http://www.judiciary.senate.gov/hearings/testimony.cfm
< Return To Hearing
Testimony of

Philip Heymann

June 10, 2008


Testimony for the Senate Judiciary Committee
June 10, 2008

"Coercive Interrogation Techniques:
Do They Work, Are They Reliable, and What Did the FBI Know About Them?"

Testimony of Professor Philip B. Heymann*

Madam Chairwoman, Members of the Committee:

I think I can be most helpful today by addressing, as precisely as I can, the six questions that have most concerned this committee. Raised by the discovery that we have been engaged in interrogation practices that we have long condemned, the questions are about what should be permissible and what is wise in the way of interrogation practices.

I think it is clarifying to break the questions into two groups: (1) four about interrogation practices that may amount to torture; and (2) two questions about the more likely coercive successors to any such practices -- practices that are not torture but may be illegal as cruel, inhuman, and degrading practices prohibited by the Detainee Treatment Act of 2005. Finally, I will address what legislation is needed.


A. Four Questions About Torture

1. Is the prohibition against torture absolute or does the prohibition depend upon the need for the information and the harms that information may prevent?

The prohibition of torture -- in the Convention Against Torture ("CAT") and in the statute passed in 1994 by Congress to enforce that convention - is an absolute prohibition of certain, defined conduct, whatever the situation or exigency. The conduct is defined in the Convention Against Torture to forbid "any act by which severe pain or suffering ..... is intentionally inflicted on a person". Article 2 of the convention then says "no exceptional circumstances whatsoever, whether a state of war or a threat of war, internal political stability, or any other public emergency may be invoked as a justification of torture." Adhering to the convention, the Senate stated its understanding of what should constitute severe mental, as opposed to physical, pain or suffering prohibited by Article 1 but it stated no objection to the unqualified obligation to avoid torture which Article 2 demanded. This issue does not seem to be contested. In its second periodic report to the Committee Against Torture, our government stated on May 6, 2005 "The United States is unequivocally opposed to the use and practice of torture .... No circumstances .... may be invoked as justification for or defense to committing torture." (As we shall see, there is a strong but contested argument that exigencies and dangers may be considered in assessing what lesser forms of coercion are prohibited by Article 16 and U.S. statutes as "cruel, inhuman, and degrading.").

What constitutes severe physical suffering is the subject of some dispute, at the margins. Any U.S. interpretation that departed very sharply from the understanding of our closest allies would be equivalent to our renouncing the treaty as far as they were concerned. An interpretation that concluded that waterboarding is not intended to inflict the severe physical suffering on which it relies as an interrogation technique would be outside the limits of plausible interpretation.

2. Should the United States renounce its obligations under several international conventions in order to be able to defend ourselves by torture in extreme circumstances?

The short answer is "no". The reason is that the case for torture being an advantageous, much less critical, way of getting information from a suspect is unproven and weak, while the costs of reserving a right to engage in torture are real, demonstrated, and large. And so would be the costs of withdrawing from the convention against torture and other international conventions prohibiting it.

Let me very quickly detail some of the costs of preserving a "right" to torture, even in extreme emergencies. It is easiest to list them in terms of some of the groups that would pay the price.

? Those against whom torture is used who are, or turn out to be, innocent of any terrorist activity.

? Our military and civilians whose safety is at risk of reciprocal behavior by other countries.

? All of us who need the cooperation of foreign allies who have frequently decided to withhold cooperation in handling suspects because of fear of our use of torture.

? Anyone threatened by the creation of new pools of potential terrorists recruited by their sense of injustice such as followed the release of the photos from Abu Ghraib.

? The interrogators asked to participate in torture and the organizations charged with employing torture in the face of public rejection of that technique.

? Those whose political support any President needs in times of national danger and whose pride in their nation and support of its policies depend on their beliefs in the traditional decency of American practices.

? The citizens of undemocratic states whose subjection to torture has previously been lessened by their governments' fear of western reaction -- a reaction which is made impossible if we accept the same practices when the government considers them necessary.


None of these groups are affiliated with terrorism. Each of them would pay a heavy price for abandoning out commitment to avoid torture.

These costs could only be justified by quite certain and substantial benefits from withdrawing from our various obligations under international conventions to avoid torture or, alternatively, from ignoring our obligations either publicly or secretly. The benefits of keeping torture available as a tool of foreign policy or war for use in dire circumstances are, instead, rare, uncertain, and undocumented. There are, of course, undocumented claims of valuable information obtained by torture, but there is no record of how frequently we have been misled by the tortured suspects, nor has there been any effort to show that the information could not have been obtained in other ways. Supporters of torture in dire circumstances refer to the possibility of a ticking bomb situation where they allege that only the speed of torture would help. Opponents of torture point out that we have never had an occasion where the ticking bomb applies and that the likelihood that torture would be useful on that occasion is small. Let me briefly review the arguments in terms of the menu of information-gathering techniques.

Information-gathering about terrorist threats begins with a tip from an observer or a general suspicion about an organization that may be planning attacks or about a particular plan. By far the most important, as well as the first, decision our intelligence agencies have to make is whether to gather information about the organization or plan by covert techniques or to rely on overt techniques whose cost is to allow the members of the organization to know that they are being investigated. The British have relied primarily on covert techniques with notable success.

The covert techniques we can use to gather information without tipping off the organization include human surveillance, technologically enhanced physical surveillance, various forms of electronic surveillance, secret agents and informants, undercover offers, secret intelligence searches, and review of records with or without computerized data mining. These techniques have two vast advantages: they produce reliable and accurate information because the suspects do not know that information is being gathered about them; and they produce information in real-time, allowing us to act at once, not only after several hours or days of interrogation during which the terrorist group knows to hide its operation and to change its plans. In contrast, overt techniques such as interrogation allow the terrorist group to know, rather promptly, that one of its members has been captured and is being interrogated. Using that information, the group will take steps to retreat to backup plans as substitutes for whatever plan was under way and to make it more difficult for us to disable the other members of the organization.

Even in the cases where we should be paying the high price of alerting the group by moving quickly to detention and interrogation of a suspect, the benefits of torture or other forms of highly coercive interrogation are widely disputed. Coercion is likely to elicit a statement, but frequently one that is false. Any accurate information furnished can therefore only be detected by time-consuming verification. It was this disadvantage of unreliability that first caused our Supreme Court to forbid the use of coerced confessions. The problem is most severe when dealing with a carefully planned operation; for the plan is likely to include a cover story that is hard to unravel in any short period of time. The information furnished is likely to be narrow as well as unreliable. A tortured suspect is unlikely to reveal matters about which we did not know to ask; nothing will be volunteered. In sum, as the new Army Field Manual of September, 2006, states, torture "is a poor technique that yields unreliable results, may damage subsequent collection efforts, and can induce the source to say what he thinks the HUMINT collector wants to hear." (at 5-21)

Unreliability and narrowness of torture-induced statements have led highly trained questioners to turn to any of a number of alternative techniques. The FBI has had notable success in investigating the embassy bombings in 1998 by developing a mutually supportive relationship between the suspect and the agent. As Inspector General Fine's report documents, this is plainly the preferred practice of the FBI. A similar technique was embraced by Hanns Scharff, the very successful Luftwaffe interrogator of allied pilots in World War II (The Interrogator, by Raymond Toliver). Some of our allies have used deception about the privacy of induced conversations while an individual is detained; or they have gathered enough information to make the suspect believe that he is simply confirming what is already known, leaving him with little reason to bear the dangers and costs of silence. A number of interrogators consider it to be crucial to create resentment or suspicion of his former colleagues, believing that loyalty is the major barrier to cooperation. Prosecutors rely for intelligence primarily upon offers of reduced periods of detention. In forbidding coercion, the U.S. Supreme Court recognized that coercive interrogation was a tempting shortcut replacing more reliable and sophisticated ways of investigating.

Pressed with this argument, the supporters of maintaining a willingness to use torture turned quickly to the example of the ticking bomb placed in a major population center. They point out that many of the interrogation techniques listed above require more time than torture requires. Even in this case the benefits of torture are greatly exaggerated. We often have the wrong person. We may have the right person but the information he has may be inadequate to prevent the event. He may deceive us in a costly way or delay us long enough for the other conspirators to make alternative plans. Even under the pressure of torture, he may fall back to a cover story that was chosen to be not only false but also difficult to disprove.

Perhaps most important, since 9/11, we have not had a ticking bomb case - a case where we can identify a guilty suspect who we firmly believe has information that is urgently needed to save lives and where we have no other covert or overt way of obtaining that information in time. Abandoning the international obligations we have solemnly accepted at the costs I have described seems a very bad trade to preserve the mere possibility of using torture in a situation unlikely to ever arise.

3. What should a President do if he finds he is really facing a ticking bomb situation where the conditions I have just described are met?

He has three alternatives. He can, somewhat like Lincoln at the time of the Civil War, announce that he is going to disobey a law that prevents him from taking the steps essential to save many lives in an emergency; and that it is not possible to go to Congress to change the law in time to avoid a disaster. Alternatively, he can obey the torture statute and our treaty obligations, but perhaps at the cost of many lives. Citizens will differ on their preferences between the first and the second. I would prefer the first as long as the President's actions are transparent and he accepts the political, if not legal, risks of violating the law in a situation that is unexpected and is highly unlikely to occur. The third option - to act secretly on the basis of classified, implausible legal opinions - is surely the worst because it invites a broad spread of unaccountable mistreatment and invites widespread fears of what is suspected but cannot be known.

4. Do the President's options depend on whether we are at war in a way that triggers the President's commander-in-chief powers?

No. The heart of the issue has nothing in particular to do with war or terrorism. The question is what we want the President to do in the face of unanticipated emergencies. Most nations grant their chief executive extraordinary powers to deal with grave emergencies. Our constitution does not have emergency powers, but that does not avoid the problem. The emergency may be the flooding of New Orleans, the sudden arrival of a lethal flu, an earthquake like that China has just experienced, or any of a number of events that the Congress can hardly have anticipated in limiting the powers the President is authorized to execute. In each of these situations, which have nothing to do with armed conflict, the President also has to decide whether to exceed his authority in order to save lives. The decision is the same.

B. Two Questions About Highly Coercive Interrogation Short of Torture

The following questions arise legally under the prohibition of cruel, inhuman, and degrading treatment found in Article 16 of the Conventions Against Torture (CAT) and made applicable to all American interrogators by the Detainee Treatment Act of 2005 (as well as Common Article 3 of the Geneva Conventions).

5. Is the prohibition of cruel, inhuman, and degrading treatment absolute or does our reservation make relevant the harms the interrogation may prevent?

The prohibition of cruel, inhuman, and degrading treatment found in Article 16 of the Convention Against Torture (CAT) was accepted by the Senate only with the reservation that, to qualify, any such treatment would have to violate the 5th, 8th, or 14th amendment to the Constitution of the United States. The Detainee Act now prohibits any interrogation that violates Article 16 as interpreted with our reservation. It is supported by the President's executive order interpreting Common Article 3 of the Geneva Conventions. Although it is not clear that other nations interpret the scope of this prohibition as variable with the danger and depending upon the importance of the information to save lives, there are certainly arguments to that effect for the United States.

The provision of the 5th, 8th, or 14th amendments that seems most applicable to interrogation with the object of gathering intelligence (and not producing any usable confession) is the "due process" prohibition of investigative activities that "shock the conscience" under the 5th and 14th amendments. That provision was most recently construed by the Supreme Court in Chavez v. Martinez 538 U.S. 760 (2003). In deciding whether damages should be paid to an individual who was interrogated while in great pain, having been shot by a police officer, the court split a number of ways. Still, much in the opinion suggests that it may not shock the conscience to continue interrogation in such circumstances if the police have "legitimate reasons, borne of exigency ... [such as] locating the victim of a kidnapping, ascertaining the whereabouts of a dangerous assailant or accomplice, or determining whether there is a rogue police officer" (Kennedy, J.) There is also no provision forbidding exceptions applicable to Article 16 of CAT - no provision analogous to the prohibition in Article 2 of any emergency exceptions to the prohibition of torture. Finally, the very broad word "degrading" is perfectly sensible in dealing with interrogations in familiar circumstances but is so inclusive that it suggests that many people would expect exceptions to be made if lives were at stake.

6. How does the 1994 statutory prohibition of torture relate to the 2005 statutory prohibition of cruel, inhuman, and degrading treatment?

If questions arise as to the propriety of a form of interrogation, the first question to be asked is whether it is torture. If it is, it is absolutely forbidden, without any legal exception. Many believe waterboarding fits in this category.

If the administration argues that something it proposes is not the intentional infliction of severe physical pain or suffering (i.e., "torture"), it will still have to satisfy the Congress that the interrogation technique is not "cruel, inhuman, or degrading" under the Detainee Act of 2005. Because these words are very broad and quite vague -- and might well encompass most of the contested techniques that were used frequently in Guantanamo, Abu Ghraib, or Baghram as described by I.G. Fine - it was wise for the Congress to require in §6(c)3 of the Military Commanders Act that:

The President shall take actions to ensure compliance [with the prohibition of cruel, inhuman, and degrading treatment], including through the establishment of administrative rules and procedures.

Besides arguing that what was done would not, even in a routine case, be considered cruel, inhuman, or degrading under our reservation to CAT because it did not "shock the conscience", an administration defending the use of techniques that others claim are "degrading" or "cruel" would have available to it an argument from exigency - that, applying the "shock the conscience" test, the Supreme Court would tolerate their use, to save lives under analogous emergency circumstances, for example by a local police department.

Whether such techniques will really help deal with an urgent terrorist danger is also widely disputed. For example, during the early days of battling the IRA, some British officials argued that such methods obtained important information. Frank Steele, a former British intelligence officer, disagreed: "In practical terms, the additional useful information they produced was ... minimal."

C. Should the Congress Attempt to Give More Specific Meaning to the Very Vague Terms "Cruel, Inhuman, and Degrading" or "Shock the Conscience" and, If So, How?

In 2005, Juliette Kayyem (now Massachusetts State Undersecretary for Homeland Security) and I proposed the following in our book "Protecting Liberty in an Age of Terror":

The Attorney General shall recommend and the President shall promulgate and provide to the Senate and House Judiciary , Intelligence, and Armed Services Committees, guidelines stating which specific highly coercive interrogation techniques are authorized....these guidelines shall address the duration and repetition of use of a particular technique and the effect of combining several different techniques together. The Attorney General shall brief appropriate committees of both houses of congress upon request, and no less frequently than every six months, as to which highly coercive interrogation techniques are presently being utilized by federal officials or those acting on their behalf.

This is not the only possibility. Let me try to list them all.

1. The Congress itself or the Administration with oversight by Congress could develop a list of:
a. Permitted forms of interrogation, or
b. Forbidden techniques


2. All or some part of that list could be classified and made available only to appropriate committees of the Congress.

3. Congress could, in addition and more broadly, provide for judicial review, by either a federal district court or the FISA court, of claims by an individual who believes his interrogation violated the due process, "shock the conscience" test.


Under the Military Commissions Act, the Congress adopted the option of directing the President to develop administrative rules and procedures for any form of coercive interrogation that might violate the "shock the conscience" standard it had mandated.

It is hard to overstate the importance of the Congress acting in one of these ways. The debates about coercive interrogation in the days to come are likely to focus on a number of the steps described by I.G. Fine, as observed by FBI agents, including: solitary confinement; hooding, prolonged standing; stress positions, sleep deprivation, and a variety of other things that do not constitute "torture" but have been considered as cruel, inhuman, or degrading by our European allies and many Americans. The FBI, the CIA, and the DOD are presently operating under three quite different sets of limits. (The FBI, like the British in recent decades, applies familiar law enforcement standards.) The Congress should provide guidance on what it considers appropriate, in light of American traditions and pride, in this realm of highly coercive interrogation.

Statutes now require every agency to comply, at a minimum, with the prohibition of interrogation techniques that "shock the conscience". But this was not to be a personal judgment of the President. Congress and federal courts should have an opportunity to express their judgments on what is consistent with the traditions of America and its closest democratic allies. That is not only required for democratic decision and essential to foreign alliances; it is also the most plausible way to give meaning to the Senate's and the Supreme Court's prohibition of what shocks the conscience of the American people.

Congress must give guidance as to what is permissible in the absence of the highly unlikely "ticking bomb". We have as a nation rejected torture, but we lack definition of what else is forbidden as a general, American practice. That lack is one of the major complaints of the FBI agents described by I.G. Fine. They are right.