Should We Create Exceptions to Rules Regarding Coerced Interrogation of Terrorism Suspects?

  • Print

Law professors Norman Abrams of the University of California at Los Angeles and Christopher Slobogin of Vanderbilt University consider the legal dilemmas surrounding coerced interrogation of terrorism suspects. Their essays are part of a forthcoming book, Patriots Debate: Contemporary Issues in National Security Law. The book—edited by Harvey Rishikof, Stewart Baker and Bernie Horowitz—is scheduled for publication this summer by the ABA Standing Committee on Law and National Security, which invited both writers to address coerced interrogation.



Norman Abrams

This paper presents the case for recognition of an exception—a cabined exception—to coerced-confession rules governing the admissibility of interrogation statements in civilian terrorism prosecutions. It has been many years since the U.S. Supreme Court has addressed coerced-confession issues in a significant way. Accordingly, it might be thought: Why address an issue that appears to lie in a judicial backwater?

A number of factors account for the high court’s long-standing lack of interest in this subject. In criminal prosecutions in civilian federal courts regarding interrogation statements, Miranda has long occupied center stage. While issues of voluntariness have occasionally been addressed in recent years by the Supreme Court, and by the lower federal courts, most such instances have involved questions of whether waiver of the Miranda warnings requirement was voluntary. And where an interrogation has followed voluntary waiver of Miranda rights, defendants have had a difficult time proving that the statements they made were coerced.

It is noteworthy, too, that in the post-9/11 period, military commission prosecutions have not as yet led to many judicial decisions on coerced confessions. Of course, the main explanation for that fact (despite allegations of torture) is that not very many military commission prosecutions have been tried to completion. More recently, the Obama administration attempted to switch the focus in the military detainee context away from torture by prohibiting in any armed conflict the use in interrogation of any method that “is not authorized by and listed in the Army field manual.”

The absence of significant court decisions relating to coerced-confession issues in terrorism prosecutions on both the civilian and military sides is likely to change. On the military side, as more cases are being tried, coerced-confession issues respecting past interrogations are likely to arise frequently. For the future, since the Army field manual rules are to be applied by interrogators, likely questions in subsequent trials will be: a) whether some of the methods permitted under the manual are consistent with the confessions provisions of the Military Commissions Act of 2009, b) whether existing Supreme Court coerced-confession doctrine is applicable in military commission proceedings, and c) if so, whether application of AFM-approved interrogation methods is consistent with that doctrine. Adoption of the AFM rules may be effective to prohibit torture by the military or the CIA, but it may turn out not to have been an especially good way to ensure that interrogations are lawful under the applicable constitutional coerced-confession doctrine.

More important to the subject of this paper, in connection with civilian enforcement against terrorism, new FBI interrogation guidelines have been adopted that, subject to being upheld by the courts, permit exigent-circumstance terrorism interrogations without Miranda warnings. These guidelines should eliminate in many terrorism interrogations the issue of the voluntariness of a Miranda waiver and open the door instead to a direct focus on whether the statements were lawfully obtained under the coerced-confession doctrine.

Accordingly, the application of the coerced-confession doctrine in civilian terrorism investigations and trials (and in military cases, too) in the future is likely to be addressed more frequently in the federal courts and, in due time, may reach the Supreme Court.

The Case for Recognition of a Cabined Exception

There are a number of grounds that, taken together, support the case for recognizing a cabined exception to the coerced-confession doctrine.

  1. Exceptions to other rules of constitutional evidentiary admissibility

The fact that other closely related exceptions to constitutional rules of evidentiary admissibility have already been recognized or proposed lends credence to the idea of formulating an exception in the confessions arena. Indeed, it can be argued that recognition of exceptions to existing constitutional evidentiary admissibility rules in civilian terrorism prosecutions is a developing phenomenon. Special rules have long been applied, for example, in connection with judicial approval of electronic eavesdropping and searches in terrorism contexts, under the heading of foreign intelligence; in recent years, a related special rule has been used by a couple of federal courts in compulsory process contexts in the cases of United States v. Moussaoui (4th Circuit, 2004) and United States v. Paracha (Southern District of New York, 2006). Also, arguments have been made by scholars for extending to terrorism cases the aforementioned dispensing-with-Miranda-warnings-in-exigent-circumstances (an approach that has been implemented through adoption of the aforementioned FBI guidelines). Similarly, it has been argued by this author that a similar extension can be adopted for the exigent-circumstance exception to confrontation recently recognized by the Supreme Court. Given such exceptions in terrorism cases, already recognized or proposed, it seems natural to ask whether a similar exception might be applied to existing coerced-confession doctrine.

  1. The New York v. Quarles social-cost argument

The strongest model for the development of such an exception is the decision in New York v. Quarles, which ruled in a particular exigent circumstance that Miranda warnings were not required before questions were put to the suspect. The court applied a social-cost calculus in reaching its conclusion—namely, there might have been an unacceptable social cost if the Miranda warnings had been given and, as a result, information about a missing gun had not been forthcoming from the suspect.

  1. The social-cost issue in an investigation-of-terrorism context.

Application of a social-cost argument in an exigent circumstance-terrorism context is likely to be based on a much greater unacceptable cost than that which was involved in Quarles. In that case, the cost of not getting the needed information from the suspect—where the gun was—would be the risk to human life if a single gun fell into the wrong hands. If the needed information is not obtained in a terrorism investigation, the social cost could be the occurrence of a catastrophic terrorist act that might cost dozens, hundreds or even thousands of lives. A strong argument can be made in favor of adopting a rule permitting the use of some limited coercive techniques, beyond methods permitted in ordinary crime cases, if the gain may be to obtain information from the suspect about an impending terrorism event. Of course, the coercive techniques used against the suspect will be a greater intrusion than that involved under the Miranda exception, but the somewhat greater intrusion arguably is counterbalanced by the potentially much greater social cost thereby avoided.

  1. Can the social-cost calculation be used to justify torture?

A problem with applying a social-cost argument in support of recognizing an exception to coerced-confession doctrine is the concern that it might be used to justify extreme coercive interrogation techniques to the point where even torture might be utilized. The use of torture or other extreme methods of interrogation should be unlawful under all circumstances. But there are interrogation techniques, not remotely comparable to torture and nevertheless unlawful under existing Supreme Court coerced-confession case law—for example, the type of deception used by police interrogators in Spano v. New York (1959) and Leyra v. Denno (1954)—that government agents should be allowed to use in trying to obtain intelligence about an impending terrorism event. The way to achieve this outcome is to recognize a cabined exception to existing coerced-confession doctrine, which would only go so far as to permit the use of nonextreme interrogation methods—including some that would be unlawful under Supreme Court precedents derived from ordinary crime cases.

  1. Is there empirical evidence that supports a need for this exception?

Is there a demonstrated need to increase the range of methods that the FBI can use in interrogating suspects? Is there any empirical evidence to support such a claim? Justification for adopting a limited exception to existing coerced-confession doctrine is not based on any claim of demonstrated need. Rather, the argument is that the prevailing doctrine is more restrictive than necessary where the purpose of the interrogation is to obtain intelligence about an impending terrorism event. Avoiding the harm of a catastrophe is worth additional intrusions on the individual that do not involve the use of extreme interrogation methods. Even a small modification of confession rules will serve to give the FBI additional flexibility in conducting an interrogation.

  1. Providing guidance to FBI field agents

Under the current state of the law, FBI field agents are likely to be uncertain what confession rules will be upheld by the courts. They are in need of clear guidance on this topic. It is desirable to establish clearly drawn rules and, in the course of doing so, to give interrogators somewhat more leeway than they have under prevailing Supreme Court precedents. For example, the AFM rules or a similar body of rules to be formulated may be used as a set of guidelines for both the courts and field agents.

  1. Avoiding a further muddling of the doctrine

Absent such an exception, the doctrine in the terrorism arena is likely to become even more muddled than in ordinary crime cases, and that kind of development could begin to infect ordinary criminal cases: If such an exception is not adopted, many courts are likely to apply a more flexible approach in addressing coerced-confession issues when the interrogation is designed to elicit information about an impending terrorist act, but not all courts will be willing to overlook the strictness of ordinary-crime coerced-confession rules. As a result, absent a recognized, clearly drawn exception, more than a normal modicum of inconsistency is likely to arise in the case law, which will put stress on coerced-confession doctrine in terrorism cases and in ordinary crime investigations, too.

  1. The somewhat reduced risk of unreliability in statements obtained through terrorism intelligence interrogations

Still another reason why such an exception is warranted arises out of one of the concerns underlying the traditional coerced-confession rules—the fact that extreme coercive interrogation methods often lead to unreliable confessions: The suspect will say anything to get the pain to stop. Even a nonextreme interrogation may produce untrustworthy statements when the interrogation is conducted by law enforcement agents trying to obtain information to be used in prosecuting the suspect. In such a context, the questioning agent is strongly motivated to direct questions in a way that maximizes the possibility of obtaining a confession that can be used to convict the suspect. This is a kind of bias that can substantially influence the form and content of the interrogation and thereby increase the risk that the answers given are unreliable.

What is contemplated under the proposed exception is an interrogation directed to obtaining terrorism intelligence: Questions are directed not to obtaining a confession usable in prosecuting the suspect but rather to obtaining accurate information that can help prevent terrorist acts. In such a context, there is a premium on accuracy, not conviction. Questioning terrorists for intelligence purposes thus modulates one of the factors that ordinarily contribute to a risk of unreliability in law enforcement interrogations.

  1. Fifth and 14th amendment underpinnings of the coerced-confession doctrine

How will such a cabined exception fit with the existing coerced-confession doctrine? Initially, the idea of applying an exception via the Fifth Amendment underpinning of the traditional doctrine seems sharply inconsistent with an idea expressed in opinions of some Supreme Court justices—that the Fifth Amendment does not brook any exceptions. New York v. Quarles carved out an exception to Miranda that was arguably based on the Fifth, but the majority opinion in Quarles characterized Miranda as a prophylactic rule not of constitutional dimension. Subsequently, however, Dickerson v. United States reinstated the view that Miranda is a constitutional rule while expressly preserving the Quarles exception to Miranda. So much for the notion that the Fifth Amendment is intolerant of exceptions.

The view that coerced-confession doctrine is based in the due process clauses of the 14th and Fifth amendments provides a more receptive path to take in support of the exception. To begin with, the notion of due process seems flexible enough to include taking into account the reasons for using the interrogation techniques in question as well as what those techniques are. Thus, weighing in the balance the justification for the intelligence interrogation and the relative “coerciveness” of the interrogation techniques used against the potential social cost of not permitting certain techniques to be used seems well within a reasonable application of a due process calculus. To be sure, there is very limited Supreme Court authority to cite in support of such a conclusion, and nothing directly in point. It is fair to say, however, that nothing in existing doctrine under either the 14th or Fifth amendments stands as a clear obstacle to the development of such an exception.

In crafting and implementing an exception, numerous issues will need to be addressed: determining the terms in which the cabined exception is to be formulated; where and how to draw the line that limits the exception to nonextreme interrogation techniques; how to define exigency; how to prevent expansion of the exception through interpretative erosion of its terms or extension to other crime areas; and what should be the first step(s) on the path to the recognition of such an exception. Great care will need to be taken in both the formulation and implementation of the exception.

In view of an upcoming likely increase in the incidence of confessions litigation in terrorism cases, developing a cabined, exigent-circumstance exception to coerced-confession rules now would be both a timely and desirable addition to the existing and proposed array of exceptions to constitutional evidence rules applicable in terrorism prosecutions.



Christopher Slobogin

In the wake of 9/11, the temptation to use high-pressure interrogation techniques on individuals suspected of being terrorists skyrocketed. Professor Abrams argues that our government should be permitted to give in to that temptation, at least up to a point. He contends that: (1) there are interrogation practices that fall short of torture and yet are more coercive than practices currently permitted under American law; (2) these intermediate techniques are important tools for combating terrorism; and (3) courts can (as a matter of precedent) and should (as a matter of policy) develop a “cabined exception” to the traditional prohibition on coerced confessions so that government officials may legally use these techniques. In this response, I comment on each of these propositions.

The Scope of a Cabined Exception to the Coerced-Confession Rule

Outside of one ambiguous reference noted below, Professor Abrams does not provide any specific examples of procedures that would fall under his cabined exception. Instead, he suggests that the rules found in the Army field manual and the FBI’s new interrogation guidelines would be a good baseline for developing techniques that, while impermissible under the Supreme Court’s current interpretation of the Fifth Amendment and the due process clause, would be permissible in national security investigations. In order to get a sense of how his cabined exception would work, some understanding of those rules is necessary.

The Army field manual’s interrogation provisions, which overlap considerably with the FBI guidelines (to the extent the latter provides specifics), prohibit “torture or highly degrading acts,” a phrase meant to include forcing the detainee to stand naked or perform sexual acts, hooding or covering the eyes of the detainee, beating or shocking the detainee, or depriving the detainee of necessary food and water. Presumably, given his reference to the manual as a guidepost, Professor Abrams would not permit these techniques either. However, as of 2006 the manual no longer explicitly prohibits two techniques that it at one time banned—putting the suspect in stress positions and sleep deprivation. The manual also does not explicitly prohibit questioning of individuals who are in extreme physical pain due to an earlier injury, nor does it bar suggestions to examinees that a failure to cooperate will increase the chance of subsequent injury by third parties. So, would Professor Abrams’ cabined exception permit use of the latter techniques even if, as seems quite likely, current Supreme Court doctrine categorically rejects such practices?

The answer is not clear. But Professor Abrams appears to shy away from defending coercion induced by the threat or fact of physical harm directed at the suspect. Rather, the only specific examples he gives of techniques that he might permit despite their rejection by the Supreme Court come from Spano v. New York and Leyra v. Denno. In both cases, the police relied on psychological rather than physical coercion, principally by lying to the suspect. In Spano, interrogators made use of a police officer who was a friend of Vincent Joseph Spano’s and whom the suspect had called in a panic shortly after the alleged crime; under the interrogators’ direction, this officer told Spano, falsely, that the officer was “in trouble” with his superiors because of Spano’s phone call and that if he lost his job his pregnant wife and kids would suffer as well. In Leyra, a psychiatrist posed as a medical doctor brought in to treat Camilo Weston Leyra’s sinus condition. He then repeatedly told Leyra “how much he wanted to and could help him, how bad it would be for the petitioner if he did not confess, and how much better he would feel, and how much lighter and easier it would be on him if he would just unbosom himself to the doctor.” In both cases, the Supreme Court found the confessions to be involuntary under the due process clause.

Professor Abrams suggests, without declaring outright, that the deceptive practices in Spano and Leyra are illustrative of interrogation techniques he would permit in national security situations under his cabined exception. But it is not clear that an exception is needed to accommodate that goal. Both Spano and Leyra were decided in the 1950s. Since that time the Supreme Court has explicitly or implicitly sanctioned police failure to correct misunderstandings about the right to remain silent and the right to counsel; permitted the police to suggest, as in Spano, that a confession will prevent harm to a third party; and blinked at police lies about fingerprint evidence and a co-defendant’s confession. Most lower courts have found nothing unconstitutional about other types of lies—and in particular have been unmoved by the “pretended friend” technique, in which, as in Leyra, government agents pose as caring individuals whose only goal is to help the suspect. Several lower courts have even sanctioned police threats. Not all of these latter decisions are necessarily consistent with Supreme Court case law. But, as I have argued elsewhere, if the police avoid direct threats, promises that an officer is not empowered to make, or deception about the scope of one’s right to remain silent or right to counsel, police ruses are not coercive and should be considered constitutionally permissible under current doctrine.

The upshot of all this is that, if Professor Abrams is principally interested in allowing interrogators to deceive their sources, he does not need an exception to current law. The due process clause already grants interrogators considerable leeway in interrogating suspects. Only if Professor Abrams wants interrogators to be able to resort to physical coercion, threats of physical harm or unauthorized promises would a cabined exception be necessary as a legal matter.

The Practical Necessity for a Cabined Exception

Let us assume, however, that deception of the type involved in Spano and Leyra is still unconstitutional, or that it is not but that Professor Abrams is willing to permit at least some types of physical coercion or threats or promises that clearly are unconstitutional at the present time. Although that would mean a cabined exception to current doctrine is necessary as a legal matter, it does not mean it is necessary as a practical matter. As many have pointed out, extraordinary interrogation techniques are not a particularly effective method of getting information out of suspected terrorists.

One of the strongest advocates for the latter point of view is former FBI interrogator Ali Soufan. He points out that Abu Jandal—Osama bin Laden’s bodyguard and a figure so intimidating that guards wore masks when they interacted with him—ended up providing highly useful information after weeks of resistance and hostility, not because of tough techniques but because his interrogators finally treated him with enough respect to get him talking and then used “sleight of hand” to seal the deal. Soufan also contends that al-Qaida operative Abu Zubaydah gave up the identities of Khalid Sheikh Mohammed (the mastermind of the 9/11 attacks) and of “dirty bomber” Jose Padilla through a combination of “guile and graft.” Several other military interrogators agree with Soufan that “the best way to get intelligence from even the most recalcitrant subject is to apply the subtle arts of interrogation rather than the blunt instruments of torture.”

Some who have researched interrogations believe that even “guile” is not necessary to get information out of most domestic suspects. But even if terrorists are harder nuts to crack than the typical street criminal, they may disclose useful information prompted by nothing more than expert application of “minimization techniques,” which involve empathizing with the subject and playing down the seriousness of his or her situation. Research indicates that this tactic, which is clearly constitutional, is extremely effective at getting people to talk.

There is also the possibility that more strenuous techniques will significantly increase the unreliability of any information obtained. Professor Abrams minimizes this possibility on the ground that, in contrast to domestic police whose eagerness to obtain a conviction can lead them to overreach, national security interrogators are not interested in prosecution, but only in obtaining information that can save lives. In many cases involving alleged terrorists, however, prosecution in a criminal or military court is very much a possibility. Furthermore, the pressure to get “good” information about a terrorist attack might create the same bad incentives that a desire for usable evidence in a civilian criminal prosecution does. Cases like those of Haji Pacha Wazir, Osama bin Laden’s personal banker who was subjected to detention and interrogation for eight years before being released, suggest that national security investigators are just as subject to tunnel vision and abuse of power as domestic police.

In short, techniques beyond those already permitted by the Constitution may not be necessary to protect the country in national security situations, and may even be counterproductive. At the least, use of more serious pressure tactics should be prohibited unless there is strong proof both that the individual has important information and that less strenuous tactics have failed. Even that limited approach is problematic, however, for reasons that are best explored in connection with debunking Professor Abrams’ third proposition, which is that a cabined exception is a good idea under the conditions he posits.

The Provenance and Dangers of a Cabined Exception

To this point the argument has been that a cabined exception to the coerced-exception requirement is not necessary either as a legal or a practical matter. But even if it is, it is not justifiable on legal grounds, nor is it a good idea on policy grounds. Contrary to Professor Abrams’ claim, current precedent does not provide even a weak basis for a cabined exception to the coerced-confession rule. In any event, recognition of such an exception would be a troubling development even if it did occasionally provide useful information.

The Supreme Court decision that comes closest to providing license for a cabined exception to the coerced-confession prohibition is, as Professor Abrams indicates, New York v. Quarles. In that case, the court recognized a “public safety exception” to the general rule that police must give a suspect subjected to custodial interrogation the famous Miranda warnings. At first glance, the Quarles holding would seem to be a perfect justification for relaxing restrictions in the national security context, where public safety is surely at risk. But several reasons counsel against such a move. First, Quarles is only an exception to Miranda’s prophylactic warning requirement, not to the ban on coerced confessions. Indeed, in Quarles itself the court made clear that a claim of undue coercion would lead to exclusion even in a public safety situation. Second, the Quarles rule only applies when public safety is imminently threatened; in Quarles, for instance, the police were looking for a gun that had just been discarded. Most interrogations of suspected terrorists are conducted in nonexigent circumstances, for the purpose of gathering intelligence or eventually prosecuting the source. Only if a situation arose in which an alleged terrorist were thought to have information about an immediate threat—the proverbial ticking bomb scenario—might danger to the public of the type contemplated in Quarles be present.

Even in the latter situation, however, the best approach as a matter of policy is to adhere to a ban on coerced interrogation, which should include a prohibition on the “softer” yet unconstitutional type of coercion apparently contemplated by Professor Abrams. As Judge Richard Posner has said about torture, “having been regularized, the practice will become regular. Better to leave in place the formal and customary prohibitions, but with the understanding that they will not be enforced in extreme circumstances.”

If an exception to the ban on coercion were recognized, however cabined it may be in theory, pressure tactics will become routine practice not only in cases involving alleged terrorists, but also will eventually find their way into interrogations of anyone connected with a terrorist organization, then anyone associated with “narco-terrorism,” and finally anyone suspected of any crime. We’ve seen similar expansionary tendencies in other areas of the law. For instance, the Foreign Intelligence Surveillance Act, initially limited to wiretaps where national security was the “primary” purpose, now permits surveillance when national security is a “significant” government goal. Even more insidious mission creep has occurred in connection with the fusion centers that have replaced the Department of Defense’s Total Information Awareness program; once confined to data mining records for evidence of terrorist threats, they now routinely collect information about illegal immigrants and deadbeat dads. It is much better to have rules that apply across the board to all interrogations, with the incentive to cabin that universal application brings, than to try to limit them to a single kind of crime. Give government officials an inch …



Norman Abrams

Stripped down to essentials, Professor Slobogin’s essay advances two sets of arguments against the cabined-exception proposal that merit discussion. Professor Slobogin argues that a cabined exception is unneeded; under existing coerced-confessions law, government interrogators have enough tools available. He also contends that the proposal would have negative consequences, predicting the inevitability of mission creep and “dangers”!

In this surrebuttal, I address these two sets of issues under the following headings: (A) the desirability of providing more guidance to FBI interrogators and (B) the risk of mission creep.

A.) Professor Slobogin argues that there is no need for the exception since interrogation methods of the sort contemplated under the proposal already are legally permissible; that the only Supreme Court cases to the contrary are 50 years old (I cited as examples only Leyra v. Denno and Spano v. New York; other similar cases include Lynumn v. Illinois from 1963 and Rogers v. Richmond from 1961. He seems certain that these techniques would pass muster today, and that the earlier cases are no longer good law. I believe, to the contrary, that confession law regarding deception, psychological stratagems and the like is uncertain, and that those earlier cases have not been overruled or repudiated. True, there are also lower court cases that support his view. What he does not acknowledge is that there are also lower court cases supportive of my position, including United States v. Anderson (2nd Circuit, 1991), United States v. Pichardo (Southern District of New York, 1992) and United States v. Tingle (9th Circuit, 1981).

At best, one can say that the state of law on this subject is uncertain, which creates a problem for government interrogators. The cabined-exception proposal will help to relieve this uncertainty when it is most important to do so, in exigent circumstances/terrorism investigations.

How does the FBI view the state of interrogation law? FBI policy as set forth in the Legal Handbook for Special Agents instructs that “no attempt be made to obtain a statement by force, threats or promises,” thereby setting up a reasonably bright-line standard of what is clearly prohibited. But the handbook also states: “Although it is not possible to predict in every case whether a court will find, under all the circumstances presented, that the statement was a product of the accused’s free will or a product of coercion, there are predictable factors that a court will examine in making its determination.”

The handbook then proceeds to list 11 factors, including “threats and psychological pressure; … isolation, incommunicado interrogation; … trickery, ruse, deception; … promises of leniency or other inducements.”

The statement of policy concludes with the following observation: “It must be kept in mind that the above factors are merely illustrative. The presence of any one or more of the factors mentioned above will not necessarily make a statement involuntary.”

So what is an FBI agent to think? Can he or she, for example, use trickery, ruse, deception or other inducements and psychological stratagems? The handbook’s legal advice is, I believe, a respectable—albeit general and brief—description of the law in this area. Undoubtedly, however, it leaves agents uncertain about what they can do.

The handbook approach also happens to be suggestive of one type of structure that might be used for a cabined-exception approach:

  1. Establish a reasonably clear statement of what is prohibited under all circumstances.

  2. Follow this statement with a list of interrogation practices that (given the uncertainty of the law) may, or may not, be permissible in interrogations involving ordinary crime or terrorism investigations where there are no indications of exigency.

  3. Provide that in an exigent circumstances/threat-to-public-safety interrogation to obtain terrorism intelligence, any interrogation methods that do not violate the prohibited-under-all-circumstances standard (see No. 1, above) may be used.

Note: I am not suggesting that the specific FBI handbook version of the always-prohibited standard nor the content of the book’s totality-of-circumstances listing should be used; the handbook is suggestive of a type of structural approach that might be used. The exact terms in which the exception is cast remain to be determined, subject to the limitation that no extreme methods may be used under the exception.

The advantage of the cabined-exception approach is not only that it will provide FBI interrogators with needed guidance and a safe harbor, i.e., more certainty, about what they can do in exigent-circumstance/threat-to-public-safety interrogations; it also will enable them to exercise greater flexibility in their choice of interrogation methods.

B.) Professor Slobogin and I share a concern about mission creep—that is, the risk that an exception, once established, might be extended without express authorization to other crime categories or extended in other ways. We differ, however, in our assessment of that risk. He cites the evolution of the Department of Defense’s Total Information Awareness Program to illustrate the seeming inevitability of mission creep, and even uses it as the basis for a dramatic “give the government an inch” statement at the end of his paper. But mission creep in one government program does not mean it will occur in another. A counterexample is the Foreign Intelligence Surveillance Act, legislated in 1978 as a broad exception to normal Fourth Amendment requirements. FISA, to be sure, can be criticized on various grounds, but for 33 years operations under the statute have yet to be seriously faulted for mission creep; the FISA approach has not spread to ordinary crime areas.

The FISA example illustrates how to control the risk of mission creep: Develop a principled basis for recognition of a special exception; establish clear standards in legislation or administrative regulations, or both; and subject the program to effective judicial review, meanwhile sensitizing all to the concern about the possibility of mission creep.

In conclusion, let me suggest another path to assess the merits of the cabined-exception proposal. Professor Slobogin speaks of “dangers” posed by the proposal. Rather, the issue that should be considered is: What are the dangers if the course of action recommended by one or the other of us is followed but that individual turns out to be wrong in his assessment?

If I am wrong, an unnecessary exception will be created, and there will be a likelihood of mission creep—that is, the exception will spread to other crime categories and, overall, may compromise constitutional standards governing coerced confessions.

But suppose Professor Slobogin is wrong? In that case, a cabined exception that could have been will not be created. As a result, some FBI agents, deterred by the uncertainty of the applicable coerced-confession doctrine, may be reluctant to use particular interrogation methods and thus fail to obtain information; and one can posit that in some instances that additional information might have enabled the bureau to prevent a serious terrorist event.

So which is the greater set of dangers? Of course, the nature of the probabilities can affect judgment on the issue. And how does one weigh the risk of constitutional coerced-confession doctrine being compromised against the chance that a serious terrorist event, possibly involving the loss of hundreds (or more) lives, might have been prevented? My judgment is that the latter is the graver risk, especially since I believe that the increase in the chances of obtaining relevant information and averting a terrorist event can be achieved without compromising basic societal values.



Christopher Slobogin

Professor Abrams is to be commended for his efforts to reconcile confessions law with the threat posed by terrorism. After reading his two submissions in this debate, I am willing to contemplate an exception to Miranda when public safety is seriously threatened. But I cannot agree to Professor Abrams’ exception to the prohibition on coerced confessions, which is the way he has framed his proposal. Here, I reiterate my concerns about his exception and set out a possible compromise position.

The Need for Clarity

Professor Abrams suggests that I mischaracterized his proposal by using words such as strenuous and extraordinary to describe the techniques he would permit. These words were only meant to indicate what he might be proposing (since other than the citations to Leyra and Spano he did not provide any guidance on that issue). I agree that Leyra and Spano do not involve strenuous techniques. But he now cites two other old Supreme Court cases that found a violation of due process (again without definitively declaring how they align with his exception): Lynumn v. Illinois, where interrogators threatened to take away the suspect’s children, and Rogers v. Richmond, where interrogators threatened to arrest the suspect’s wife. The techniques at issue in Lynumn and Rogers should not be legitimized even when terrorism is suspected because they create too much pressure to say anything in order to avoid the threatened harm.

Professor Abrams argues that FBI agents need clarification as to the techniques they may use during interrogation. I agree, but an exception to the coercion prohibition is not needed for this purpose. Rather, current law, which admittedly is murky, should be elucidated. In the work that I mentioned earlier I argued, in relevant part, that under current law a deceptive technique may be used during interrogation if: (1) “it is necessary (i.e., nondeceptive techniques have failed)”; and (2) “it is not coercive (i.e., … would not be considered impermissibly coercive if true).” Under factor No. 2, the question raised by the cases we have discussed is whether coercion would exist if the interrogator really were a friend (Leyra) or if, in the absence of a confession, the interrogator really would lose his job (Spano), the suspect really would lose her children (Lynumn), or the suspect’s wife really would be arrested despite her innocence (Rogers). My brief answer (and probably the Supreme Court’s answer as well) would be negative in the first two situations and, for the reason expressed above, affirmative in the last two.

Professor Abrams acknowledges the potential for mission creep but does not think the latter techniques, if permitted when a national security exigency exists, would find their way into ordinary interrogations. Consider two reasons why he might be wrong. First, the notion of a “national security” investigation is extremely amorphous. I’ve already given some examples of this fact. But consider one more. Shortly after 9/11 the FBI rounded up 4,800 Arab-Americans in Detroit, selected, according to Attorney General John Ashcroft, because “generic factors” suggested they might be terrorists; at the same time, around the country another 15,000 other men and women, mostly of Middle Eastern descent, were subject to FBI interviews. If a cabined exception existed, it could easily apply here.

Professor Abrams might respond that his exception would not come into play unless “exigent circumstances” exist. But here the second mission-creep problem arises. How does one define exigency? Professor Abrams himself states that he is willing to contemplate a concept “many degrees short” of the ticking bomb scenario. Although he probably does not mean to say so, this latter language could easily be construed to mean that anyone the government thinks has intelligence about terrorism—in other words, anyone who might have fraternized with fundamentalists, traveled to Middle Eastern countries, or visited suspect mosques—may be subject to coercive interrogation, a situation that is hardly “cabined.”

Professor Abrams raises the specter of a failure to prevent “a serious terrorist event” if his exception is not recognized. Again, there is little evidence agents need coercion to get this type of information. But if they do, coercive techniques should still be illegal; interrogators should be prosecutable or civilly liable for assaultive behavior, with the knowledge that they could assert a necessity defense. This is the approach taken in Israel, which has a much longer history of dealing with real terrorism than we do.

An Exception to Miranda?

Professor Abrams makes several comments about the FBI’s guidelines and Quarles’ public safety exception, but they do not advance his argument appreciably. I did not discuss the FBI’s approach in any detail because, in contrast to the Army field manual, it is noticeably lacking in specifics of the type just discussed. And, as I noted before, Quarles declined to recognize a public safety exception to the coerced confessions rule, stating that Benjamin Quarles was “certainly free on remand to argue that his confession was involuntary under traditional due process standards.”

However, the references to the FBI’s guidelines, which in part lay out when Miranda warnings are required, and to Quarles, which adopts a public safety limitation on Miranda, do suggest a more limited exception than the one proposed by Professor Abrams. Consider a rule declaring that, when national security is threatened, interviewers need not inform suspects of their rights to silence and counsel, and can even state that there is no immediate right to cut off questioning or to obtain counsel. This rule would allow continuous questioning despite an invocation of rights, as long as the interrogation does not become coercive through means “calculated to break the suspect’s will.” Its constitutional justification would be that the Fifth Amendment only prohibits “compulsion,” not the prophylactic rules created by Miranda and its progeny.

Even this lesser exception may not be necessary, however. After all, the only remedy for a violation of Miranda (as opposed to a due process violation) is exclusion of a suspect’s admissions from his or her criminal trial. If the goal of the government is to prevent “a serious terrorist event” or to obtain intelligence against some third party, the government can ignore Miranda with impunity, because statements made during interrogation will only be used to stop the threat or to prosecute a person who lacks standing to exclude them. Even if the government wants to use the statements against the person interrogated, a simple Miranda violation would not require exclusion from a civil proceeding focused on preventive detention, nor would it necessarily require exclusion from a military commission.

Current interrogation rules do not stymie government efforts to sniff out and prevent terrorism. Adoption of an exception to the coerced-confession rule would do real damage to suspects (most of whom will probably be innocent of any serious wrongdoing) and to the constitutional integrity of the American criminal justice system.

Norman Abrams is a professor of law emeritus at UCLA. A longtime member of the faculty, Abrams has held several senior positions, including vice chancellor of academic personnel. He is an expert on federal criminal law and anti-terrorism law.

Christopher Slobogin of Vanderbilt University, an expert on criminal procedure and mental health law, is director of the law school’s criminal justice program. He also holds an appointment as a professor in Vanderbilt medical school’s psychiatry department.

The ABA Journal’s Patriots Debate series has been publishing advance versions of the essays that will appear in the committee’s book; full versions of the essays will appear in the forthcoming book.

Give us feedback, share a story tip or update, or report an error.