September 16, 2004

September 16, 2004 | Legal Memorandum on Department of Homeland Security

Preventive Detention and Actionable Intelligence

Law enforcement and intelligence agencies often have solid, actionable intelligence about a terrorist threat, but the information is not of the sort sufficient to secure a conviction under existing criminal laws. In those circumstances, what should be our response? Right now, America has no coherent legal system in place to answer that question. One of the gaping holes in America's response to the terrorist attacks of September 11 is that we have not yet undertaken the difficult task of defining a legal regime in which actionable intelligence may be acted upon.

Rather, our response relies on a patchwork of laws that pre-date September 11 and that were designed with other goals in mind--laws relating to material witnesses, immigration, and enemy combatants in a traditional war between nations. And that response has been only partially successful in the courts. We cannot afford to continue with this inadequate and haphazard legal system. America needs to develop a coherent structure of laws and procedures to deal with the interaction of intelligence information and the judicial system.

America's situation is not unique. Other countries, most notably the United Kingdom, have faced similar terrorist threats and have developed legal structures that permit the government to act upon intelligence information (through preventive detention) while also ensuring (through burden of proof standards, procedural protections, and oversight) that civil liberties are not infringed upon. We should follow, generally, the English model in developing an American legal system that meets our contemporary needs.

The Criminal Justice System
and Actionable Intelligence:
A Fundamental Disconnect

To see the need more clearly, consider two recent real-world examples: New Jersey police observed a man who appeared to be of Middle Eastern descent, dressed in business clothes near a vital bridge. His actions, though innocent on their face, were somehow out of the ordinary and they aroused the suspicion of a police officer, who approached the man. It turned out that the man was wearing a wet suit beneath his business suit, and was carrying a knife and an underwater camera. In the second case, an individual approached a border station in Texas seeking entry into the United States. A check of the Terrorist Screening Center's watch list of suspected terrorists disclosed he was on the list.1

What then?2 In other words, what should we do if we identify individuals whose actions (or other information about them) gives us grounds for believing they are terrorists who seek to do us harm? The question is obviously quite serious, and the easy answer--to arrest them--is often not available. Indeed, the man in New Jersey was never detained, and the individual at the border was merely turned back from the entry point (and, incidentally, turned back at three more entry points over the next several days). Did the government do the right thing in those cases?

The hearings before the National Commission on Terrorist Attacks Upon the United States have made one thing clear: Prior to September 11, there were systematic problems that prevented appropriate coordination among the various intelligence agencies and between the intelligence community and law enforcement.3 Fixing that problem is a long and difficult task--one that America has already begun.

But even if we achieve a much better coordination structure that seamlessly integrates intelligence and law enforcement information, that would answer only half the problem.

As those involved in intelligence collection know, much intelligence information--even the best and most accurate information that is directly actionable--is not suitable for use in our existing criminal justice system. Consider just a few of the possibilities:4

  • Highly accurate information may have been provided by a foreign government, but only on the condition that the information never be publicly disclosed, or indeed, that the fact of the government's cooperation with us never be disclosed. Using this information, even in the controlled setting of a criminal trial governed by the procedures of the Classified Information Procedures Act (CIPA)5, would dry up the source of information and end its utility for all future events.
  • Information of unquestioned veracity might have been gathered through sources and methods that are not known to the public or to foreign powers and terrorists. Public disclosure that the information is in the possession of the United States would compromise the source or method and render it useless.6 There is some anecdotal evidence that this has already occurred. During the first World Trade Center bombing trial, the government disclosed that it had the capacity to intercept Osama bin Laden's satellite phone calls. It is reported that, naturally, he ceased using satellite phones.7
  • Evidentiary rules requiring disclosure of evidence can conflict with national security needs. One need only look at the difficulties created by the trial of Zacarias Moussaoui to recognize this problem. Criminal trial rules require that he have access to al-Qaeda operatives (reported to be Khalid Sheikh Mohammed and Ramsi Binalshibh) as a potential witnesses with allegedly favorable evidence.8 Yet allowing Moussaoui (or his lawyers) access to Mohammed and Binalshibh while they are still being interrogated would be a foolhardy compromise of vital intelligence assets.
  • The rules of evidence in a court of law strictly limit the admissibility of information. Documents and photographs must be authenticated. Hearsay is not allowed.9 Yet often the best, most useful intelligence information cannot meet these legal requirements. A stolen document (or one intercepted by electronic means) often cannot be authenticated. The individual who surreptitiously took a photograph might not be available in an American court to authenticate it. And sometimes the best oral intelligence ("At a meeting last week, Osama said . . .") is rankest hearsay.
  • Finally, one must confront the new reality posed by the "problem" of interrogation. Virtually every practitioner of interrogation maintains that one of the most successful means of productive interrogation is isolation. As the Fourth Circuit has acknowledged, there is substantial force to the general argument that interruption of the interrogation process could "have devastating effects on the ability to gather information" from those who have been captured.10 Of course, the inability to gather information of this sort could well result in the failure to prevent future terrorist attacks. However, isolation of a terrorist suspect--that is the denial of access to him by anyone--is inconsistent with existing practices, including rules relating to the provision of counsel.

These are but a few examples of the ways in which the intelligence-gathering function does not mesh with our conception of law enforcement and the legal system. One can readily imagine any other number of conflicts (disclosure of witness identities would endanger them; premature disclosure of the penetration of a terrorist network would cause the network to evaporate; etc.) in which the methods of intelligence, though perfectly adequate and acceptable in their own sphere, are unacceptable in the traditional criminal legal realm.

Moreover, not all intelligence information is as substantial as that required in the legal system. Certain intelligence may be enough to raise substantial suspicion (and, thus, warrant turning someone away at the border) but it may (and often does) fall far short of information that would establish someone's terrorist intent in a forum requiring proof beyond a reasonable doubt.

Thus the question: What do we do? If, indeed, U.S. law enforcement and intelligence agencies have solid actionable intelligence of a terrorist threat, and if that intelligence is sufficiently particularized that it allows the identification of an individual or group of individuals, what should be the response in those limited situations in which existing legal rules are not fully consistent with what seems to be the most prudent response? Commitment to the existing legal system may, as noted, be impossible to sustain without unacceptable risks and costs. But detention outside the existing legal structures is also unacceptable. What then?

The American Post-9/11 Response

Our lack of a coherent answer to this question explains some of the most problematic and controversial actions taken by our government after September 11. The use of material witness warrants; the questioning and arrest of Arab immigrants in the aftermath of 9/11; and the detention of "enemy combatants" are all responses to this conundrum. On the merits, these policies have plausible legal grounds. They have engendered much discussion.

Our concern here is not, however, with their legal justification. Rather it is that, examined objectively, the responses reflect a simple fact: We lack a settled, structured legal system for dealing with actionable intelligence. In other words, we lack a regularized process for preventive detention in lieu of criminal trial. As one observer has put it (albeit in a different context): "America was put off balance by September 11."11 This section outlines the temporizing responses the government has taken, which were mostly in the nature of trying to accommodate existing legal procedures to a new reality--in effect, squeezing a square peg into a round hole.

Material Witness Warrants

One of the first legal responses to the problem of actionable intelligence without a satisfactory legal structure was the use of material witness warrants.12 Material witness warrants traditionally functioned exactly as their name implies--they were used to arrest and detain material witnesses to criminal events when it was anticipated that the witnesses would flee the jurisdiction to avoid the obligation of giving testimony. Immediately after 9/11, a number of individuals suspected of involvement in terrorist activities were detained through the use of material witness warrants. Some detainees fit the traditional paradigm (at least in part) because they were held pending their testimony before a grand jury. But the traditional paradigm was broken when the detention of some witnesses continued for a time despite their willingness to give testimony and have it preserved. It was further broken, to the extent that it was a constraint, when one detainee (Jose Padilla) initially held on a material witness warrant, was remanded to military custody as an "enemy combatant." The material witness provisions became, in effect, a proxy for a preventive detention program.

Challenges to this practice were swift in coming. Those opposing the use of material witness warrants in this manner challenged their use in pre-trial investigations. Although the law had historically been construed to approve the use of material witness warrants for grand juries,13 at least one court initially held that the post-9/11 jailing of witnesses pending investigation was a constitutional violation.14 Eventually, however, the courts held that the material witness statute was applicable to pre-trial investigations, and that the government could lawfully detain a witness, notwithstanding the witness's willingness to have his or her testimony preserved, provided that the grounds for detention were subject to judicial review under the standard bail statutes.15

However, the approval of material witness warrants as a legal tool cannot obscure the practical reality that they were being used for a purpose different from that which Congress initially intended--the detention of witnesses despite the lack of any real need for their testimony. Because this legal structure is not designed to adjudicate issues now before it--the long-term detention of suspected terrorists rather than the short-term detention of potential witnesses--it lacks any number of legal checks on the exercise of executive authority. Judicial review is either standardless or based upon standards (such as the bail statute) designed for a different context. The burden of proof placed upon the government to justify its detention is ill-defined. There is only a limited opportunity for those detained to challenge their detention and secure the assistance of counsel. Moreover, because the use of material witness warrants occurs at the interstices of law and intelligence gathering, there is little, if any, sustained congressional oversight of the use of the power.

"Hold Until Cleared"

The policies instituted immediately after September 11 included the arrest of many Arab immigrants, and their detention, mostly on relatively minor immigration law violations.

In reviewing these detentions, the Inspector General of the Department of Justice concluded that approximately 762 aliens were arrested in connection with the investigation of the September 11 attacks. The vast majority of those arrested were detained pursuant to civil immigration law authority for remaining in the United States after expiration of their entry visas or for entering the country illegally.16 Thereafter, pursuant to government policy, each of the detainees was held in the United States until such time as they were "cleared" by the FBI of any suspicion that they were connected to terrorists.17

The Inspector General did not criticize this policy, vel non.18 Moreover, he acknowledged that in all but one instance the detainees arrested were held on valid immigration charges.19 It nonetheless remains the case that the implementation of the "hold until cleared" policy was over-inclusive and resulted in identifying far more immigration violations than individuals connected with terrorism.

Given the extremity of the times in which these detentions occurred (most arrests were in New York City or surrounding areas within three months of the attack) the reaction of government authorities was understandable. With the World Trade Center smoldering in lower Manhattan a robust response was appropriate. In the end, however, immigration law served as a substitute for terror detentions--a substitution made necessary by the absence of any other viable legal mechanism.

Enemy Combatants

Many of the same observations can be made regarding the far better known cases involving the detention of Yasser Hamdi and Jose Padilla as "enemy combatants." Hamdi and Padilla were both American citizens, one caught on the field of battle in Afghanistan and the other detained as he entered the United States at Chicago's O'Hare airport. The government designated both as enemy combatants subject to special detention rules. These included, at least initially, a prohibition on contact with counsel and a contention that the detention decision was subject to limited judicial review.

Once again, the government's conduct was a response to a genuine need. Although we need not take the government's concerns at unquestioned face value, there is certainly some factual basis for thinking that Padilla, for example, may have planned significant terrorist activities in the United States.20 Yet it is by no means clear that at the time of his initial detention, Padilla could have been criminally charged, or that if he had been charged, the case against him could have been sustained in a federal court under prevailing constitutional and evidentiary standards. Thus the conundrum recurs: how to treat those whom credible evidence suggests are engaged in terrorist activities, but for whom the legal system was not designed.

The Supreme Court has now concluded that such detentions must be subject to more rigorous review than that initially afforded by the government.21 In doing so, the Court recognized, that the capture and detention of those who would wage war against the United States was a universal historical practice. The Court recognized, however, that the "war" in which we are engaged has unusual characteristics--not the least of which is that some aspects of the war have no foreseeable termination. (Others, such as the conflict in Afghanistan, are more reasonably viewed as subject to closure.) In recognition of this, the Court determined that the Constitution required some form of process for reviewing detention claims, including notice of the factual basis for detention and an opportunity to rebut the factual assertion before a neutral decision maker. The Court left for another day more detailed questions about precisely what form this process would take, including questions about the admissibility of evidence, the burden of proof, and the identity of the decisionmaker.22

The Court's determination makes palpable the need for congressional action in this area. The question remains: Exactly what sort of process ought to be provided?

What is needed is a new legal architecture to govern the detention of suspected terrorists. Any such system must provide a legitimate process by which we can adjudge those suspected of terrorism as threats and a means of ensuring that the process is used appropriately, not abused. The challenge is an exceedingly difficult one--one that perhaps admits of no ideal solution. But the absence of a comprehensive, integrated legal structure in current law has led to an unsatisfactory ad hoc approach. It is likely that the detention powers described in this paper will be needed sometime in the future. It is certain, in our view, that the exercise of those powers is better constrained, and civil liberties better protected, when the exercise is regularized. Indeed, in the absence of any thoughtful effort to construct such a legal system before another terrorist attack on American soil, it is all too likely (indeed, almost inevitable) that the reaction will be similar, if not more extreme. Thus, the need for some new system of law appears palpable.23

English Legal Architecture for Detention

Fortunately, the problem is not new. The English response to Irish terrorism affords a useful model for the structure of an appropriate U.S. legal architecture. To see this, it is useful to begin before the advent of an English legal system of detention, when British detention policy was ad hoc and irregular--that is to say, when the English were in the same position with respect to Irish terrorists that America is in today with respect to al-Qaeda.

British Policy in Northern Ireland 1971-1972

In response to escalating violence from the Irish Republican Army, on August 9, 1971, British Prime Minister Edward Heath authorized a policy of detention (then called "internment") in Northern Ireland. In the months leading up to the decision, bombings occurred nearly twice daily, and the number of violent incidents continued at a high rate. One observer, Tony Geraghty, described the policy of internment as flawed, and noted that the violence only increased in 1972, including nearly four bombings per day.24

Once decided, implementation of the policy was left to lower level military personnel without adequate oversight from senior officers or political leaders. Interrogations were designed to produce tactical level information for battlefield commanders, and required that the prisoner's resistance be broken without leaving visible evidence. Hence, interrogators used such techniques as sensory deprivation, exhaustion, humiliation, and fear. In spite of the harsh interrogations, no prisoners died during internment in 1971. By December, nearly 1,600 people had been detained at three separate facilities, including the main interrogation center at Ballykelly Barracks, a former Royal Air Force base, although over 900 had been released. Some were physically abused, but most of the treatment was psychological. The European Court labeled some treatment as "degrading and inhumane."25

The process yielded significant short-term intelligence, which enabled a marked increase in the collection of arms and munitions, but the success was short-lived. Because of the techniques employed and resulting scandal, as Geraghty notes, the IRA received a boost in credibility. The British suffered from a lack of inside intelligence sources for a long time after, and were further hampered by limitations placed on interrogations. 26

Detention in England Today 27

Nonetheless, the English were faced with a quandary. Although the interrogations had been stopped and large-scale internment finished, there remained a small class of individuals for whom release was not a reasonable option. Initially, this perceived necessity applied almost exclusively to members of the Irish Republican Army. Today, the English law has been updated to include within its ambit other terrorists whose detention might be appropriate. In outline form, the English law provides:28

  • Upon certification by the Home Secretary, a suspected international terrorist may be detained;
  • That detention is subject to review by a Special Commission;
  • The detained individual is entitled to notice of the grounds for his detention;
  • The detainee is entitled to representation in the appeal of his detention from amongst a special cadre of lawyers cleared to handle classified information;
  • During the appeal, the detainee's lawyer may cross-examine witnesses and submit written argument to the Commission;
  • The Home Secretary appoints an independent reviewer to examine the operation of the detention laws and to review each individual case of detention.29

The Outline of An American Response

America needs a more thoughtful, comprehensive legal response to terrorism. It must "regain [its] balance"30 and adopt a system of laws that regulate and constrain executive behavior, while simultaneously allowing an effective response in those very narrow circumstances where a response is necessary, but the existing legal structures are inadequate. Here we outline what such new legal system might look like.31

Structure of a Proposed Law

First, the regime of preventative detention that we envision should be limited to cases of terrorism. This requires a narrow definition of terrorism--narrower than that used in current law. Detention will be appropriate only in situations involving individuals who:

  • act or threaten to act in a manner that involves serious violence against a person and/or in a manner that risks the health and safety of the public; and
  • does so to influence government policy or intimidate the public; and
  • does so for the purpose of advancing a political, religious, or ideological cause.

In other words, the "gate" for the preventative detention system must be very narrow.


  • There must be a rigorous certification process at the front end. No individual should be subject to detention unless the Attorney General first certifies as to its necessity. The certification should aver that credible evidence exists that: (a) the individual to be detained intends to commit a terrorist act; (b) the individual is affiliated with a terrorist organization; and (c) the existing criminal legal justice system cannot be applied to the individual without compromising national security.
  • This certification should be subject to review in court. The preferred method of review is in an adversarial process in which the detained individual is represented by counsel. To effectuate assignment of counsel for proceedings in which classified materials are considered, the government should create a group of pre-cleared defense counsel (perhaps as part of the Federal Public Defender system) available for assignment to the detained individual(s).
  • Proceedings to determine whether detention is appropriate would have the following components (as required by the Hamdi) decision:
  • Notice to the detainee of the factual basis for detention;
  • An opportunity to rebut the detention evidence;
  • A neutral decision maker. To regularize the process and insulate it from executive influence, we would recommend the creation of a new adjudicative court akin to the Foreign Intelligence Surveillance Court;
  • Evidentiary rules would be relaxed and procedures modified to account for the necessity of using classified information;
  • The government would bear the burden of proving the grounds for its detention decision by proof that meets the clear and convincing evidence standard.
  • In rare circumstances the government may have grounds for wishing to delay the detainee's appearance in court and access to counsel, so that ongoing interrogation can continue. The initial period of delay should be no more than 30 days and any extension should be periodically reviewed and justified (to the court) by a clear and convincing showing that a further delay is likely to provide additional intelligence.
  • To ensure that the preventive detention authority is not abused, there should be routine, systematic oversight. Review of the general process should, of course, reside with the congressional intelligence committees. Beyond that, as in the United Kingdom, each individual detention decision should be independently reviewed and the subject of a public report. Possible entities for conducting the review might include the President's Intelligence Oversight Board or the Inspector General of the CIA.


In calling for Congress to enact legislation concerning this matter, we are not alone.32 There are some who will disagree with making the effort at all, who will say that Americans should never allow preventive detention in any form because it is an unwarranted threat to liberty.

The response to this criticism is threefold: First, the criticism blinks reality. We already have incomplete and irregular forms of preventive detention because it is a necessity. We advance liberty when we regularize the practice, cabin it to narrow circumstances, and use it sparingly. Second, as detailed above, other countries (such as the United Kingdom) have managed to adopt very limited forms of preventive detention without becoming noticeably "unfree" or "authoritarian." Adoption of similar legal forms in the United States will not render us an authoritarian regime either.

Finally, and most important, to reject preventive detention in those rare circumstances in which it is necessary is to exalt liberty at the expense of security.33 The founding of the American Republic was for the purpose of constructing a political system of ordered liberty. It simply cannot be right to unilaterally prefer liberty. Liberty is not an absolute value; it depends upon security (both personal and national) for its exercise. As Thomas Powers has written: "In a liberal republic, liberty presupposes security; the point of security is liberty."34 The growth in danger from the consequences of the failure to stop terrorism necessitates altering our tolerance for governmental order. More fundamentally, our goal should be to maximize both order and liberty.

We do that best, not by closing our eyes to the necessity of security, nor by allowing security concerns to run rampant without oversight, but rather by taking appropriate steps to ensure that the powers given to the executive branch are exercised thoughtfully and with care, and subject to continue review and oversight by both the judiciary and the legislative branch. This concept of checks and balances was the fundamental insight of the Framers of the Constitution--and is as applicable today as it was at the time of the Founding.

Paul Rosenzweig is Senior Legal Research Fellow in the Center for Legal and Judicial Studies, at The Heritage Foundation and Adjunct Professor of Law at George Mason University. James Jay Carafano, Ph.D., is Senior Research Fellow in Defense and Homeland Security in the Kathryn and Shelby Cullom Davis Institute for International Studies at The Heritage Foundation. This paper will be part of a larger project of The Heritage Foundation, a book-length treatment of homeland security issues generally, entitled Winning the Long War, that is scheduled for publication in early 2005.


1. These stories were related by Gen. (Ret.) Matt Broderick, Director of the Homeland Security Operations Center at a briefing on May 20, 2004. He avers that they are true, recent events.

2. The question is borrowed from a provocative talk given by Judge Michael Chertoff (former Assistant Attorney General) to the ABA Standing Committee on Law and National Security in Washington, D.C., on April 13, 2004.

3. See The 9/11 Commission Report: Final Report of the National Commission on Terrorist Attacks Upon the United States, (2004). The USA Patriot Act eliminated many of these systematic problems. For a brief discussion of the emerging consensus as to the utility of the Patriot Act, see James Jay Carafano & Paul Rosenzweig, "A Patriotic Day: 9/11 Commission Recognizes Importance of Patriot Act," Heritage Foundation Web Memo No. 480, April 15, 2004 at For an in-depth treatment of various aspects of the Patriot Act, including the information sharing provisions, see Paul Rosenzweig, "Civil Liberties and Terrorism," 42 Duq. L. Rev. 663 (2004).

4. Here, too, we are indebted to Judge Chertoff's speech, in which he outlined some of these possibilities.

5. The Classified Information Procedures Act is codified at 18 U.S.C. App. II. CIPA is a procedural statute; it neither adds to nor detracts from the substantive rights of the defendant or the discovery obligations of the government. Rather, the procedure for making these determinations is different in that it balances the right of a criminal defendant with the right of the government to know in advance of a potential threat from a criminal prosecution to its national security. See, e.g., United States v. Anderson, 872 F.2d 1508, 1514 (11th Cir.), cert. denied, 493 U.S. 1004 (1989); United States v. Collins, 720 F.2d 1195, 1197 (11th Cir. 1983); United States v. Lopez-Lima, 738 F. Supp. 1404, 1407 (S.D.Fla. 1990). Each of CIPA's provisions is designed to achieve those dual goals: preventing unnecessary or inadvertent disclosures of classified information and advising the government of the national security "cost" of going forward.

6. See CIA v. Sims, 471 U.S. 159, 176 (1985) (noting potential for harm if identity of a source is revealed).

7. According to Anthony Blinken, a former National Security Council official now at the Center for Strategic and International Studies, the disclosure of bin Laden's use of cell phones "ended that channel of information." Tony Raum, "US Stingy on Bin Laden Evidence," Associated Press, October 30, 2001.

8. United States v. Moussaoui, 365 F.3d 292 (4th Cir. 2004).

9. See Fed. R. Evid. 801 (defining hearsay); 802 (prohibiting hearsay as evidence except in limited circumstances). See also Crawford v. Washington, -- U.S. --, 124 S.Ct. 1354 (2004) (prohibiting admission of testimonial hearsay, even if otherwise reliable).

10. Moussaoui, 365 F.2d at 306.

11. David Ignatius, "Small Comfort" Wa. Post A23 (June 15, 2004).

12. See 18 U.S.C. § 3144.

13. E.g. In re Bacon, 449 F.2d 933 (9th Cir. 191).

14. United States v. Awadallah, 202 F.Supp.2d 55, 202 F.Supp.2d 82 (S.D.N.Y. 2002), rev'd 349 F.3d 42 (2d Cir. 2003).

15. United States v. Awadallah, 349 F.3d 42 (2d Cir. 2003).

16. See Office of the Inspector General, "The September 11 Detainees: A Review of the Treatment of Aliens Held on Immigration Charges in Connection with the Investigation of the September 11 Attacks," at 34 (April 2003).

17. Id. at 37. The report also identified 84 instances of abuse at the Metropolitan Detention Center in Brooklyn. Id. at 118.

18. Id. at 70.

19. Id. at 15 and n.22.

20. See U.S. Dept. of Justice, Summary of Jose Padilla's Activities with Al-Qaeda (May 28, 2004) (transmitted to Congress June 1, 2004).

21. See Hamdi v. Rumsfeld, -- U.S. --, 124 S.Ct. 2633 (2004). Though Padilla's petition for release was dismissed on jurisdictional grounds, see Rumsfeld v. Padilla, -- U.S. --, 124 S.Ct. 2711 (2004), there is little doubt that his detention will be subject to standards similar to those applied to Hamdi.

22. See Hamdi, 124 S.Ct. 2646-52; see also John Yoo, "The Supreme Court Goes to War," Wall St. J. (June 30, 2004).

23. Indeed, the recently reported decision of the government to allow Hamdi's release, upon certain conditions, see Thomas Ricks & Jerry Markon, "US Nears Deal to Free Enemy Combatant Hamdi," Wa. Post at A2 (Aug. 12, 2004), only serves to emphasize the need for a regularized system of detention review. There is little doubt that, had the system envisioned in this paper been in place, Hamdi's detention would have been reviewed and resolved long before now.

24. Tony Geraghty, The Irish War: The Hidden Conflict Between the IRA and British Intelligence at 41 (Baltimore and London: The Johns Hopkins University Press, 1998, 2000).

25. Id. at 46-48.

26. Id. at 51.

27. We focus on England because of the commonality of experience and legal traditions. But the English are not unique. Preventive detention regimes (with widely varying legal structures and rules) can also be found, for example, in France (see, e.g.,, BBC News, "France's Top Anti-Terror Judge" (July 1, 2003) (available at, Germany (see German News, "Penal Law Tightened" (June 18, 2004) (available at, and at the International Criminal Tribunal--Yugoslavia (see Rule 65 of the Rules of Procedure and Evidence of the International Tribunal (allowing preventative detention because of "extreme gravity" of offenses)). A detailed survey of these laws is well beyond the scope of this paper.

28. The English laws are embodied in two acts, the Terrorism Act of 2000 (available at and the Anti-terrorism, Crime and Security Act of 2001 (available at The English system applies only to non-English citizens, but the procedures it adopts would appear to satisfy American constitutional standards under Hamdi. See Laura K. Donohoue, Counter-Terrorist Law and Emergency Powers in the United Kingdom, 1922-2000 (Dublin: Irish Academy Press, 2001).

29. Though the law does not require it, the Home Secretary has appointed a notable attorney who is not a member of the governing Labor Party. The independent reviewer is currently Lord Carlile of Berriew Q.C., a well known and well regarded Liberal Democrat. The law also provides for review of the anti-terrorism law, more broadly, by a Committee of Privy Counselors.

30. Ignatius, op. cit.

31. Some of our thinking in this has benefited from discussions with our colleague Robert Levy of the Cato Institute.

32. See, e.g., Benjamin Wittes, "Enemy Americans," The Atlantic Monthly (July/August 2004); Stuart Taylor, "The Fragility of Our Freedoms in a Time of Terror," National Journal (May 5, 2004).

33. For a provocative reflection on the ethics of the constrained use of legal authority in a liberal democracy, see Michael Ignatieff, The Lesser Evil (Princeton Univ. Press. 2004).

34. Thomas Powers, "Can We Be Secure and Free?" The Public Interest (Spring 2003).

About the Author

Paul Rosenzweig
Edwin Meese III Center for Legal and Judicial Studies

James Jay Carafano, Ph.D. Vice President for the Kathryn and Shelby Cullom Davis Institute for National Security and Foreign Policy, and the E. W. Richardson Fellow