these acts surely come nowhere near the legal lines of a proper interrogation. Even techniques like stress positions, sleep deprivation, and temperature manipulations, which can be torturous in some iterations, can also be merely unpleasant and harassing in others. The CIA might with perfect propriety draw its policy lines in a different spot than the military, and Congress should tolerate its going deeper into the gray area than the military does.

Ideally, Congress would insist that this document, like the Army Field Manual, be openly disclosed, so that all approved interrogation techniques available under the law could withstand debate and scrutiny. This may not be possible with the CIA, however. The agency may, rather, need to maintain a certain level of ambiguity about its interrogation palette, but Congress should still require of it as much transparency and accountability as possible. One possibility would be the publication of a document that describes the techniques in general terms while leaving the more granular details to a classified version shared only with the congressional intelligence committees. The key, however, is for Congress to force the CIA to articulate—in significant detail and in public to the maximum extent possible—what it will do and what it will not do and to put the force of law behind this list. To the extent members of Congress believe that any listed techniques violate the prohibition against “cruel, inhuman, or degrading” treatment or Common Article 3 of the Geneva Conventions, they would then have ample leverage and information to force its removal.

Such a rule would, at once, ameliorate a great deal of public and international anxiety about American interrogation policy and provide ample flexibility for all but the most exigent interrogation circumstances. The public concerns about American interrogation policy, after all, do not have roots in the talismanic magic of the specific techniques approved by the Army Field Manual. There is a big gray area between what the military permits and the actual legal lines—wherever they may be. The anxiety, rather, flows from a combination of the brutality of certain specific tactics and the sense of the agency’s interrogations as constituting a lawless zone in which anything goes. Clarifying that anything does not go and that what does go—while perhaps more permissive than the military’s rules—falls within international humanitarian protections would go a long way toward addressing the anxiety. Such a rule would likely suffice for all but a tiny number of the highest-value detainees.

This brings me to my third point: A subset of high-value detainees exists which will stress the rules. And Congress cannot pretend that the executive does not sometimes interrogate highly resistant detainees in time sensitive efforts to avert catastrophes. In these efforts, the executive will face extraordinary pressure to get information and will sometimes make a decision to breach the rules in order to get it. That reality requires that the rules do something genuinely extraordinary: contemplate the circumstances of their own violation. The question—and it is a tremendously difficult question—is what legal zone the president and his administration will occupy when they do what they deem necessary in such dire circumstances.

This question has two distinct components: the legal status of the action by the agents in the field and the legal status of the order to take those actions. Many people believe in structuring the law so as to render the interrogator in the field culpable of a felony for the interrogation expected of him under these circumstances. This seems to me perverse. These people are not the guards at Abu Ghraib, people who sated their sadism by violating policies designed to protect detainees from abuse. Rather, to the extent their conduct breaches America’s international obligations—even its domestic laws—it breaches them precisely because they are following instructions, ultimately of the president, in a situation of presumably surpassing national importance. To ask that they subject themselves to prosecution for these acts is unlike anything we ask of other officials in government service for acts we expect them to take in foreseeable situations. We don’t ask police officers to face murder charges when they fire their guns and kill people in accordance with departmental policies, for example. Congress needs to create a mechanism to recognize that in such situations, the president has the authority to immunize officers for the effectuation of his orders, for which he alone stands accountable.

Congress, in fact, gives up little in recognizing this—because a crude such mechanism already exists as a matter of constitutional law. The president has the unreviewable and plenary power to pardon people for any infraction under federal law, and he has the implicit authority to promise to do so. He can already, in other words, make clear to agents in the field that he will shield them from the consequences of any illegality associated with carrying out his will—and shoulder the burden of responsibility himself.

Congress should enact a law that provides a more refined device for the president to use, in lieu of the pardon power, in such situations, one that clarifies his legal and political accountability for breaches of the normal rules and also ensures that the legislature is kept informed and has the opportunity to object. The idea is to fix accountability where it belongs—in the person of the president—and to make clear that field personnel do not commit crimes in carrying out the deviations for which the president is willing to take responsibility.

Congress can accomplish this relatively simply. The law should forbid the president to authorize any deviations from CIA interrogation policies except by written finding to the congressional intelligence committees, identifying the need for enhanced tactics in the specific case and the individual techniques he is ordering. The law should insist that these techniques under no circumstances violate the prohibition on torture, and the finding should require the personal signature of the president. Congress should require as well that the White House annually publish the number of such findings the president issues, so that while each finding would remain classified, the public may determine whether coercive interrogation has remained an exception or is drifting towards more of a norm. The law should further immunize against all criminal and civil liability those personnel carrying out the enhanced techniques specified within such a finding.

The idea here is to create a stopgap for the true emergency, an arrangement that recognizes what the president has the raw power to do and will face an overwhelming temptation to do, without overtly approving of it and while affixing the clearest of accountability for it.

Congress has already taken substantial steps towards a healthier statutory environment for interrogations. It is time to finish the job.