Sixth public hearing of the National Commission on Terrorist Attacks Upon the United States

Statement of David Martin to the National Commission on Terrorist Attacks Upon The United States
December 8, 2003

Mr. Chairman and Members of the Commission: Thank you for the invitation to address these important questions. Our nation has responded vigorously to the terrorist threats that became vividly apparent to all of us on September 11, 2001. Some of those responses have been admirable and successful and worthy of continuation. Others, we should be learning, are far more questionable, either hampering true progress in our battle against terrorism or ignoring the vital need to protect core liberties no matter what the challenge. Some, but certainly not all, of the recent uses of immigration enforcement tools fall into the questionable category. And the current mode of application of the notion of "enemy combatant" or "unlawful combatant" to this new type of struggle poses serious threats to human rights and fundamental liberties - threats that government officials responsible for the policy seem not to treat with sufficient seriousness, I regret to say. There is a role for such a category of combatants, and the Administration is right to work to adapt some of the measures authorized by the law of war to this novel kind of struggle against non-state adversaries. But our government must be vigilant to develop the doctrine in a way that observes precise limits, in order to preserve liberty and to honor our national commitment to human rights for all persons.

I. Use of the Immigration Laws

1. Introduction: the Role of Immigration Controls and the Need to Sustain Support in the Immigrant Community

   Because the September 11 attacks were carried out by foreigners admitted to the United States for temporary purposes, some of whom had violated the terms of their admission, frustration with our immigration control machinery has been an understandable reaction. There are many problems with that machinery, and it would be good if we used the reaction to September 11 to repair it systematically, overcome our enduring national ambivalence about illegal migration, and begin to enforce the stated rules more effectively - amending those rules if we think they reach too severe a result in certain categories of cases. Unfortunately, that seems not to be happening. Instead we are selectively applying some enforcement tools with great severity, while continuing to toy with other proposals (for ill-designed amnesties or guest worker programs) that might compound the problem of mixed signals about immigration violations.

   A sense of perspective about the use of immigration enforcement tools is badly needed. Immigration enforcement by itself could eliminate a significant portion of the foreign terrorist threat only if we were to shut our borders and expel all aliens. That the nation possesses that power and, for the most part, that legal authority probably feeds popular frustration and anger at the work of our immigration control agencies. But much of that anger has been and remains unfair. We have not shut our borders and should not do so. Immigration, both temporary and permanent, serves many vital national interests. It is an indispensable part of our tradition, of who we are as a nation. We will always want to admit foreigners in fairly high numbers, just as our citizens increasingly want to travel abroad for purposes of business, tourism, study, and the like. As long as we sustain a system that includes a high volume of admissions of aliens, we will run risks. Absolute certainty in immigration control comes only with absolutely closed borders.

   Given the many reasons for a high-volume admission system, even in the wake of September 11, immigration responses to terrorism will always be secondary. It is far more effective to capture, try, and sentence terrorists than simply to exclude or deport them. Immigration tools remain important, because sometimes the best that one can do with the evidence available is to remove a potentially dangerous person from our soil. Nonetheless, it will rarely be immigration enforcement as such - visa interviews or questioning at the border or internal immigration investigations - that develops the crucial information. Such information will continue to come primarily from terrorism-specific investigations and intelligence, and the main players for these purposes will continue to be the FBI, the CIA, and related agencies. Although we must develop a better system to apply any such leads more reliably at the time when immigration decisions (visa issuance, border admission, or deportation) are made, we need to be realistic about the agencies that are best positioned to acquire such information initially.

   A major source of intelligence along these lines is the immigrant community itself, which includes naturalized citizens and their families. An immigrant community that is supportive of US government enforcement will let the authorities know of suspicious activity. It will applaud targeted efforts to find and either convict or remove dangerous persons. All immigration-enforcement responses to terrorism must be crafted with careful attention to sustaining that support and minimizing adverse reaction among the immigrant community or the segments of it that are under primary scrutiny. Some of the enforcement measures undertaken after September 11, particularly the call-in registration targeted at male nationals from specific Middle Eastern and South Asian countries, have been counterproductive in this respect. The blanket rule for closing the immigration hearings of persons on a list of special-interest aliens (whose exact criteria were unclear) compounded this alienation.

   I do not know what sorts of valuable information, if any, have been generated by the call-in registration regime or the blanket closure of these hearings. I seriously doubt whether the harvest has been enough to offset the alienation those actions fed. Selectively thorough registration at the border, in contrast, does not apply solely by nationality, can be fine-tuned more readily, potentially casts a better calibrated and less discriminatory net than the call-in registration, and is more likely justifiable. It should not be ruled out as a tool in the struggle against terrorism. But we need to be sure we apply any such controls in a manner that focuses as much as possible on individual characteristics and not on group generalizations that may serve to alienate resident communities whose support we should value.

2. Immigration Detention

   1. Framework for detention and the validity of using immigration charges as part of anti-terrorism efforts

      Detention is a legitimate response to evidence of an immigration violation, as part of the process aimed at removing the violator from the United States. The Immigration and Nationality Act clearly authorizes detention, even for what are generally seen as garden-variety immigration violations (visa overstay violations or entrance without inspection (EWI)). Before September 11, 2001, our system had evolved over many decades to a point where detention during proceedings became the exception, not the rule. Most aliens charged with immigration violations either were never detained or had an opportunity to seek release on bond or on personal recognizance. All but arriving aliens - those whose right to enter is challenged right at the border - have also had a chance to seek a reduction in the bond set by the immigration authorities, in an expedited procedure before an immigration judge known as "bond redetermination." The standard criteria for release have been the same as in the pre-trial criminal justice context: an assessment of flight risk and dangerousness. This salutary evolution in immigration practice, however, gave no one an entitlement to release, particularly given that in almost all removal proceedings the evidence of an immigration violation is solid - indeed is uncontested in the vast majority of immigration cases.

      Some commentators and judges have suggested that the most common immigration charges (overstay, entrance without inspection) are merely "technical" or are sometimes used as a pretext to deport or detain persons that the government wants removed from our soil for other reasons. Both these views should be rejected. Insisting that people come in through established procedures rather than entering surreptitiously, and also insisting that those who enter properly abide by the terms of their admission, are both fundamental to the system and should not be considered technicalities. Those whose immigration violations are unearthed because of enforcement actions started for other purposes, such as steps against terrorism, have no legitimate complaint if they are then placed into removal proceedings (and much less so if they are picked up because they already have a deportation order issued against them). In the proceedings, of course, they should be given a full opportunity to put forward any defense or request for relief from deportation that may be available.

      For these reasons, in the wake of the September 11 attacks, the government was justified in using immigration charges as a basis for detention, and to deny release on bond based on evidence that the respondents might be involved in terrorism, even if the person was not charged under one of the terrorism-related grounds of removal. It is important to be clear about this point. Sometimes the government decides to file only immigration charges rather than criminal charges, despite serious evidence of criminal wrongdoing. It may not be clear, for example, that the evidence of crime amounts to proof beyond a reasonable doubt. Or some of the key evidence may come from an informant or wiretap other intelligence method that the government decides not to reveal, but instead to preserve in order to obtain ongoing information. When the immigration agency is seeking only deportation as the ultimate sanction, it is simply good government to file only the simplest and most straightforward charge - typically an overstay or EWI charge, which can be proven from government records or the absence thereof. Such charges are rarely even contested. Terrorism charges pose, in contrast, significant challenges of case management and proof. They were among the most difficult and challenging cases that the Immigration and Naturalization Service (INS) had to deal with during my tenure there as General Counsel in the mid-1990s. There is simply no point in adding those complications if there is another easily provable charge. The person's removal from the country is the same whether the charge is overstay or terrorism.

      Nonetheless, evidence of involvement in terrorism or crime may well be highly relevant in deciding whether the person should be released pending completion of the process, because it can shed light on both flight risk and dangerousness. To exclude such evidence from the proceedings that set or review release terms, simply because the precise immigration charge does not embrace such bad acts, would be misguided. For this reason, the system has always allowed the government to use evidence that reaches beyond the precise ground of deportation as a basis for determining whether continued detention is justified. Hence, INS (now reorganized into the Department of Homeland Security (DHS)) properly used evidence of terrorist involvement in establishing the initial release terms, if any, and INS also properly presented such evidence to the immigration judge in a bond redetermination proceeding to defend its initial decision on detention. This basic system design, governing both charge and detention decisions, is sound. But it must make adequate provision for the government occasionally to use confidential information in this process - and also to build in safeguards that can substitute for the normal informed adversarial confrontation when that happens. I address those issues briefly in the Part II below.

   2. Limiting principles

      To say that it is proper to use immigration detention as part of our national response to terrorism and that the basic system design is sound does not entail approval of all uses of detention in the immediate aftermath of September 11. This system still must be run with sense that immigration detention is a serious intrusion on liberty, to be kept to the minimum justified by the facts of the specific case. The Supreme Court reminded us of these values in a ruling handed down just a few weeks before September 11, 2001. In Zadvydas v. Davis, a decision that placed limits on lengthy immigration detention, the Court began its substantive discussion with this affirmation: "Freedom from imprisonment . . . lies at the heart of the liberty that [the Due Process] Clause protects." The government's use of detention after September 11 often fell short of this standard. An important report by the Inspector General (IG) of the Department of Justice spells out these shortcomings. Many times the Department, although initially justified in its use of detention, did not follow through with needed measures to file charges in a timely fashion, to keep detention to a minimum, or to release or deport promptly those whose continued detention should not have been seen as justifiable. I will not recount the critique so thoroughly developed in the IG report, but instead will look more closely at a few particular issues. One should note, however, the unfortunate tone of the Justice Department spokesperson's primary response to this report. It was largely dismissive of the careful criticisms and constructive suggestions the report contained, emphasizing instead that the Department was making "no apologies" for its use of immigration detention powers.

   3. The time for filing charges and making an initial custody determination.

      Until late 2001, the regulations required a determination on charges and on custody or release within 24 hours of the warrantless arrest of an alien. That regulation properly reflected the sense that immigration arrest and detention constitute a serious intrusion on liberty, and so must be subject to prompt review and justification. Within a week after the September 11 attacks, the regulation was amended to extend the period to 48 hours, and the regulation allowed an escape hatch in the event of "an emergency or other extraordinary circumstance," in which case the determination is required within "an additional reasonable period of time." An extension to 48 hours may well have been justifiable, particularly in the circumstances prevalent immediately after September 11, when law enforcement agencies were under additional pressure and there was a particular need to establish an alien's true identity and to check databases for any additional terrorism-related information. But the escape hatch allowing for delayed charges in some circumstances was apparently misused. The IG report found several cases in which aliens remained detained for up to several weeks without charges being filed.

      For the future, it would be better to restore a firm and fixed deadline, no later than 48 hours after arrest, and preferably 24 hours, for the initial custody or release decision. Escape hatches, especially ones bounded only by vague terms like "reasonable period" simply invite sloppy implementation and departure from the limiting principles that should govern the use of detention. The system worked for decades with a 24-hour deadline, and we have now had over two years since September 11 to develop more expeditious ways to share relevant information on identity and terrorist involvement. Although there is apparently a long way to go before we can be satisfied with the data-sharing system, the exception for emergency or other extraordinary circumstance should now be repealed.

   4. The rules for staying release orders issued by immigration judges pending appellate review.

      In most removal cases (essentially all except those involving arriving aliens), the initial bond set by the immigration agency (or a decision to refuse release on bond) is subject to redetermination by an immigration judge. This case-by-case review is concluded expeditiously, usually within a few days of initial detention, and without many of the usual formalities required for the merits hearing. It is often conducted by telephone. Done properly, this procedure provides a useful check on the initial determination of the alien's flight risk or dangerousness, requiring the government to convince a neutral decision-maker of its case for detention. The government under some circumstances has presented confidential information on these matters in camera to the immigration judge (IJ), but the IJ has authority to probe such evidence, question the government attorney about it, and if necessary demand additional supporting information. Both sides are entitled to appeal an immigration judge's bond redetermination decision to the Board of Immigration Appeals (BIA), which also generally handles such an appeal on an expedited basis.

      In late October 2001, the Department of Justice issued an interim regulation revising the rules providing for an automatic stay, in certain circumstances, of release pending appeal of an immigration judge's reduction of bond - a regulation that has been widely misreported and discussed in overly alarmist tones. I think that regulation should be revisited, but some perspective on its history is useful.

      Immigration regulations have long provided the BIA with discretionary authority to stay the release of an alien when the government appeals an IJ's redetermination decision, and this individualized authority is largely uncontroversial. INS began considering a new rule for automatic stay of IJ release orders in the mid-1990s, when it found that immigration judges sometimes issued release decisions late on a Friday afternoon, after the close of business at the BIA's home office in Virginia, making it hard for INS to contact the BIA and request an emergency stay. Sharing the concern that some genuinely dangerous persons might be released and disappear into the populace before the BIA could be engaged, the Department of Justice therefore added a new regulation in 1997. It provided for an automatic stay upon the speedy filing of an INS appeal of the IJ's release decision, for a limited class of aliens deemed especially likely to merit two layers of consideration before release. This issue arose during my tenure as INS General Counsel. We debated at some length the best way to identify the classes of individuals that should be covered by the automatic stay provision. The final regulation ultimately provided this power only for persons removable as criminal or terrorist aliens and for whom the INS district director had originally set a bond of $10,000 or more (or denied release altogether). That level of bond was then relatively infrequent and signaled a significant degree of concern by the district director about the risks of release. The stay was to remain in effect only until the BIA ruled on the bond redetermination appeal (not the full merits of the case) - ordinarily a matter of two to three months, and sometimes concluded more quickly. We expected INS attorneys to look closely, case by case, at the merits of the IJ decision and not to deploy the automatic stay mechanism routinely but instead to reserve it for cases provoking serious concerns about the release.

      In the wake of September 11, the Department of Justice expanded the application of the automatic stay to cover all aliens for whom a bond was originally set at $10,000 or higher by INS, and in whose case the INS files a notice of appeal of the IJ's bond redetermination. Press accounts tended to portray this as an authority for the INS to overrule the immigration judge's determination - an unfortunate distortion that totally ignored the provisional nature of the action. The new amendment gave INS only the right to assure that the person remained in custody and could reliably be found while the ultimate review of his release terms took place. INS could not overrule the IJ's release decision - that question remained for the BIA to determine.

      Nonetheless, in my view it is time to revisit the automatic stay mechanism and return to a system that requires the immigration agency to take the initiative and obtain an emergency stay in any case where it believes the IJ's order poses an unacceptable risk. I say this because there are indications that the automatic stay mechanism is now being used routinely and without careful calculation by the enforcement agencies of the individual merits that led the IJ to reduce the bond in the first place. Today these are rarely cases involving hints of terrorist involvement, but instead usually arise in the context of deportation based on criminal convictions - often old criminal convictions of persons who have lived law-abiding lives since finishing their sentences. Of course, many deportation respondents who have committed violent crimes should be detained throughout the proceedings. But a case-by-case stay procedure can identify those instances. The immigration enforcement system needs to retain a sense of perspective about the level of risk such persons pose if released during proceedings. Repealing the automatic stay provisions while still assuring the BIA's power to issue a discretionary stay upon an adequate showing provides sufficient safeguards, both of public safety and of the core interest in liberty. If necessary, repeal could be accompanied by changes in BIA operations to assure that DHS can access a Board member at any necessary hour, in order to assure that the BIA can consider the stay request before the release takes place.

   5. The detention provisions in the USA PATRIOT Act

      The USA PATRIOT Act also addressed the issue of pre-charge detention, and the choices Congress made in that context are highly instructive. Section 412 of that Act added a new § 236A to the Immigration and Nationality Act (INA). It allows the Attorney General to make a special certification that an alien who is covered by the terrorism and related national-security grounds of removal should be detained essentially throughout deportation proceedings. The Administration's original legislative proposal would have been far more sweeping, providing not only for detention on the basis of such a certification, but also for essentially the rejection of any defenses to deportation for an alien so certified. Wisely, the Department of Justice quickly retreated from that position and thereafter focused only on creating a new detention power.

      Congress debated this detention provision carefully and trimmed it to its current dimensions, adding several additional safeguards. Congress specifically provided that such a certification must be done personally by the Attorney General or Deputy Attorney General, and may not be delegated to any other official. This requirement focuses personal responsibility in a Cabinet-level official for actions that run the risk of serious breach of personal liberties. Even with that personal involvement, Congress mandated that immigration or criminal charges must be filed within seven days after certification or else "the Attorney General shall release the alien." In a departure from the original Justice Department proposal, Congress required the Attorney General to review certification every six months, including after the issuance of a final removal order, in case the order could not be promptly executed. It also made the Attorney General's determinations subject to judicial review on a petition for habeas corpus filed in any federal district court, but with unique and carefully crafted appeal rights running only to the Court of Appeals for the District of Columbia.

      INA § 236A has apparently never been invoked, for reasons to be discussed below. But the fact that Congress took such pains to protect against lengthy and unjustified detention, in a statute passed in the immediate shadow of the September 11 attacks, speaks volumes about the priority that we should continue to place on minimizing executive intrusions on liberty.

      Why has the certification mechanism of INA § 236A never been invoked? I am not privy to any of the internal decision-making, but I would guess that this transpired because the normal detention and release procedures have proven adequate, in a way that might well have been doubted as of October 2001 when the law was passed. In fact, although I am only speculating, I would guess that the initial interest in a detention-upon-certification provision arose largely because of court decisions shortly predating September 11 that seemed to foreclose or render highly complicated the use of confidential information not shared with the individual - in immigration cases generally and in decisions on detention specifically. Perhaps the drafters of this section thought that the courts would be more likely to defer to a detention decision resting on confidential information if it had been made by the Attorney General or Deputy Attorney General personally.

      As it happened, one of the initial decisions hostile to the use of confidential information in this setting was vacated on appeal, and the other, though not directly appealed, came in for sharp appellate court criticism in later ancillary proceedings. Immigration judges and federal courts have lately proven more accommodating to the use of secret evidence - and so have become somewhat more likely to uphold continued detention of someone for whom the detention decision rests in part on confidential information of terrorist involvement. Thus the perceived need for a provision like INA § 236A, I am speculating, proved evanescent.

      Should INA § 236A, added by § 412 of the PATRIOT Act, then be retained? As noted the final version of this provision contains several safeguards. But the fact that we have not used it in two years strongly suggests that it should be allowed to lapse. The traditional process of case-by-case consideration of release risks, in a more diverse system of checks and balances involving several players, better honors the liberty interests at stake. At the same time, it still leaves ample room to assure continuing detention of genuinely dangerous persons, and the government now appears to be given sufficient scope, under safeguards, to present confidential information in defending detention. (See Part II below on secret evidence.)

   6. Post-removal-order detention

      So far I have considered primarily detention before the hearing or pending appeal. The IG report also identified several incidents wherein persons were detained for a lengthy period even after they received a removal order and when there were no other apparent barriers to removal to the other country. In February 2003, the Office of Legal Counsel, disagreeing with INS, determined that such extended detention in many such circumstances is statutorily authorized and constitutionally defensible, and that there is no blanket requirement to proceed with reasonable dispatch to execute a final removal order.

      OLC's stance, in my view, is out of keeping with the principles underlying the Supreme Court's Zadvydas decision, and it shows an insufficient appreciation of the real intrusion on liberty that ongoing immigration detention represents. An obligation to proceed in securing removal with reasonable dispatch does not saddle us with a wholly rigid and unrealistic timetable. It might well be within the spectrum of "reasonable dispatch," in extreme times like those we faced immediately after the September 11 attacks, to allow some play in the joints of the post-order removal schedule to permit the authorities to take extra steps to verify identity and speedily consider the possibility of lodging other charges. But such extreme times do not last forever. By now, removal should certainly occur promptly after the order is final. In any event post-order detention should not extend beyond the 90-day guideline for effectuating removal already established by Congress (INA § 241(a)), provided that the country of nationality will accept return and the alien has cooperated in completing all the necessary formalities.

II. Secret Evidence

The government has sought to use confidential information unshared with the alien respondent in a small fraction of removal cases. Traditionally such use has been authorized in connection with exclusion proceedings (now properly known as removal proceedings involving "arriving aliens") and in connection with applications for discretionary relief - which the government has argued includes release on bond. Only since 1996 has the government been authorized to use confidential information as part of the case in chief supporting removability of an admitted alien, and only in the context of unique proceedings before a special tribunal known as the Alien Terrorist Removal Court (ATRC), set up by Title V of the INA. The members of the ATRC are appointed by the chief justice for 5-year terms from among sitting federal judges with life tenure. They oversee the entire process, including the preparation of an unclassified summary of the confidential information. Title V also provides for appointed counsel for ATRC proceedings (otherwise generally forbidden in immigration cases) and for specially cleared counsel in some settings, who may see the classified information and then offer arguments or cross-examination on behalf of the respondent. To date the ATRC has not been used, probably owing to the very narrow range of circumstances that come within its jurisdiction - a statutory restriction that is not well understood.

The use of such confidential information against an individual raises exceedingly difficult issues. Obviously an individual confronted with only vague indications of the derogatory information is left almost wholly at sea in trying to prepare a defense. For an innocent respondent, such a trial could be wholly Kafkaesque. Moreover, the chance to shield information from close scrutiny has sometimes made room for abuses or shoddy practices on the part of the enforcement agencies, relying too much on rumor or information planted by personal enemies or estranged spouses. On the other hand, sometimes the government does have important information with a legitimate bearing on the case that cannot be revealed publicly lest it endanger a valued intelligence source who has forwarded the information at great peril, or who can continue to supply important information if not compromised. Therefore very weighty interests can be found on both sides of the debate over the use of secret information.

In 2002 I wrestled with these questions at some length in an article published in The Supreme Court Review. The relevant portions of that article are being made available to the Commission, and I will not repeat in this statement the full argument I made there. Instead, I offer this summary of my position:

1. Secret evidence should not be usable as the basis for deporting persons admitted as lawful permanent residents. They have been specifically invited to make a permanent home in this country, and we should not uproot them without giving them a full chance to confront the information against them.

2. The government should retain the right to use classified information in other immigration cases, but only subject to substitute safeguards.

   1. First, the executive branch should implement its own high-level internal review mechanisms to assure that the secret evidence it proposes to use is reliable and well-supported and that there are no reasonable alternatives to its use. (Such a disciplined DOJ review process, overseen by the Deputy Attorney General's office, was initiated in the late 1990s, after several adverse court decisions on secret evidence questions.)

   2. Beyond this, in many settings, including decisions on detention during proceedings or in admission decisions, the minimum safeguard should consist of sharing the classified information in camera with the immigration judge and eventually with any reviewing federal court. The judge should generally have the authority to cross-examine a government agent familiar with the evidence. Although this questioning cannot be as thorough or as well-informed as would be full cross-examination by the alien respondent's counsel with access to the information, it nonetheless provides a real check against misuse of this procedure. Moreover, government uncertainty about how thorough and probing this inquiry will be provides an added inducement for disciplined internal review before the immigration agency decides to rely on confidential information..

   3. It may be wise to go beyond the minimum indicated above, however, particularly in settings where the stakes are potentially higher, such as asylum cases. For these purpose, the ATRC model provides a range of useful additional protections, some or all of which could be employed in that small number of cases where confidential information is used. We should consider amending Title V of the INA to make these procedures available in a wider array of circumstances.

III. Enemy Combatants

1. The Government's Position and the Court Decisions to Date

   In the wake of September 11, the Administration began to assert broad authority to detain foreign nationals believed to be involved with Al Qaeda or the Taliban as "enemy combatants" or "unlawful combatants." This practice commenced in connection with theater-of-combat prisoners eventually moved to a detention facility at the US naval base at Guantanamo, although some are apparently also held in facilities in Afghanistan and perhaps elsewhere. The government eventually applied this doctrine in full measure to US citizens as well. This stance became clear when one of the Guantanamo detainees, Yasser Esam Hamdi, turned out to have a solid claim to US citizenship, though he had grown up abroad. He was thereupon brought to US territory but has been subjected to the same basic sort of detention as an enemy combatant, in a naval brig in Norfolk, Virginia. The government expanded the sweep of this claimed authority much further when it declared, in June 2002, that a US citizen picked up in Chicago and initially held on a material witness warrant, Jose Padilla, is an unlawful combatant affiliated with Al Qaeda. He was transferred to a military prison in Charleston, South Carolina, where he has so far been denied access to counsel and has not been charged with any crime.

   The government asserts the authority to hold such enemy combatants under the law of war "at least until the end of hostilities" - apparently meaning the war against terrorism, whose end is not anticipated for a very long time. It asserts this authority against both citizens and aliens, whether picked up on a battlefield or even at O'Hare Airport (as was the case with Padilla). It has strongly resisted court review of such detentions, or failing that, has argued for an extraordinarily deferential standard of review. Moreover, it has asserted the authority to hold such persons virtually incommunicado, as part of a "stress and duress" interrogation regime meant to break down the person's resistance to providing the information the interrogator seeks. This technique is described in remarkably candid terms in a declaration annexed to a motion filed in the second stage of the Padilla case, imploring the district court to reconsider its rather mild initial order that Padilla, who had then been under a highly restricted detention regime for seven months, be allowed to consult with counsel in connection with his habeas corpus proceedings. Even a brief contact with counsel, the declaration asserted, would damage its interrogation strategy, perhaps irrevocably. (The court rejected the motion.) Furthermore, the President made a blanket determination that all these "enemy combatant" detainees, even Taliban fighters, were not entitled to prisoner of war status under the Third Geneva Convention of 1949, in part because those forces did not themselves observe the laws and customs of war. The US government did eventually go on to say that it would still provide the core protections of that treaty for the detainees. Precisely what this means is unclear, but monitoring visits by the International Committee of the Red Cross (ICRC), which plays a unique role under the Convention, have been taking place.

   The government to date has prevailed in its view that no court has jurisdiction to consider the habeas petitions filed by the Guantanamo detainees - but the Supreme Court granted certiorari in November to consider precisely this question. The Fourth Circuit ruled in the Hamdi case that habeas jurisdiction lies, given Hamdi's US citizenship. But because he was captured in a "zone of active combat operations," the court treated as virtually conclusive an affidavit by a government official giving the broad outlines of the basis for the executive branch's determination that Hamdi is an enemy combatant. Hamdi was not allowed to be heard in court in order to controvert those factual assertions. The district court in the Padilla litigation accepted most of the government's proposed enemy combatant doctrine, but it did set forth a slightly more searching standard of review. The government's determination that Padilla is an enemy combatant would be sustained, the district court said, if the authorities provide "some evidence" to support this factual conclusion. But Padilla is entitled to challenge the government's evidence, and must be permitted to consult with counsel in connection with this process. Both sides appealed, and the Padilla case was recently argued in the Court of Appeals for the Second Circuit.

2. Evaluation

   If the government's position regarding enemy combatants is valid, then the congressional and regulatory debates recounted in Part I of this paper look silly in retrospect. The question whether it should be permissible to detain a suspected alien terrorist for 48 hours or seven days or a "reasonable period" before the filing of immigration charges is dwarfed by the assertion that he may be detained indefinitely based solely on an executive announcement that he is an enemy combatant. If habeas does not lie when the enemy combatant label is rolled out, even with regard to a US citizen, or if the review standard is satisfied by an unchallengeable affidavit filed ex parte by a Defense Department official, then Congress's disputes over the precise form of habeas corpus review to incorporate into the USA PATRIOT Act for that circumscribed immigration detention provision are really beside the point. Even the wide-ranging debates over the validity or wisdom of the military commissions that President Bush authorized in his Military Order of November 13, 2001, seem to be irrelevant. Detainees in most instances (absent a capital charge) would almost surely prefer trial by commission, whatever its defects, to indefinite incommunicado detention under coercive interrogation. The momentum for trial by commission, which appeared powerful shortly after the President's Order, seemed to dissipate when, several months later, the administration's doctrine regarding the detention of enemy combatants became fully formed.

   1. Applying the laws of war to the battle against terrorism

      In my view, what is objectionable is the extraordinary sweep of the government's asserted powers with regard to "enemy combatants" - not the underlying idea that the law of war should be applied in some fashion in our struggle against terrorism after September 11. For here our government has faced a very real dilemma. Until 2001, we tried generally to deal with terrorist threats through the law enforcement model - apprehending terrorists whenever possible and placing them on trial before the ordinary courts, with all its customary protections, including jury trial and the exclusionary rule. Maybe we should have realized earlier the shortcomings of that model for dealing with a terrorist threat on the scale posed by Al Qaeda - but we certainly saw its limits after September 11. It is not mere rhetoric to suggest that the United States (along with much of the world community) is now engaged in a war against terrorism and should use the wider range of tools that a state of war brings into being.

      If we are in a war against terror, a different set of rules applies to our actions in the struggle. Most important for present purposes are two changes: (1) War allows preventive detention of captured combatants until the end of hostilities, without any need to prove criminal charges. (2) War privileges certain acts that would otherwise be considered murder or summary execution or other serious crimes. A nation at war may validly track down and kill those who would do it harm and may deploy weapons of vast destructive power - obviously without any need for warrants, indictments, trials, or sentences. Given the types of attacks we have experienced, the United States was justified in invoking these powers in the struggle against terrorism. Congress strongly blessed this approach in its September 18, 2001, resolution authorizing the use of force in response to the attack. Resolutions adopted by the UN Security Council and by NATO fully support treating the September 11 violence as an armed attack that brings into play many of the rules governing armed conflict.

   2. With the added powers must come certain protections

      The key point, however, is that war brings into play a different set of rules, not the absence of rules altogether. The world community has gone through a painful evolutionary process over the past 100 years to try to refine the laws of war so that the important additional war powers mentioned in the preceding paragraph would not lead to indiscriminate cruelties. The United States has participated actively in this development of well-crafted protections, and US military lawyers have been among the leaders in asserting the importance of these humanitarian limits. Because of the United States' far-flung military role in an era when we are the only superpower, we above all have a major stake in avoiding erosion of these protective provisions.

      The protective provisions most relevant here are: (1) the safeguards of the Third Geneva Convention applicable to prisoners of war (POWs) throughout their detention, and (2) the rules requiring that privileged violence be directed only at military targets and adversary combatants, carefully distinguishing civilians to the greatest extent possible (the principle of distinction).

      POWs. The trade-off for wide powers of preventive detention has been a set of specific protections for those thus detained, recognizing basic human dignity and protecting them from outrages, mistreatment, and coercion. Article 17 of the Convention specifically provides:

      No physical or mental torture, nor any other form of coercion, may be inflicted on prisoners of war to secure from them information of any kind whatever. Prisoners of war who refuse to answer may not be threatened, insulted, or exposed to any unpleasant or disadvantageous treatment of any kind.

      The detention of POWs, in short, is preventive, not punitive - meant to keep them from engaging any further in the conflict. Importantly, however, such detainees are not immune to punishment, provided their guilt of an offense is properly established in a proceeding that is basically equivalent to the kinds of trials used for punishing members of the military of the detaining power. The proceedings must also meet certain other minimum protections spelled out in Articles 99-107 of the Convention. These articles represent an important stage in a successive expansion of protections, through different generations of treaties, in order to close loopholes that various warring parties had invoked or distorted under prior treaty regimes, often to the distinct detriment of US POWs. The United States, in response to these abuses, has been in the forefront of refining the protections and insisting on their observance. .

      Distinguishing civilians. Protecting civilians in the fashion required by the Geneva Conventions is problematic in the context of a war against terrorism, precisely because terrorists try to appear as ordinary civilians up until the moment when they unleash the violence. This very different kind of conflict therefore may well call for some new approaches, some modification of inherited doctrine. But it does not mean that we can or should simply abandon the principle of distinction and the need to build in mechanisms that work hard to distinguish combatants from true civilians. Indeed it is vital to try to shore up this principle during in the face of terrorism, which is designed precisely to tear it down.

      I have no easy insights as to how to do this, but a good starting point would be an acknowledgment on the part of our government that the principle of distinction remains valid, and some sign of creative action to reinforce it in the new circumstances. At the very least, in the context of prolonged preventive detention, the principle calls for an effective mechanism that gives detainees a timely opportunity to show that they are in fact innocent civilians and to win prompt release in that case. On December 1, 2003, a National Public Radio report stated that as many as 20 percent of the Guantanamo detainees will soon be released, largely based on a judgment that these 140-plus persons had no significant ties to Al Qaeda or the Taliban. The report suggested that most of these people were simply in the wrong place at the wrong time and that many were handed over to the United States by warlords in Afghanistan in return for a bounty the United States was offering. If this report is true, we should be ashamed that it has taken this long, nearly two years in most such cases, to recognize and rectify the error. It apparently also took the added inducement provided by the Supreme Court's recent acceptance of the Guantanamo habeas case to push the US authorities to speed up the review process that could result in release of these "civilians."

      The key flaws in the current approach. The main objection to the Administration's "enemy combatant" doctrine is this: The Administration appears to want the added powers that come with treating the anti-terrorist struggle under the law of war, but without most of the concomitant restrictions or protections, even in some form appropriately modified to take account of the different character of this conflict. Government spokesmen on this question are fond of invoking the Supreme Court's Quirin decision, the Nazi saboteur case, as though it blesses all the current actions against unlawful combatants. But that decision simply cannot carry us that far. Quirin arose in the context of a declared war and applied to a handful of individuals who were put ashore in uniform by German submarines. Even after their capture, they were allowed counsel and were placed on trial, not held incommunicado in conditions that included coercive interrogation.

      Equally chilling is the fact that the government's arguments regarding its enemy combatant doctrine give little sense of how profound a line is being crossed, particularly with the Padilla detention. We suffered a terrible blow on September 11, but it must not be allowed to obliterate the accumulated learning of centuries, dating all the way back to Magna Carta, about the dangers of unchecked executive detention authority or about the need to maintain important humanitarian lines even as we engage in armed conflict against a new and ruthless enemy. If the Supreme Court ultimately sustains Padilla's detention on the basis of pure executive dictate that he is an enemy combatant, with only the most limited judicial review of the basis for detention, then I fear greatly for our liberties. Supreme Court legitimation of indefinite incommunicado detention of a citizen picked up far from any battlefield would surely result in expanded use of that power thereafter, particularly since the war against terrorism will be with us for many years or decades to come. Invoking that power against other targets would be a continuing and enormous temptation for government officials. I deeply hope that the Supreme Court will declare the practice imposed on Padilla flatly unconstitutional. To do so in the current climate will require courage - but what is judicial life tenure for if not to permit such a ruling, even in the face of justifiable fears about dangerous foes?

      Conditions of confinement. I am also dismayed at the lack of public outcry against the conditions of confinement and the interrogation techniques our government has decided to employ. The government's Padilla declaration described techniques designed to establish the subject's unrelieved dependency on the interrogator, who becomes essentially his only connection to the outside world - a process the declaration says may take years. This sounds remarkably like techniques used by Communist countries that we denounced in an earlier era as "brainwashing." By the 1970s and 1980s, when other countries were using such techniques, the United States often criticized the practices at least as cruel, inhuman, and degrading treatment, which is outlawed by international human rights treaties. And we did not then hesitate to question the validity of information ultimately tendered under such circumstances, believing it tainted by a broken subject's eventual desire simply to say whatever he believes the interrogator wants to hear.

      Another sign of the disturbing nature of the conditions of detention may be found in the reaction of the International Committee of the Red Cross to what it found in Guantanamo, where it has been allowed a limited monitoring role. In October, the ICRC publicly raised concerns about "a worrying deterioration in the psychological health of a large number" of the detainees, owing in large part to the complete uncertainty about their ultimate fate and the lack of any means of legal recourse. "The ICRC's main concern today is that the US authorities have placed the internees in Guantanamo beyond the law." The organization also made brief and unexplained reference to other "significant changes" it has asked the United States to implement. The press reported these ICRC comments, but few press accounts gave the full picture of their true significance. The ICRC makes complaints of this kind public only on rare occasions. It considers that its role is primarily to raise any concerns directly with the government involved, and it prefers that such a process remain confidential. Only when it feels that its concerns are highly serious and that the government at issue is unresponsive does it go public. For the United States, a traditional leader in pressing compliance with humanitarian law, now to find itself the subject of public ICRC criticism is deeply regrettable.

   3. For the future

      What we need in the new context is a serious effort to adapt the traditional laws of war to this new kind of conflict - an adaptation that recognizes both expanded powers (as compared to dealing with terrorists under the law enforcement model) and the abiding need for limits and protections against possible abuses. It would be best if the executive branch were to take the lead in this elaboration process, but as I said, its actions so far fall almost all on the expanded power side and show little sign of wrestling seriously with the need for limits. It is not too late to change that unfortunate course, although I doubt we will see such a change in the absence of court decisions invalidating key pieces of the current approach.

      Congressional action. One might also have expected Congress to take the lead in developing a modified code for this new type of conflict, but Congress has so far shown little interest in plunging into the "unlawful combatant" fray. Nonetheless, the Padilla case and the Hamdi case, involving US citizen detainees, may well result in the stimulus for a healthy congressional debate on these questions. That is, suppose the Supreme Court is deeply troubled about the prospect of broad executive authority to detain US citizens as enemy combatants (at least absent circumstances like Quirin, which involved a declared war and undisputed members of an organized enemy military), and yet is also concerned about placing constitutional limits on executive war-fighting authority that might later prove impracticable. Current statutes provide a subconstitutional way for the Court essentially to remand the issue to Congress while leaving the ultimate constitutional boundaries for another day.

      The key statute was enacted in 1971, as 18 U.S.C. § 4001(a). It provides that: "No citizen shall be imprisoned or otherwise detained by the United States except pursuant to an Act of Congress." This provision was specifically intended to help guard against any recurrence of citizen roundups like that of Japanese-Americans during World War II. A court could base a decision ordering the release of Padilla or Hamdi on this statute (unless he were then properly charged under an act of Congress), without having to reach the question of the outside constitutional limits on the President's powers to detain enemy combatants. In § 4001(a), Congress has in essence already enacted a provision telling the President that he must come back to the legislature to spell out the exact framework for any new power of this sort and win its enactment, before beginning to impose it on US citizens.

      It must be acknowledged that both Hamdi and Padilla invoked 18 U.S.C. § 4001(a), without winning this form of remand to Congress. That is because both courts found in Congress's Authorization for the Use of Military Force resolution of September 18, 2001, to be the empowering Act of Congress required by the 1971 statute. I believe that that conclusion is erroneous (although it may have somewhat stronger warrant in the Hamdi case, where the detainee was captured on the battlefield in an area of active combat operations). The language of the September 18 resolution is extremely general. It simply authorizes the President to "use all necessary and appropriate force against those nations, organizations or persons" he deems responsible for the terrorist attacks. To my knowledge, no one had suggested as of that date, one week after the initial terrorist attacks, anything like the sweeping powers to detain US citizens labeled enemy combatants that the government was asserting the following spring. Title 18 U.S.C. § 4001(a) should be seen as a framework statute that requires a clear statement of later Congressional intent, enacted after a genuine effort to grapple with the citizen detention question. Broad enactments like the September 18 Authorization to Use Force almost surely are not what Congress had in mind in 1971, and they should not be deemed an adequate foundation in any event for the kind of intrusions on liberty we see in Padilla..

      Although many congressional debates today are bitter, overly partisan, and designed more for sound-bites than for enlightenment, I still think such a remand to Congress would be a worthwhile development. Congress's efforts, during the debates on the USA PATRIOT Act, to refine and narrow the far more limited form of executive detention of aliens ultimately enacted as INA § 236A, described above, revealed a healthy and serious attempt to come to grips with a difficult balancing problem. Those same debates strongly suggest that Congress, if somehow brought to engage these issues, would enact a far narrower detention power than the current sweeping "enemy combatant" doctrine propounded by the executive branch. If it takes up the issue in the citizen context, Congress should use the occasion to develop clearer standards and protections that would apply to Guantanamo-type detentions as well, at the very least addressing conditions of confinement, detainee rights to contact the outside world, and limits on coercive interrogation techniques.

      I would personally favor departing as little as possible from the classical protections as we gradually learn how to adapt the law of war to this new kind of conflict. In this light, persons subjected to preventive detention because the government believes them to be combatants should presumptively be entitled to POW status. If they are to be given less favorable treatment, it should come only after a tribunal has individually ruled that they transgressed the relevant rules that allow such a downgrade (cf. Article 5 of the Third Geneva Convention). Such status questions should be resolved promptly. And of course detainees could be tried for war crimes or other unprivileged offenses, before a military commission if that is the chosen course. But we should make sure that the commissions conform to the minimum standards laid out in the Geneva Conventions. Since President Bush first authorized the use of military commissions in November 2001, successive implementing orders and regulations have brought the current plan for such trials a long way toward meeting those criteria, although further issues remain, particularly regarding limits on defense counsel.

      Congress should also wrestle with the thoroughly vexing issue of the permissible length of preventive detention. Classic international wars have had a more readily discernible endpoint, marked by a surrender or armistice that brings the state of belligerency to an end and allows for mass release of POWs. The war against terrorism will probably be with us for decades and cannot be expected to find such a clean endpoint. Yet it would be wrong to countenance preventive detention without any limits. Some have suggested a fixed number of years as an outside boundary, and perhaps there are other mechanisms that would help deal with this question. But the issue should be confronted. In the long run, it would be best to proceed to enshrine new doctrine in new international treaties - as has often happened in the past based on experience gained in recent conflicts. We owe the existence of the 1949 Geneva Conventions to such a process. But I recognize that conditions now, in the wake of the Iraq conflict, are about as inauspicious as could be imagined for such an endeavor.

      The deeper need for a legislative check. There is another good reason to require deliberate congressional action before recognizing an authority in the President as potentially sweeping as the asserted power over enemy combatants. When, during the Korean War, President Truman ordered the seizure of the steel mills to avoid labor strife, the Supreme Court struck down his action, despite a plausible argument that avoiding a strike in this fashion would advance the war effort. It found insufficient authorization for this kind of executive action in congressional enactments. In his much-admired concurring opinion, Justice Jackson, a former Attorney General and chief prosecutor at Nuremberg, reflected broadly on the use of claimed inherent powers in the executive to respond to emergencies. He surveyed the practices of several other countries before concluding:

      This contemporary foreign experience may be inconclusive as to the wisdom of lodging emergency powers somewhere in a modern government. But it suggests that emergency powers are consistent with free government only when their control is lodged elsewhere than in the Executive who exercises them. That is the safeguard that would be nullified by our adoption of the 'inherent powers' formula.

      If this insight fits the executive seizure of domestic factories, it applies with even greater force to the domestic seizure of citizens.