ńňđ. 5
(âńĺăî 23)



The concentration of counter-terrorism powers in the executive is a common theme in
this volume: see, for instance, Powell, Chapter 25 (expressing concerns about ‘dispropor-
tionate executive power’), and Monar, Chapter 20 (arguing that the ‘EU co-operation
framework has clearly contributed to the strengthening of the role of the executive
branches in the fight against terrorism’).
Section 83.01(1) of Canada’s Criminal Code (RSC 1985, c. C-46). See generally Roach,
Consequences for Canada, at 36–38.
See Roach, Chapter 23, in this volume at 512, referring to Canada’s Immigration and
Refugee Protection Act, SC 2001 c. 27.
Internal Security Act, (Cap 143, 1985 Rev Ed Sing); Internal Security Act 1960 (Act 82,
Laws of Malaysia).
See Military Order on Detention, Treatment and Trial of Non-Citizens in the War Against
Terrorism (George W. Bush, November 13/01). Note, however, that the decision of the US
Supreme Court in Rasul v. Bush, 124 S Ct 2686 (2004), may well have opened the door to
greater judicial oversight of executive detentions by the United States at its naval base in
Guantanamo Bay, Cuba, of foreign nationals captured during the US campaign against
al Qaeda and the Taliban regime in Afghanistan.

security intelligence community and other security experts within the
bureaucracy, is in a much better position to assess and respond to the risk
of terrorism than the public, the legislature or the judiciary. When it comes to
risk assessment, experts, particularly in their area of expertise, are more likely
than ordinary people to be right.45
But experts are not immune from error and the nature of risk assessment is
hardly an exact enterprise. When, as the aftermath of September 11 and the
American-led invasion of Iraq demonstrate, expert evidence is interpreted by
the executive through a political lens, it is hardly surprising that serious risk-
assessment mistakes result.46 We ought therefore to be sceptical of claims that
executive expertise per se is enough to justify our automatic deference to the
executive in matters of national security. This is all the more so given the
problem of transparency. In the realm of national security, executive power is
itself often shielded not only from full pubic scrutiny, but from judicial
scrutiny as well.47 And expert advice, particularly advice based on secret
intelligence-gathering and which forms the basis of executive judgments
about risk, is even less accessible to the public. But a significant possibility
of abuse of power arises when the executive and its intelligence-gathering
branches are able to operate absent some form of direct oversight of their

B. Arguments from trust
A second, related argument concerns the need for trust. In a complex
modern society, much of the business of regulation needs to be done by a
professional bureaucracy and effective governance requires that the

See above note 26.
Some recent examples include the failure of American forces in Iraq, well over a year after
the invasion, to discover the weapons of mass destruction that provided the ostensible
justification for the invasion (see generally, ‘Ex-U.N. Inspector Has Harsh Words for
Bush’, The New York Times (16 March 2004), A3, on the charge by the former chief United
Nations weapons inspector, Hans Blix, that ‘the Bush administration convinced itself of
the existence of banned weapons based on dubious findings before invading Iraq and was
not interested in hearing evidence to the contrary’); the claim prior to the invasion made
by British Prime Minister Tony Blair to the effect that Iraq had the power to deploy its
weapons within 45 minutes (see Patrick E. Tyler, ‘Threats and Responses: An Assessment:
Britain’s Case: Iraqi Program to Amass Arms Is ‘‘Up and Running’’ ’ The New York Times
(25 September 2002), A12; and the allegation by Richard A. Clarke that the Bush adminis-
tration had failed to take heed of warnings by its own intelligence experts prior to
September 11 that the threat of terrorism had to be taken seriously: above, note 30.
See Young, Chapter 18, criticizing the lack of consultation and transparency in the process
of drafting security laws in Hong Kong.
Philip B. Heymann, Terrorism, Freedom, and Security (Cambridge, MIT Press, 2003),

bureaucracy be trusted by the public and by the other branches of govern-
ment.49 While constitutional checks might theoretically be available, checks
on the bureaucracy must also come from an internal ethic – the profession-
alism and integrity of those within it, and their respect for the rules that
govern them. A bureaucracy that exhibits this sort of professionalism and
integrity will command the trust of the public, making external checks less
The argument for trust in government can be pitched at a very general
level, but a cultural particularist version of this argument has also been made
in Singapore. It is argued, for instance, that the Confucian idea of govern-
ment by ‘honourable men’ (junzi) plays an important role in public life, in
contrast with the ‘Western idea that a government should be given as limited
powers as possible, and should always be treated with suspicion unless proved
otherwise’.50 But the public reaction to 9/11 suggests that trust in government
might be as circumstantial as it is cultural. Trust in government in the United
States rose markedly in the weeks and months after 11 September 2001, with
double the number of people (at 64 per cent) saying they ‘trusted the
government to do what is right most of the time or virtually always’ than in
the spring of 2000.51
Whether the call for trust in government is circumstantial or cultural, the
more general problem here is that even a once trustworthy government can
subsequently (or secretly) abuse its power and when such a change occurs,
laws authorizing greater flexibility and deference are no longer able to check
the abuse. Indeed, despite its reliance on the idea of government by honour-
able men, the Singapore government has, in recognition of the possibility of
some future abuse of governmental power, created additional institutional
safeguards52 to prevent abuse. Moreover, even in a trustworthy, thoroughly
professional, and competent bureaucracy, experts responsible for making
policy do make normative judgments (as noted earlier) and might even

While the plurality of the US Supreme Court in Hamdi v. Rumsfeld, 124 S Ct 2633 (2004)
did not insist on explicit Congressional authority for the detention of US citizens regarded
by the government as ‘enemy combatants’, it nevertheless acknowledged the need for
judicial oversight in national security cases to prevent abuse, arguing that ‘crucial as the
Government’s interest may be in detaining those who actually pose an immediate threat to
the national security of the United States during ongoing international conflict, history
and common sense teach us that an unchecked system of detention carries the potential to
become a means of oppression and abuse of others who do not present that sort of threat’
(at 2647).
White Paper on Shared Values (Singapore: 2 January 1991), Cmd. 1 of 1991 at para. 41.
David S. Broder, ‘The Trust Factor’, The Washington Post (3 October 2001), A31.
One example is the office of the ‘Elected President’ which has the power to safeguard,
among others, the use of government reserves: see generally Kevin Tan and Lam Peng Er
(eds.), Managing Political Change in Singapore: The Elected Presidency (New York,
Routledge, 1997).

display an ‘affiliation bias’,53 holding general views that are favourable to the
industry or organization with which they are affiliated.

C. Arguments from extraordinary circumstances
But even if executive expertise should ordinarily be carefully circumscribed, it
might be argued that in extraordinary circumstances greater deference to the
executive and its professional bureaucracy is needed. The stakes may well be so
high that ordinary rules simply do not apply; constitutional rights would have to
yield to security concerns as the very survival of the state is at stake. One version
of this argument, advanced by Oren Gross, is that in the immediate aftermath of
an emergency, the executive will often act illegally or extra-constitutionally. In
these circumstances, the conduct of the executive cannot properly be judged by
ordinary constitutional standards, and ought instead to be subject to democratic
ratification after the fact.54 This argument is part of a sophisticated defence of
extra-legal measures; Gross argues that by containing these measures to truly
exceptional circumstances, and allowing after-the-fact democratic ratification,
we can avoid diluting normal constitutional principles and uphold the rule of
law. A full response to this argument is provided by David Dyzenhaus in this
volume,55 but suffice it to say that whatever the merits of Gross’s proposal as a
way of preserving the rule of law, the idea that through popular democratic
ratification we could exonerate extra-constitutional government action56 in
emergency situations is deeply problematic. The problems arising from risk
perception and trust suggest that the outcome of a ratification vote would turn
largely on how closely after the event the vote takes place and how emotionally
charged the atmosphere remains.

IV. Judical review and anti-terrorism policy
The argument thus far suggests that we need to be cautious about conferring
excessive power over anti-terrorism policy-making to the legislative and

Nancy Kraus et al., ‘Intuitive Toxicology: Expert and Lay Judgments of Chemical Risks’ in
Slovic, The Perception of Risk, 285–315 at 311 (observing that ‘toxicologists working for
industry see chemicals as more benign than do their counterparts in academia and
Oren Gross, ‘Chaos and Rules: Should Responses to Violent Crises Always Be
Constitutional?’ (2003) 112 Yale Law Journal 1011, at 1111–15.
Chapter 4.
One difficulty with Gross’s argument concerns the very idea of justifying extra-constitu-
tional measures. If we accept Gross’s argument, it is not entirely clear what we are saying
about the legality of illegal government conduct that is subsequently ratified. If, as Gross
seems to suggest, it then becomes legally unobjectionable, his theory might then be seen as
an attempt to articulate higher order constitutional norms as to when a government, faced
with an emergency, can depart from lower order or prima facie constitutional norms.

executive branches. The argument is not that these branches of government
should be denied a say, only that they should not have the final say as to what
sorts of anti-terrorism policies should be adopted. The dysfunctional nature
of populist democracy is especially pronounced in a fearful and emotionally
charged atmosphere in which judgments about risk are likely to be distorted,
resulting in ill-conceived and hastily enacted laws that unnecessarily restrict
individual freedom. And, of course, executive expertise in matters of national
security remains fallible. Yet arguments persist in favour of insulating the
executive from judicial review in times of crisis. In particular, it may be
argued that the judiciary lacks the necessary institutional competence to
provide a legitimate safeguard in matters of national security and, in any
event, that national security cases involve sensitive intelligence information
that cannot be disclosed.

A. Objections to judicial review
A common argument against judicial review is that the judiciary has less
knowledge and expertise than the executive with respect to national security
matters. The purported institutional incompetence of the courts figured
prominently in the famous wartime preventive detention case of Liversidge
v. Anderson, in which the House of Lords deferred to the subjective discretion
of the Secretary of State on questions of national security.57 According to
Lord Wright: ‘In my judgment, a court of law could not have before it the
information on which the Secretary of State acts, still less the background of
statecraft and national policy which is what must determine the action which
he takes on it.’58 A similar approach can be seen in Justice Thomas’s dissent in
Hamdi v. Rumsfeld, on the issue of whether the courts should have the power
to review the executive’s decision to detain without trial a suspected ‘enemy
combatant’. For Justice Thomas the ‘question of whether Hamdi is an enemy
combatant is ‘‘of a kind for which the Judiciary has neither aptitude, facilities
nor responsibility’’ ’.59 This sort of hands-off approach defers both to the

[1942] AC 206 (HL). See also Lord Romer at 281: ‘There are (among others) acts
prejudicial to the public safety or the defence of the realm committed in a time of grave
national danger and of such a nature that ‘‘by reason thereof’’ it is necessary to exercise
control over the person suspected of committing them. Whether or not the acts of some
individual appear to be of this description is a question which the Secretary of State must
plainly be a better judge of than any court of law.’
Ibid. at 267.
124 S Ct 2633 (2004) at 2678, quoting Chicago & Southern Air Lines v. Waterman S. S. Corp.,
333 US 103 at 111. See also 2676: ‘Congress, to be sure, has a substantial and essential role in
both foreign affairs and national security. But it is crucial to recognize that judicial inter-
ference in these domains destroys the purpose of vesting primary responsibility [to protect
the national security and to conduct the Nation’s foreign relations] in a unitary Executive’.

executive’s empirical assessment of the extent of the risk and to its determin-
ation of whether that risk is sufficiently grave as to deny ordinary due process
and restrict individual freedom.
The second argument against judicial review is that there are legitimate
concerns about the sensitivity of intelligence information, particularly in
times of crisis. For instance, in Liversidge, Viscount Maugham stressed the
importance of protecting confidential sources,60 while Lord Romer expressed
concerns about leakage to the enemy, even in the course of an in camera
hearing.61 Moreover, when it comes to intelligence-gathering activities, the
willingness of an intelligence agency to provide information required by an
oversight body that monitors its activities ‘depends on providing a very high
measure of assurance that a body whose independence the public can trust
will not make secret information public in the course of receiving secret
reports, calling hearings, and inquiring about complaints.’62 Similar concerns
arise as to the disclosure of secret information to the courts on judicial review.
These are compelling arguments against judicial review, but they are not
insurmountable, once we recognize the judiciary’s institutionally unique and
important area of competence and consider imaginative ways of designing
our institutions63 to address these concerns.

B. The unique role of judicial review
Even allowing that the executive has a special expertise in assessing the magnitude
of the risk of terrorism because of its privileged access to intelligence inform-
ation, it still has no special expertise in measuring the risk of terrorism against
state incursions on fundamental, but intangible public values, such as liberty.
It is important, then, to distinguish clearly between two sorts of questions –
questions about the nature and extent of the risk, in which experts may well
have a special expertise, and questions about what ought to be done given
the magnitude of the risk, which involves a normative judgment in respect
of which the executive, even with its security experts, holds no special
expertise.64 Clearly, our response to normative questions will influence our
answer to quantitative questions about risk, and this would be so even for

Liversidge at 241. See also David Dyzenhaus, ‘Intimations of Legality Amid the Clash of
Arms’ (2004) International Journal of Constitutional Law 244 at 254.
Liversidge, at 280.
Heymann, Terrorism, Freedom, and Security, at 153.
I borrow this turn of phrase from Dyzenhaus, Chapter 4, in this volume.
The executive might well give insufficient weight to the rights or interests of minorities, for
instance, and, as Davis suggests, enforce an anti-terrorism regime irrationally or arbitrarily
by using it ‘against some of the most vulnerable members of society such as members of
racial or ethnic minorities’ (Chapter 9, in this volume, at 197).

experts.65 But to the extent that we could hold these two sorts of judgment
apart, experts would be able to assist the executive only to determine the extent
of the risk; their expertise would not extend to the more complex, normative
In this respect, the judiciary holds a comparative advantage over the other
branches. While judges might personally be susceptible to the same social
phenomena that amplify the perception of risk, they remain in a unique
institutional position to distinguish empirical questions about the extent of
the risk from the normative questions that hang in the balance. The courts
can provide a forum in which empirical evidence is presented and policies
that interfere with individual freedoms are defended and justified; they serve
as an institutional safeguard against policy-making motivated primarily by
public fear. But crucially, judicial review allows for a careful and sober
consideration of risks and responses in a forum in which limitations on
liberty are given their due and where the effectiveness of anti-terrorism
policies can be assessed against its impact on fundamental freedoms.

C. Specialized administrative tribunals and the role of the courts
But even if the courts are competent in relation to the complex normative
questions that arise in national security cases, and have the ability to measure
security concerns against normative values, they still do not have knowledge
or expertise in relation to security intelligence. And the concern about
the sensitive nature of intelligence information remains unaddressed.
Dyzenhaus’s proposal in this volume may be useful in helping to answer
these concerns. He proposes a model of legality which maintains the rule of
law within an institutionally flexible system. Modern common law, he
explains, provides us with the tools to recognize the importance of specialized
administrative tribunals within a legal order committed to the rule of law,
allowing the courts to play a more detached, supervisory role. An institution
such as the Special Immigration Appeals Commission,66 he argues, can
answer many of the objections to judicial review, including concerns about
institutional competence and sensitive intelligence information.67
Curiously, a similar conclusion was reached in Singapore from a very
different starting point. In the aftermath of Chng Suan Tze, which (until
quickly reversed by legislative and constitutional amendment) asserted the
power of judicial review over preventive detention under the Internal

Slovic, ‘Trust, Emotion, Sex, Politics and Science: Surveying the Risk-assessment
Battlefield’ in Slovic, The Perception of Risk, 399–412.
On the powers of and problems with SIAC, see generally in this volume: Fenwick and
Phillipson (Chapter 21); Harvey (Chapter 8).
Dyzenhaus, ‘Intimations of Legality’, at 260.

Security Act,68 Sin Boon Ann criticized the courts for not deferring suffi-
ciently to the executive.69 The arguments that he made are the now familiar
ones concerning, for instance, the expertise of the executive at intelligence
gathering and the need for executive discretion.70 Yet despite his stinging
criticism of the judiciary, Sin comes to a similar conclusion to Dyzenhaus,
arguing that what is needed is an independent advisory board, ‘empowered to
make political judgments’,71 which could take a ‘more active role in making
inquiries like reserving the right to determine whether the Minister’s decision
to withhold information on grounds of national security has any valid
Dyzenhaus is right to stress the importance of imaginative institutional
design. We should not be constrained by an overly formal understanding of
the separation of powers if this means that only the courts can check executive
power in an emergency. And Sin may well be right to stress that emergencies
require a unique set of executive powers and distinct legal response.
A specialized administrative tribunal could scrutinize executive actions in a
manner that takes into account the sensitivity of intelligence information
and, if carefully constituted, can respond to concerns about institutional
For a specialized administrative tribunal to protect sensitive intelligence
information and sources while preserving legality, it must be both indepen-
dent and effective. It must be able to scrutinize the executive’s arguments and
have the power, as needed, to reverse executive decisions. The independence
that is required here is both institutional and substantive. It is crucial that the
reviewing body is independent in the sense that it is institutionally separate
such that the usual standards of independence (such as security of tenure and
remuneration) are present. But it is also imperative that the institution also
have a distinct institutional culture. It must be able to view the matter through
a different lens, looking not only at the magnitude of the risk, but also the
values at stake, so as to counter-balance an executive institutional culture
dominated by a security imperative with its particular understanding of
risk.73 Subject to these caveats, a specialized administrative tribunal may well
be able to scrutinize the factual and legal basis of executive decisions in
individual cases (e.g. in the context of preventive detention) in a manner that
preserves the confidentiality of intelligence sources, with the appropriate level
of expertise in security matters.

See Hor, Chapter 13, in this volume.
‘Judges and Administrative Discretion – A Look at Chng Suan Tze v. Minister of Home
Affairs’ [1989] 2 MLJ ci.
Ibid. at ciii, cvi. 71 Ibid. at cvii. 72 Ibid.
See discussion of ‘affiliation bias’ in Kraus, ‘Intuitive Toxicology’, at 311.

Be that as it may, the courts still have an important role to play. This can be
seen by distinguishing two ways in which the courts or tribunals can provide a
check on executive power. First, as we have just seen, a specialized adminis-
trative tribunal might scrutinize the factual and legal basis of executive
decisions in individual cases, with access to the sensitive intelligence inform-
ation on the basis of which those decisions are made. Here, a specialized
administrative tribunal plays an important role in scrutinizing the exercise of
executive power in an emergency. Second, the ordinary courts can never-
theless provide a public forum for contesting the broader policy decisions of
both the legislative and executive branches in as much as they purport to be
weighing rights against security considerations. Neither the legislature nor
the executive has an institutional expertise in rights and the courts can play an
important role here in compelling the government to be forthright about its
assumptions about risk.
The opinions of Justice O’Connor and Justice Souter in Hamdi implicitly
recognize the role that specialized administrative tribunals can play in check-
ing executive power. For instance, neither judge was satisfied with a mere
declaration by the executive that Hamdi was an enemy combatant who could
therefore be detained without trial, and saw a need for some form of scrutiny
of the decision to detain him:
Often the executive will argue that ‘security considerations’ led to a govern-
ment action and request that the court be satisfied with this argument. Such
a request should not be granted. ‘Security considerations’ are not magic
words. The court must insist on learning the specific security consider-
ations that prompted the government’s actions. The court must be per-
suaded that these considerations actually motivated the government’s
actions and were not merely pretextual. Finally, the court must be con-
vinced that the security measures adopted were the available measures least
damaging to human rights.74

A detainee must be given a fair opportunity to contest the factual basis for the
detention before the courts. But as William C. Banks observes, ‘the neutral
decision maker prescribed by Justice O’Connor’s plurality opinion . . .
could be a military commission rather than a civilian court’.75 There are,
no doubt, reasons to be cautious here, but if it is designed carefully enough
within an institutional culture already committed to principles of legality, a
specialized administrative tribunal may well be able to address the need for
independent ‘judicial’ scrutiny.
While opening the door to specialized administrative tribunals, Hamdi also
signals an important role for the courts by insisting that they will scrutinize

Aharon Barak, ‘Forward: A Judge on Judging: The Role of a Supreme Court in a
Democracy’ (2002), 116 Harvard Law Review 16 at 157–8.
Chapter 22, in this volume, at 506.

the use of preventive detention as a general policy. By assuming this role, the
courts could well demand evidence both that measures damaging to constitu-
tional rights will be effective in reducing the risk76 and that any restriction on
rights is proportionate to the risk. By insisting on learning as much of the factual
basis for a particular policy as might be disclosed without compromising con-
fidential sources (which would be easier where a general policy, rather than an
individual instance of preventive detention, is in question), the courts can force a
careful examination of the gravity of the risk and the effectiveness of the
proposed countermeasures, as against the other values that might be sacrificed
in the hope of preventing the risk from materializing.
On this approach, the executive would have to convince the court both
that the measures in question, such as the authorized use of preventive
detention, were a rational response to the security threat, and that alternative
measures, less restrictive of individual freedom, were insufficient. The execu-
tive might be compelled to explain, for instance, why criminal proceedings
are markedly less adequate than preventive detention in preventing acts of
terrorism, particularly when some terrorist suspects (such as John Walker
Lindh, an American captured in Afghanistan while fighting with the Taliban)
are tried in the ordinary courts. Or it might be asked to explain why the
extraordinary powers it claims are needed when considered in light of the
wide range of policy options at its disposal.

D. Judicial review, democracy, and the limits of law
Critics of judicial review might still object to its anti-democratic nature.
However, it is crucial to distinguish between populist conceptions of demo-
cracy and other more complex notions of democracy designed to ensure fair
collective decision-making or, more ambitiously, to secure substantively just
laws. Judicial review may well encroach on democratic populism, but its goal
is to articulate and express the wishes of the people at the level of ‘the most
profound values of society in its progress through history’, not at the level of
‘passing vogues’.77 There is, of course, an element of judicial paternalism
here, but perhaps a ‘softer’ and more palatable kind of paternalism than a
courts-know-best variety, one that seeks only to ensure that public decisions

Banks argues in Chapter 22 of this volume that at least some of the successful terrorism-
related prosecutions in the United States post-September 11 were due not, as claimed by
President Bush, to the Patriot Act and expanded investigative powers, but rather to
information obtained ‘in the old-fashioned way – from an anonymous letter left at a
local FBI office’ (at 509).
As Justice Barak of the Supreme Court of Israel puts it, ‘the choice is not between the wishes
of the people and the wishes of the judge. The choice is between two levels of the wishes of the
people. The first, basic level reflects the most profound values of society in its progress
through history; the second ad hoc level reflects passing vogues’ (‘A Judge on Judging’, at 51).

are fully informed and considered.78 In this way, judicial review (or even the
prospect of judicial review)79 can serve as a check on democratic law-making
to ensure that in times of heightened emotion and widespread fear, a decision
to limit individual liberty is not lightly taken. The courts might not always be
able to prevent excessive responses to terrorist attacks, but their willingness to
scrutinize new laws may stimulate public debate and provide civil society
groups with the means to ask critical questions about anti-terrorism
policies.80 They might empower concerned citizens to ask more pointed
questions about the extent of the risk and remind them of the sacrifices
that they are being asked to make.
One final point is in order. It is tempting to turn to the law to solve our
most pressing problems. This is a tendency that we must resist. The law should
function in tandem with other measures – political, diplomatic, and otherwise –
to address those problems that most concern us. But while the law might not be
able to solve our problems, including the threat of terrorism, it might still play an
important role in ensuring that our legal response to particular problems is an
appropriate one. Of course, even in legal systems with a long-standing tradition
of judicial review, the courts have often in times of emergency deferred to the
executive on matters of national security. It takes a confident judiciary with a
strong culture of judicial review to challenge the concentration of power and the
implementation of reactive anti-terrorism policies. So it can hardly be expected
that the courts in legal systems that do not have such an institutional culture
would be in any position to take on the robust, supervisory role recommended in
this chapter.81 In the absence of judicial or even specialized administrative

See Gerald Dworkin, ‘Paternalism: Some Second Thoughts’ in Rolf Sartorius (ed.),
Paternalism (Minneapolis, University of Minnesota Press, 1983), 105–111 at 107, where
he defines ‘soft paternalism’ as the view that ‘(1) paternalism is sometimes justified, and
(2) it is a necessary condition for such justification that the person for whom we are acting
paternalistically is in some way incompetent’.
See Barak, ‘A Judge on Judging’, at 159: The impact of the court’s willingness to engage in
judicial review should be measured ‘not merely in the few cases brought before it, but also
in the many potential cases not brought before it, since governmental authorities are
aware of the court’s rulings and act accordingly’.
Roach, Consequences for Canada, at 202.
On this front, I am less optimistic than Dyzenhaus (see Chapter 4, note 67) about our
ability to shape perceptions of agency simply through the design of institutions. Having
the ‘right’ legal framework alone does not guarantee a robust approach to judicial review.
An institutional culture supportive of notions of legality and the rule of law may depend
on a host of other factors including the educational background of the judges, their social
status and influence, and their relationship to the political elite. So even if Dyzenhaus is
right that respect for the rule of law plays an increasingly important role even in ‘well-
ordered societies’, it may well be that the number of societies that have the deep institu-
tional capacity to respect and enforce the rule of law is fewer than we might think. There is
enough evidence in this volume alone of societies past and present whose legal institutions
are weak to suggest that the law is not enough.

review, other institutions, such as civil society groups, academics, and indepen-
dent experts, would have to step in to play a supervisory role. Without the firm
backing of the courts, however, they would find it all the more difficult to
encourage public deliberation and influence policy.

V. Conclusion
At a time when anti-terrorism policy is high on the public agenda, the
literature on perception of risk serves as an important reminder that the
fear inspired by terrorism might far exceed the danger it poses, especially
relative to the dangers we face in everyday life. In our efforts to be vigilant, we
must keep the threat of terrorism in perspective. When we step into a car or
board a plane, when we buy our next meal or take medication, we live with
and manage these risks. Likewise, the risks of terrorist acts need to be
managed, and a level of tolerable risk must be set. The real challenge is not
to eliminate the risk, but to manage and learn to live with it, confident that
what might reasonably be done (and no more) is being done. We do need to
be conscious of the limits of the law. But the courts can still provide an
important institutional role in times of anxiety and fear to ensure that a
dispassionate assessment of risk takes place and that the response to the threat
of terrorism is a measured one.

A Comparative Study of Anti-Terrorism Measures

The criminal law and terrorism

Many societies instinctively and quickly reach for the criminal law as a
response to terrorism. The criminal law has frequently been expanded as a
direct response to acts of terrorism. In the first part of this chapter, I will
provide an overview of how new criminal laws have often been produced in
response to terrorism and relate this to narrative, memorial, and commu-
nicative uses of the criminal law and increased concern about the rights of
victims and potential victims of crime. New anti-terrorism laws, however, are
not solely based on a symbolic focus on state punishment and denunciation.
They also incorporate a more modern approach that sees crime as one of the
many risks of modern society. The risk of crime can only be managed by
enlisting non-state actors in the crime control enterprise.
In the second part of the chapter, I will examine some of the dangers of the
criminal law solution to terrorism from both instrumental and normative
dimensions. The enactment of new criminal laws after acts of terrorism
implies that the existing criminal law was inadequate to respond to acts of
terrorism. The accuracy of such a claim, however, depends on the baseline
established by the ordinary criminal law in each particular jurisdiction. In
many jurisdictions, ordinary crimes with respect to attempts, conspiracy and
accomplice liability for crimes such as murder and bombings could already be
applied to terrorists should they be apprehended by police and intelligence
officials. In such countries, criminal law reform after acts of terrorism may
falsely suggest that the criminal law, as opposed to its enforcement, was to
blame for the failure to stop the terrorism. Another danger is that repressive
criminal laws may backfire by provoking successful due process challenges
and/or widespread resentment of the law. A normative danger is the distort-
ing influence that strong anti-terrorism laws may have on the criminal law.
The incursions on criminal law principles may intensify should new criminal
laws fail to stop terrorism and should other serious crimes command the
same public concern as terrorism. Broad definitions of terrorism, as well as

I thank William Banks, David Dyzenhaus, Trish McMahon and Victor V. Ramraj for helpful
comments on an earlier draft.


claims that tough new anti-terrorism laws are consistent with rights protec-
tion instruments, may also encourage the spread of incursions on criminal
law principles beyond the terrorism context. Perhaps the ultimate corruption
of the criminal law are wrongful convictions such as occurred in terrorism
cases in the United Kingdom in the 1970s.
Although it is relatively easy to be critical and even cynical about reliance
on criminal law as a response to terrorism, the virtues of the criminal law
should not be ignored. They emerge when comparisons are made to a some-
times lawless and no-holds barred war that has, at times, been waged against
terrorism. Although principles of individual responsibility, legality and due
process have been stretched by many new anti-terrorism laws, these princi-
ples have also exercised an important restraining influence. In the third part
of this paper, I will examine some of the restraints of the criminal law as
compared to the use of war, assassination, extraordinary rendition, torture,
preventive detention and immigration law. These less restrained responses are
inspired by the idea that many societies cannot afford to rely on the legalities
of the criminal law and its costly and public emphasis on establishing
individual fault in order to incapacitate or deter suspected terrorists. The
less restrained alternatives to the criminal law relate to what Oren Gross
has argued is an extra-legal approach to terrorism1 and to what Lord Steyn
has called a ‘black hole’ approach that abandons the restraint of legality
altogether.2 Although such approaches may avoid corrupting the criminal
law with attempts to accommodate the threat of terrorism, they also bring
anti-terrorism policies uncomfortably close to the techniques used by terrorists.
As Laura Donohue suggests in her contribution to this volume, the imposition
of violence and fear without regard to guilt or innocence and in order to
impress a larger audience are key features of terrorism and, regrettably, of
some anti-terrorism measures.3
In the final part, I will begin the task of situating the criminal law in a more
comprehensive and multi-faceted anti-terrorism strategy. As in the second
part of this paper, there will be both instrumental and normative dimensions
to my argument. Normatively, I share the uneasiness expressed by David
Dyzenhaus in this volume about the less restrained and extra-legal
approaches that have been used in the war against terrorism.4 The criminal
law with its emphasis on the presumption of innocence, no detention without
charges being laid, proof of guilt beyond a reasonable doubt, and the right to
full answer and defence is the repository of many values associated with the

Oren Gross ‘Chaos and Rules: Should Responses to Violent Crises Always Be
Constitutional’ (2003) 112 Yale Law Journal 1011.
Johan Steyn, ‘Guantanamo Bay: The Legal Black Hole’ (2004) 53 International and
Comparative Law Quarterly, 1.
Donohue, Chapter 2, in this volume. 4 Dyzenhaus, Chapter 4, in this volume.

rule of law. At the same time, however, a rule of law approach, especially one
based on the criminal law, leaves one open to criticisms of being naıve about
the threat of terrorism since 9/11. The answer in my view is to construct a
comprehensive anti-terrorism policy that employs much more than the
criminal law. Drawing on work in the fields of public health, technology
and crime prevention, I will suggest that administrative and private sector
strategies can play an important role in making it more difficult for terrorists
to have access to sites and substances vulnerable to terrorism and can take
steps to limit the harms of terrorism. There may also be a role for harder
strategies including incapacitation of terrorists and perhaps even war. At the
same time, escalating force against terrorism should be matched with corres-
ponding systems of legality. If war must be employed, it should be war that
respects the law of war. This is necessary to preserve the vital distinction
between terrorism and anti-terrorism efforts and to avoid the moral danger
of legal black holes in which no rules apply.

I. New criminal laws against terrorism
There is a long history of new criminal laws being enacted as a direct response
to horrific acts of terrorism. For example, new criminal laws were enacted in
the United Kingdom as a response to the IRA bombings in Birmingham in
1974 and in Omagh in 1998.5 In the United States, new terrorism offences
were created in reaction to the murder of Leon Klinghoffer upon the Achille
Lauro when it was hijacked by the PLO in 1985 and in response to the
bombing of the World Trade Centre in 1993 and the Oklahoma City bombing
in 1995.6 One danger of reactive legislation is that there may often be
inadequate time for debate either in the legislature or in civil society about
the proposed measures. For example, there was only seventeen hours of
legislative debate before the draconian Prevention of Terrorism (Temporary
Provisions) Act 1974 was enacted after the Birmingham bombings killed 21
and injured another 180 people. The dangers to civil liberties and general
principles of criminal law and legality may be particularly great when new
anti-terrorism laws are enacted as a direct and immediate response to terrible
acts of terrorism.7

Philip Thomas, ‘Emergency Terrorist Legislation’ (1998) Journal of Civil Liberties 240;
Philip Thomas, ‘September 11 and Good Governance’ (2002) 53 Northern Ireland Law
Quarterly 366.
Laurie McQuade, ‘Tragedy as a Catalyst for Reform: The American Way?’ (1996) 11
Connecticut Journal of International Law 325.
Note, however, that even anti-terrorist legislation that is not enacted as a quick response to
terrorism and in a state of crisis or emergency may still threaten such values. See for
example the United Kingdom’s Terrorism Act 2000 examined in Fenwick and Phillipson,
Chapter 21, in this volume.

Reactive law reform that occurs quickly after acts of terrorism often
proceeds without a full understanding of why the terrorists succeeded. In
the absence of such an understanding, it is often tempting to conclude that
new offences and new investigative powers could prevent terrorism. There
may also be a tendency to ‘fight the last war’ by focusing on the means used by
the successful terrorists or even a particular terrorist organization. Reactive
law reform in response to terrorism may not only threaten legal principles
and civil liberties, but also produce high profile laws that fail to use all the
policy instruments that can be used against terrorism.
Governments often experience pressure to enact new criminal laws after
well publicized and shocking crimes. In North America this has led to a new
memorial style of law-making where new laws are at times named after
prominent victims of crime and crafted in response to the powerful narratives
of horrific crimes. New criminal laws have become something of the ultimate
tribute to crime victims. As David Garland has observed ‘the new political
imperative is that victims must be protected, their voices must be heard, their
memory honoured, their anger expressed, their fears addressed’. This leads to
the enactment of new criminal laws as ‘retaliatory measures’ designed to be
cathartic denunciations of crime and ‘to respond with immediate effect to
public outrage’.8 The attempt to mould laws to particular crimes places
considerable strain on general and rational principles of criminal law. The
nature of terrorism as a crime that targets innocent people and is designed to
maximize fear in the general populace makes it particularly likely to produce
reactive and emotional criminal law reform.
Reactive anti-terrorism laws fit into a pattern that Jonathan Simon has
coined ‘governing through crime’. Writing in the American context, Simon
observed that even as crime rates decrease, media and governments have been
drawn to ‘crime as the preferred metaphor for all forms of social anxiety and
highlighted acts of punishment or retribution as the primary way of resolving
disputes of almost any kind’.9 I have suggested elsewhere that the focus on
crime has led to a ‘criminalization of politics’ in which criminal law reform
has been offered as a symbolic and relatively cheap response to a broad range
of social, economic and cultural problems.10
The new emphasis on the criminal sanction and retributive punishment is
also tied to a sense that ‘nothing works’ except perhaps punishment. There is
an impatience with strategies that address the broader causes and

David Garland, The Culture of Control: Crime and Social Order in Contemporary Society
(University of Chicago Press, 2001), at 11, 133–4.
Jonathan Simon, ‘Governing through Crime’ in L. M. Friedman and G. Fisher (eds.), The
Crime Conundrum: Essays on Criminal Justice (New York, Westview Press, 1997), at 173.
Kent Roach, Due Process and Victims’ Rights (University of Toronto Press, 1999), at

determinants of crimes or that compare the harms of crime with other threats
to human security such as disease and poverty. This process also fits into a
retrenchment from the welfare state and increasing privatization of many
services. In such an environment, criminal law becomes an even more
important focus of formal state activity. It becomes the primary means for
the neo-liberal state to be seen as doing something about social problems and
to communicate society’s disapproval of crime and its solidarity with victims.
The tendencies to use criminal law as a means both to govern and to
control risks are magnified in the terrorism context. Terrorism is designed
to cause maximum fear and damage to innocent victims. This is turn creates
enormous pressures on governments to act. Despite having enacted a com-
prehensive and quite severe anti-terrorism law the year before 9/11, the
United Kingdom joined countries throughout the world in enacting another
new anti-terrorism law shortly after 9/11. The government entered into a
formal derogation of rights under the European Convention demonstrating
what Garland observed before September 11 was a decreasing concern about
‘the risk of unconstrained state authorities, of arbitrary power and the viola-
tion of civil liberties’ by a public that was ‘decidedly risk-averse’ and in
‘a perpetual sense of crisis’ on issues of crime.11
The worldwide expansion of anti-terrorism laws in the wake of September 11
was facilitated by Resolution 1373 of the United Nations Security Council
which required all states to ensure that terrorist acts, including the financing of
terrorism, ‘are established as serious criminal offences in domestic laws and
regulations and that the punishment duly reflects the seriousness of such
terrorist acts’.12 The bulk of this resolution enacted under mandatory provi-
sions in the United Nations Charter contemplated criminalization and punish-
ment as the primary response to terrorism.13 It facilitated the pattern of
reactive law reform by calling for countries to report back to the Counter-
Terrorism Committee within ninety days on the steps taken to comply with the
resolution. Some countries took this as a virtual deadline for enacting new anti-
terrorism laws. Domestic criminal law reform can be shaped by international
standards and bodies.
Blanket media coverage of 9/11 increased a sense of fear and anxiety that
demanded both punitive measures by the state and self-protective measures
by individuals. The quickest response came from the United States. The Patriot
Act was introduced into Congress on 23 October 2001. It was approved by the

Garland, The Culture of Control, at 12, 19.
UN Security Council Resolution 1373. See Kim Lane Scheppele, ‘Other People’s Patriot
Acts’ (2004) 50 Loyola Law Review 89 at 91–3.
The resolution placed great stress on criminalizing the financing of terrorism and also
contemplated cooperative exchanges of information and immigration law reforms as
secondary responses to terrorism.

House of Representatives by a vote of 357–66 and by the Senate in a 98-1 vote.
It was signed into law by President Bush on 26 October 2001.14 The Patriot Act
responded to the United Nations Resolution to ‘deny safe haven to those who
finance, plan, support or commit terrorist attacks, or provide safe havens’ with
a new criminal offence that punished with up to ten years’ imprisonment
‘whoever harbors or conceals any person who he knows, or has reasonable
grounds to believe, has committed or is about to commit’ a long list of offences
associated with terrorism. The crime of providing material support for terrorism,
which was first created in 1996 in the wake of the first World Trade Centre
and Oklahoma City bombings, was broadened to include the provision of
monetary instruments and ‘expert advice and assistance’ to terrorists groups.
The maximum penalty for this offence was increased from ten to fifteen years
with the possibility of life imprisonment if death results.15 These parts of the
Patriot Act demonstrate a faith that broadening and toughening the criminal
law will help stop terrorism.
The phenomenon of enacting new criminal laws as a response to acts of
terrorism was not limited to the West. A new anti-terrorism law was pro-
posed in Indonesia shortly after September 11, but met significant resistance
in civil society. After the Bali bombings killed over 200 people on 12 October
2002, however, a new anti-terrorism regulation was enacted as an emergency
measure on 18 October 2002. Unlike the Patriot Act, the new law was made
effective with retroactive force. The 2004 bombing of the Australian embassy
in Jakarta has started another round of proposed amendments to toughen the
law raising concerns about human rights.16 Another act of terrorism in the
United States would have produced a similar momentum to enact a Patriot
Act II. Indonesia, like the United States, employs the death penalty for the
most serious acts of terrorism subject to the qualification that it would not be
imposed on those who committed their crimes as juveniles. Both countries
turned to instant criminal law reform in the wake of terrible terrorist attacks
and relied on broader and tougher offences to prevent acts of terrorism. In
both countries there was pressure on the government to be seen to be doing
something about terrorism. Tough new anti-terrorism laws were enacted to
denounce acts of terrorism, express solidarity with the victims and attempt to
deter future acts of terrorism.
New anti-terrorism laws fit into a process of the state governing through
the enactment of new criminal laws and by increasing terms of punishment.
At the same time, however, the new anti-terrorism laws also incorporate a
newer less state-centred approach to crime that is designed to encourage

John Whitehead and Steven Aden, ‘Forfeiting ‘‘Enduring Freedom’’ for ‘‘Homeland
Security’’’ (2002) 51 American University Law Review 1081 at 1087, note 26.
Patriot Act ss. 803, 805 and 810.
See Hikmahanto Juwana, Chapter 14, in this volume.

corporate and private actors to take responsibility for crime. New laws against
the financing of terrorism combine punitiveness on behalf of the state with
newer security strategies that require individuals to play their own role in the
fight against crime. Much of UN Resolution 1373 was devoted to having
states enact laws aimed at those who provide terrorists with financial and
others forms of support, consistent with the most recent international con-
vention against terrorism: the 1999 Convention for the Suppression of the
Financing of Terrorism. The objects of such financing laws are not so much
terrorists or even their ideological supporters but third parties such as bankers
and landlords.17 Duties were placed on financial institutions and others to
report dealings to the authorities. These new laws represented both an expansion
of the traditional scope of anti-terrorism laws and the impact of security
strategies that relied less on state imposition of punishment and more on risk
management strategies throughout society.
Although they were featured in Resolution 1373 and many new anti-
terrorism laws, there are also reasons to doubt the effectiveness of laws against
the financing of terrorism. The 9/11 Commission stated that ‘trying to starve
the terrorists of money is like trying to catch one fish by draining the ocean’.
It expressed doubts that the financing laws can be a ‘primary weapon’ against
decentralized terrorist groups that can fund destructive missions such as 9/11
at a cost of about half a million.18
Even in the area of national security, privatization has occurred alongside
the emergence of the punitive state that governs through crime. As Ulrich
Beck has argued, the ability to assess the multiple risks of modern society has
outstripped the ability to control such risks.19 September 11 reaffirmed a
sense that the state’s experts, this time in the field of intelligence and policing,
had once again failed. Part of the response has been to download some of the
tasks of anti-terrorism enforcement on private actors. Patrons on the London
subway are asked to be on the lookout for suspicious and abandoned
packages. The Mounties in Canada have established a 24-hour toll free
national security hotline asking for tips on suspicious people. Private com-
panies that manage data volunteer or are asked by governments to provide
information about high risk people. The United States has introduced a
national system of colour coded risk assessments that attempt to assess risk
and place both state and non-state actors on higher levels of alert. Although
the state has got tougher on terrorism since 9/11, it does not have a monopoly
on anti-terrorism efforts.

See Davis, Chapter 9, in this volume.
The 9/11 Commission Report (New York, Norton, 2004), at 12.3.
Ulrich Beck, Risk Society: Towards a New Modernity (London, Sage Publications, 1992).

II. Instrumental and normative limits of the criminal law
The instrumental value of new criminal laws in preventing terrorism can only
be determined by a careful examination of the criminal law as it existed before
it was expanded in response to a successful act of terrorism. In many coun-
tries, laws against being an accessory to murder or conspiring to commit
murder already existed. People alleged to have been involved in the 9/11 plot
have been prosecuted under existing criminal laws in both Germany and the
United States. 9/11 was much more a failure of law enforcement and intelli-
gence coordination than a failure of the criminal law. At the same time, this
does not mean that new investigative powers and more broadly defined
crimes could not have some utility in terrorist investigations. My point is
that the utility of such new laws is a marginal one and one that can be
exaggerated by ignoring the existing laws that could apply against terrorists.
The value of formal criminal law reform in some societies cannot be
discounted. The Indonesian Penal Code for example does not have a general
conspiracy offence and it defines attempted crimes restrictively to require the
accused to have commenced performance of the crime and that the perfor-
mance was ‘not completed only because of circumstances independent of the
[accused’s] will’.20 In such a context, new criminal laws applying to various
forms of preparation and financing of terrorism may have a significant
instrumental value in the apprehension of terrorists before they strike. As
the Indonesian experience of subsequent bombings demonstrates, however,
formal law reform is not enough. Police, prosecutors and courts must have
sufficient capabilities to investigate apprehended acts of terrorism.
In July 2004 the Indonesian Constitutional Court held in a 5:4 decision
that the law making the new terrorism law retroactive to the Bali bombings
violated the prohibition against retroactive punishment in the 1999
Constitution, but it was not clear whether its decision would overturn con-
victions already rendered under the retroactive law, including three death
sentences.21 The effectiveness of tough new anti-terrorism laws in general
may be diluted by successful due process challenges. If controversial anti-
terrorism laws are perceived as unjust or illegitimate by large segments of the

Indonesian Penal Code Article 53.
The majority of the Court stated that ‘The non-retroactive principle is strict. Otherwise, it
would open the door for certain regimes to use the laws as a tool to take revenge against
political opponents. There must be no chance for it to happen’ while the minority used
precedents such as the Nuremberg trials to uphold the retroactive law. Various commen-
tators indicated that the court’s decision did not annul previous verdicts but meant that
the regular penal code would have to be applied to subsequent prosecutions arising from
the Bali bombings. ‘Terror law ‘‘against constitution’’’ Jakarta Post 24 July 2004; ‘Justice
perturbed’ The Economist 31 July 2004.

population, the ordinary criminal law may also be more effective in denoun-
cing acts of terrorism. Reasonable people should agree that murder is murder
while reasonable people can disagree about definitions of terrorism. In
Indonesia, the Bali bombers could likely have been charged under the existing
law relating to murder and explosives. This would have minimized the
dangers of convictions being overturned on appeal because of the use of a
retroactive criminal law.22 Even in international criminal law, there might be
some advantage to prosecuting catastrophic acts of terrorism such as 9/11 as
crimes against humanity rather than trying to define and add crimes of
terrorism to the jurisdiction of the International Criminal Court.
The danger that controversial anti-terrorism laws may undermine some of
the denunciatory and communicative value of criminal law is particularly
high in the many countries that have followed the British example of requir-
ing proof of a political, religious or ideological cause as an essential element of
new crimes of terrorism. Although designed to differentiate terrorism from
ordinary crime, such an approach will require prosecutors to prove religious
or political motivations for the crime. This could lend some support to claims
that accused persons are being prosecuted because of their politics or religion.
The new Indonesian anti-terrorism law is sensitive to this danger as it
provides that terrorism should not be considered a political crime and that
it does not discriminate against any particular religion.23
New criminal laws against terrorism often feature enhanced penalties for
acts defined as terrorism. Although higher and harsher penalties may serve
communicative and symbolic purposes in denouncing terrorism, their value
in deterring terrorism is likely to be very marginal. A person planning a
serious act of terrorism already faces relatively high maximum penalties.
The marginal deterrent value of a new criminal law depends not only on
the severity of punishment, but its certainty and celerity. Terrorists, especially
international terrorists, may also be unfamiliar with the relevant penalties in
the jurisdiction in which they plan to commit their crime and they may be
prepared to die for their cause. Third parties who may provide assistance to
terrorists may, however, be more amenable than suicide bombers to deterrence.
There are normative dangers when the criminal law is ratcheted up after acts of
terrorism in an attempt to deter acts of terrorism. Some criminal laws may
challenge basic principles of criminal law such as the requirement for a clear illegal
act that is committed with fault. As in the United Kingdom, membership in a
terrorist organization may be criminalized. Various forms of association or parti-
cipation in a terrorist group may also be punished. The possession of instruments
or documents that could be used for terrorism may also be punished. Terrorism

Ross Clarke, ‘Retrospectivity and the Constitutional Validity of the Bali Bombing and East
Timor Trials’ (2003) 5 Asian Law 128.
Indonesian Anti-Terrorism Law arts. 2 and 5.

may be defined broadly to include much more than violence against civilians.
Many new laws against terrorism apply to the disruption of essential public and
private services including electronic systems and not all of them make adequate
allowance for political protests or stoppages of work.
Alan Dershowitz has suggested that terrorists may be amenable to deter-
rence because of the calculated and political nature of terrorism. Even
assuming the empirical accuracy of Professor Dershowitz’s claim, the deter-
rence strategies that he proposes depend in part on collective punishment,
something that he admits is ‘the most immoral technique for combating
terrorism’.24 The idea of collective punishment is implicit in Dershowitz’s
argument that the political cause of the terrorists should be punished for acts
of terrorism because ‘the cause hopes and expects to benefit collectively from
terrorism’.25 The problem is that ‘causes’ do not commit acts of terrorism,
individuals do. From the perspective of the criminal law, the punishment of
the cause imposes punishment on the innocent. Although at times he
seems aware of the injustice of collective punishment, Professor Dershowitz
concludes that ‘any effective attack calculated to reduce terrorism – especially
suicide bombers – must include an element of collective responsibility and
punishment for those supporting terrorism’.26 This departs from the funda-
mental focus on individual responsibility under the criminal law and the idea
‘that punishing the mentally innocent with a view to advancing particular
objectives is fundamentally unfair. It is to use the innocent as a means to an
end’.27 Although the notion of collective guilt can influence public discourse,
it is alien to legal discourse.28
Laws against the financing of terrorism are not aimed at terrorists or even
those who may sympathize with their cause, but business people who are
required, on pain of criminal conviction, to use their own resources to ensure
that they are not assisting terrorists. Such systems are also encouraged by lists
distributed by international, regional and domestic agencies of people who
are designated as terrorists, lists that are often incorporated in the domestic
law of many nations. Those listed are not generally given an opportunity to
make submissions before they are listed and the provisions for removing
those mistakenly added to the list may be slow and not repair the damage of
being officially listed as a terrorist. The International Bar Association has
criticized the UN for not providing ‘for any judicial review or right of appeal
to an individual or entity whose assets are incorrectly frozen’.29 There is also a

Alan Dershowitz, Why Terrorism Works (New Haven, Yale University Press, 2002), at 117.
Ibid. at 174. 26 Ibid. at 181.
R. v. Hess [1990] 2 S. C. R. 906 at 923–4 per Wilson J.
George Fletcher, Romantics at War (Princeton University Press, 2002).
International Bar Association, International Terrorism: Legal Challenges and Responses
(Ardsley Park, Transnational Publishers, 2003), at 126.

danger that executive listing mechanisms will spread into domestic criminal law
and undermine the role of the judiciary in determining who is a terrorist.30
Another danger is that the new laws may encourage third parties to err on
the side of caution by not dealing with those that they suspect may be
involved in terrorism. If this occurs, the process of punishment will have
been contracted out by the state to financial institutions. The sanction will be
applied not after a court has found the accused guilty beyond a reasonable
doubt or even the executive has made a listing decision, but rather when a
private institution decides it does not want to run the risk of association with
a terrorist and possible prosecution. The sanction imposed by such third
parties will not, of course, be imprisonment, but it could amount to a form of
civil death in which the suspect is denied access to a wide range of essential
services including banking and shelter. Although this process may not meet
all definitions of state-imposed punishment, it will be one in which people
suffer harm and stigma because of suspicions that they are terrorists.
There are normative dangers of distorting criminal law principles in order
to facilitate the apprehension of terrorists. One danger is that extraordinary
powers may be introduced and justified in the anti-terrorism context but
then spread to other parts of the criminal law. Incursions on the right to
silence were first made in the United Kingdom in anti-terrorism law but have
now spread throughout the entire law.31 The United States Department of
Justice has proudly boasted that parts of the Patriot Act have been used
against those charged with child pornography and other crimes that have
nothing to do with terrorism.32 Broad definitions of terrorism and claims
that new anti-terrorism regimes are consistent with rights protection instru-
ments may facilitate the spread of anti-terrorism law to other parts of the
criminal law. In any event, anti-terrorism measures will be intensified should
they not be successful in stopping other acts of terrorism. One of the reasons
why a widely rumoured Patriot Act II has not yet been introduced in the
United States is the fortunate absence of other acts of terrorism on American
soil since 9/11. As the experience of the first part of this chapter suggests, anti-
terrorism law will expand and build on its own failures.
Another danger is that changes to criminal law principles will produce a
greater risk of miscarriages of justice. In the United Kingdom, a series of
wrongful convictions occurred with respect to IRA bombings. Suspects were

In Pakistan a new anti-terrorism law allows either groups or individuals to be listed as
terrorists and allows for the arrest of terrorist suspects on this basis. See ibid. at 45.
Oren Gross, ‘Cutting Down Trees: Law-Making Under the Shadow of Great Calamities’ in
Ronald J. Daniels, Patrick Macklem and Kent Roach (eds.), The Security of Freedom: Essays
on Canada’s Anti-Terrorism Bill (Toronto, University of Toronto Press, 2001).
United States Department of Justice, Report From the Field: The USA Patriot Act at Work
(July 2004), at 19.

identified in part because of their nationality and political sympathies. They
were mistreated in custody and did not have adequate disclosure of evidence,
including dubious forensic evidence, used against them. Although there is
always a risk of wrongful convictions even under the ordinary criminal law,
some features of new anti-terrorism laws produce even greater risks of
wrongful convictions. Some anti-terrorism laws allow restrictions on the
disclosure of relevant evidence to the accused in an attempt to protect
intelligence sources and other national security information. Some terrorism
offences are defined in such a broad manner that they resemble both status
offences and guilt by association. Other offences neglect principles of sub-
jective fault and the presumption of innocence. Extreme interrogation tech-
niques may increase the risk of false confessions and false intelligence.33
Although new criminal laws against terrorism may increase the risk of
wrongful convictions, a possible safeguard is the role of the judiciary in
applying such laws. Most countries have not derogated from rights protection
instruments and this allows the accused to argue to the independent judiciary
that aspects of the new laws violate rights. In the United States, parts of the new
offence of providing material support for terrorism have been held to be
unconstitutional because of the vagueness of some of the terms used. In the
United Kingdom, reverse onuses imposed on the accused have been changed to
evidential burdens in recognition of the accused’s right to a fair trial. At the
same time, judges are not entirely immune from the pressures that are placing
increased emphasis on security. They may be willing to accept arguments that
anti-terrorism laws are justified in part by reference to the rights of victims and
potential victims and they may decide that it is a mistake to evaluate these laws
only through the traditional due process optic of the balance of power between
the individual accused and the state.34 Ronald Dworkin has eloquently warned
of the dangers of concluding that ‘the requirements of fairness are fully
satisfied, in the case of suspected terrorists, by laxer standards of criminal
justice which run an increased risk of convicting innocent people’.35 It is
important that criminal law does not lose sight of its foundational principles
such as the presumption of innocence and the necessity of proof of individual
fault beyond a reasonable doubt. These demanding standards, however, create
another risk, namely that states will find the crime model, despite its political
and moral appeal, to be too weak for fighting a war against terrorism.36 When

Kent Roach and Gary Trotter, ‘Miscarriages of Justice in the War Against Terror’ (2005)
109 Pennsylvania State Law Review 967.
Irwin Cotler, ‘Terrorism, Security and Rights: The Dilemma of Democracies’ (2002) 14
National Journal of Constitutional Law 13.
Ronald Dworkin ‘The Threat to Patriotism’ New York Review of Books 28 Feb. 2002.
For arguments that the crime model is not sufficient to deal with terrorism and new rules
are required see Bruce Ackerman, ‘The Emergency Constitution’ (2004) 113 Yale Law
Journal 1029. For arguments against Professor’s Ackerman’s proposed regime including

the state goes beyond the criminal law, however, the restraining rules become
much less clear and demanding. Indeed, at times, there appears to be no rules
at all.

III. Less restrained alternatives to the criminal law
Although there are many instrumental and normative dangers in reliance on
the criminal law as a means to combat terrorism, the focus on individual
responsibility and deserved punishment in the criminal law has many virtues,
especially when compared to some of the other techniques that have been
used against terrorism. Since September 11, many countries have chosen to
use other instruments that, like the criminal law, rely on coercive force and
detention, but do so without most of the safeguards and restraints associated
with the criminal law. These less restrained alternatives to the criminal law
have included wars in Afghanistan and Iraq, assassinations,37 the use of
torture and humiliation in prisons in Iraq and elsewhere, extraordinary
rendition to third countries, the use of detention without judicial review at
Guantanamo Bay and the use of detention with limited judicial review under
immigration or internal security laws. All of these measures starkly reveal the
virtues of the criminal law and the dangers of opting out of the crime model.
In his 2004 State of Union address, President George W. Bush made clear
that the United States under his leadership would not rely on the criminal law
in its war against terrorism. He stated:
I know that some people question if America is really in a war at all. They
view terrorism more as a crime, a problem to be solved mainly with law
enforcement and indictments. After the World Trade Center was first
attacked in 1993, some of the guilty were indicted and tried and convicted,
and sent to prison. But the matter was not settled. The terrorists were still
training and plotting in other nations, and drawing up more ambitious
plans. After the chaos and carnage of September the 11th, it is not enough
to serve our enemies with legal papers. The terrorists and their supporters
declared war on the United States, and war is what they got.38
The idea that is not ‘enough to serve our enemies with legal papers’ is a
recurring theme in American counter-terrorism policy. It is seen in President

its use of preventive detention, see David Cole, ‘The Priority of Morality: The Emergency
Constitution’s Blind Spot’ (2004) 113 Yale Law Journal 1753.
Plans by both President Clinton and Bush to capture and preferably kill bin Laden before
the September 11 attacks are discussed in The 9/11 Commission Report, chs. 3 and 4. Justice
Thomas (in dissent) has expressed concern that the Supreme Court’s decision in Hamdi v.
Rumsfeld might have required some due process in a case such as the November 2002 use
of a Hellfire missile fired from a CIA predator drone at a vehicle in Yemen believed to
contain al Qaeda members including an American citizen.
State of the Union Address 20 January 2004.

Bush’s military order authorizing the detention and trial of people captured
in Afghanistan. The order provides that the rules of evidence will not apply
and purports to preclude the jurisdiction of either the domestic courts of the
United States and international courts over the detainees. It is also seen in the
attempts made by the United States to dispense with the requirements of due
process by declaring American citizens Yaser Hamdi and Jose Padilla to be
enemy combatants under the control of the military and by detaining over
600 non-citizens in Guantanamo Bay in an ultimately futile attempt to evade
the habeas corpus jurisdiction of American courts.
The war against the Taliban regime in Afghanistan was in part justified by
the same language of disrupting terrorist networks that is used to justify
criminal laws against financing terrorism.39 The use of the language of the
criminal law can give war, and its inevitable killing of innocent people,40 a
moral legitimacy that is not deserved. The war against the Hussein regime in
Iraq was also justified in part as an anti-terrorism effort. Efforts were made to
link Iraq with al Qaeda and unfounded claims about weapons of mass
destruction in Iraq were made in a manner that appealed to public fears
about the possibility of biological, chemical or even nuclear terrorism.
Although wars are governed by laws, including criminal law pertaining to
war crimes, war is a much more indiscriminate instrument of force than a
criminal prosecution. Modern technology may allow more innocent lives to
be spared than ever before, but so-called collateral damage is an accepted part
of war. This can be contrasted with the widespread revulsion at the conviction
of the innocent under the criminal law. To be sure, wrongful convictions do
occur and may at some level be inevitable given widespread use of the
criminal law, but they are a matter of grave concern in most criminal justice
Immigration law, particularly in Western countries, has been used as
perhaps the prime instrument to counter international terrorism since 9/11.
It routinely employs what in criminal law would be seen as problematic
status-based offences and standards of proof well below the criminal law
standard of proof beyond a reasonable doubt. Although new anti-terrorism
criminal laws have been influenced by concepts developed in immigration
law,41 they still provide a higher standard of fairness and due process than

For an account of how language borrowed from the criminal law has been used to justify
military action in the war against terrorism, see Note ‘Responding to Terrorism: Crime,
Punishment and War’ (2002) 115 Harvard Law Review 1237.
For an estimate of 100,000 additional deaths in post-invasion Iraq, most attributable to
Coalition Forces, see Les Roberts et al, ‘Mortality Before and After the 2003 Invasion of
Iraq: Cluster Sample Survey’ Lancet 29 Oct. 2004, at http://image.thelancet.com/extras/
Audrey Macklin ‘Borderline Security’ in The Security of Freedom; David Cole, Enemy
Aliens (New York, New Press, 2003).

immigration laws. Although it is not a crime to be a member of a terrorist
group in either the United States or Canada, it is a ground for apprehension
and removal under both countries’ immigration laws. Immigration law is also
more accepting of preventive, investigative and indefinite detention than the
criminal law.42 The United Kingdom has derogated from fair trial rights
in order to provide for indefinite detention of non-citizens suspected of
involvement with terrorism who could not be deported because of concerns
that they would be tortured. There are strong arguments that this law discri-
minates against non-citizens. Immigration proceedings in many countries are
closed to the public and make use of evidence not disclosed to the accused.
Western immigration laws have fulfilled some of the same roles in facilitating
preventive detention without trial as the Internal Security Acts in Malaysia and
The use of immigration law to detain and deport suspected terrorists is
striking given the alternative that is available under most new anti-terrorism
criminal laws of charging these people with a broad range of criminal offences
relating to terrorism. Most new criminal anti-terrorism offences define sup-
port for terrorism in a broad fashion and apply extraterritorially. In the
United Kingdom reliance on immigration law in the Anti-terrorism, Crime
and Security Act 2001 is especially striking given the broad crimes, including
crimes of membership in a terrorist group, that are available under the
Terrorism Act 2000. To this end, the Newton Committee recommended that
as ‘a matter of urgency’ the immigration law powers of indefinite detention
contained in Part 4 of the 2001 Anti-terrorism Act be replaced and steps taken
to rely more on criminal prosecutions against non-citizens and citizens alike.43
The United Kingdom government has rejected this suggestion concluding
that the Newton Committee did ‘not offer a solution to the need to protect
sensitive information whilst enabling the defendant to know the full case that
has been put against him’.44 The easiest route for Western states since 9/11
has been to proceed under immigration laws which allow preventive and
indefinite detention and closed hearings based on evidence not disclosed to
the non-citizen rather than attempt to prove guilt beyond a reasonable doubt
in a criminal trial.

See Roach, Chapter 23, in this volume.
Lord Newton (chair) Anti-terrorism, Crime and Security Act 2001 Review Report 18 Dec.
2003, at para 203–5. The Newton Committee suggested that some reforms to the criminal
law such as the use of a security cleared judge to assemble the case, greater incentives for
plea bargains and ‘a more structured disclosure process that is better designed to allow the
reconciliation of the needs of national security with the rights of the accused to a fair trial’
might be required. Ibid. at para. 241.
Home Secretary, Counter-terrorism Powers: Reconciling Security and Liberty in an Open
Society Feb. 2004 Part II, at para. 37.

Immigration law can tolerate a degree of over and under-inclusiveness that
would not be accepted in the criminal law. Some countries have restricted
their acceptance of refugee applicants for security reasons even though
terrorists would only be a small percentage of such applications and most
terrorists would want to avoid the scrutiny that accompanies a refugee
application. Immigration law is also under-inclusive because it cannot be
applied against home-grown terrorism or internationally inspired terrorism
that is committed by citizens of the particular country.
The demands of the criminal law can be seen in the few 9/11 related
criminal prosecutions that have been undertaken in the United States and
Germany. Zacarias Moussaoui, the so-called twentieth hijacker, has been
charged with a variety of existing criminal conspiracy offences under
American law. The government has refused Moussaoui access to Ramzi bin
al-Shibh who is in American custody and is believed to have played a key role
in the attacks. The trial judge eventually ruled that the death penalty should
not be applied and the matter is now the subject of appellate litigation. An
appeal court has affirmed Moussaoui’s right to have access to witnesses who
can provide material evidence essential to his defence, but has also held that
the government has acted in good faith and overturned the sanction of
making the accused ineligible for the death penalty.45 There has been spec-
ulation that should the courts require the production of key al Qaeda
witnesses, the United States government may halt the criminal prosecution
and declare Moussaoui, a French citizen, to be an enemy combatant subject to
indefinite detention and trial before a military tribunal. The prospect of the
American government opting out of a criminal prosecution to avoid a full
adversarial exploration of all relevant issues in a criminal trial would demon-
strate the impatience of the United States government with the idea that ‘legal
papers’ must be served on those suspected of involvement with 9/11.
The experience in Germany with prosecutions under the criminal law
illustrates in even more dramatic form the commitment of the criminal law
to basic principles of legality. The conviction of Mounir el-Motasssedeq in a
Hamburg court of over 3,000 counts of accessory to murder, including the
accused’s fifteen-year sentence, was reversed on appeal. Although noting that
the accused ‘is certainly far removed from being clear of suspicion’, the appeal
court stressed that the accused should have access to evidence of a key
witness, Ramzi bin al-Shibh. The German court reasoned that ‘a conflict
between the security interests of the executive and the rights to defence of the
accused cannot be resolved to the disadvantage of the accused’. The court
defended the legality model of the criminal law over the war metaphor by
stating that ‘we cannot abandon the rule of law. That would be the beginning

United States of America v. Moussaoui 22 April 2004 (4th Cir.)

of a fatal development and ultimately a victory for the terrorists . . . The fight
against terrorism cannot be a wild, unjust war.’46 This appeal court ruling
followed on the heels of an acquittal of another alleged member of the
Hamburg al Qaeda cell a few weeks earlier, with both courts emphasizing
the importance of providing the accused access to relevant evidence.47
The lack of success of 9/11 criminal prosecutions reveals that even under
the sway of that terrible event, criminal courts in both the United States and
Germany are still drawn to the due process and legality norms associated with
the criminal law. In the German criminal courts, victims of 9/11 were
represented and in one case tried without success to introduce incriminating
evidence. Even with respect to the initial conviction, the maximum fifteen-
year sentence was criticized as too lenient given the enormity of the damage.
Although the criminal law is often defended in the language of the rights of
victims, the ability of the criminal law to deliver satisfying justice to victims
remains very much in doubt.
The willingness of the United States government to risk criminal convic-
tions in order to keep intelligence sources secret48 is not novel. Even before 9/11,
governments often preferred covert surveillance or infiltration of terrorist
groups to criminal charges and prosecutions. Such an approach appears to be
driven by utilitarian and instrumental rather than retributive demands for
punishment. It is striking that even in relation to 9/11, fears about exposing
intelligence sources still prevailed over the need to turn over evidence to
support a criminal prosecution. At the same time, the desire for retribution
and punishment arising from 9/11 has been channelled into instruments less
restrained and discriminating than the criminal law. As discussed above,
these measures include war, targeted killings, torture and humiliation of
prisoners and the rendition of suspects to face torture. These blunt and
bloody alternatives to the criminal law reveal the virtues of the criminal law
in focusing on questions of individual responsibility within a due process
framework, as well as the danger that the criminal law will be abandoned in
the combat of a perpetual war against terrorism.

As quoted in Jeff Sallot, ‘Guilty verdict overturned in al-Qaeda suspect’s case’ Globe and
Mail (Toronto) 5 March 2004, A14; Desmond Butler, ‘German Judges order a Retrial for
9/11 Figure’ New York Times 5 March 2004.
Luke Harding, ‘German court clears student of plotting with 9/11 terrorists’ The Guardian
6 Feb. 2004.
The US government has subsequently indicated that it is prepared to release ‘unclassified
summaries’ of intelligence. The accused objected to the use of such documents because of
concerns about how such intelligence had been obtained, raising objections from one of
the victims that such arguments desecrate ‘the memory of 3000 people who died, includ-
ing my mother’. ‘US Pledges to Share Evidence as 9/11 Retrial Begins in Germany’ New
York Times 10 Aug. 2004.

IV. Instrumental and normative dimensions of a broader
anti-terrorism strategy
The criminal law should not be abandoned, but it cannot be the exclusive
instrument used against terrorism. There is a need to situate the criminal law
within a more comprehensive anti-terrorism strategy. Such work should account
for the reality that while criminal law reform often figures prominently in public
discourse and in country reports concerning compliance with Resolution 1373,
the criminal law only plays a partial role in most anti-terrorism strategies. It is
important to account for both the instrumental and normative dimensions of a
broader anti-terrorism strategy that does not rely on the criminal law. Why from
an instrumental perspective have many states not relied upon the criminal law
in their actual fight against terrorism? What are the normative dangers of
abandoning the criminal law as society’s ultimate response to terrorism?
The full development of a comprehensive anti-terrorism policy is obviously
beyond the scope of this chapter, but the project can be advanced by selective
incorporation of regulatory strategies from outside the field of terrorism or
crime. A promising construct for situating the criminal law in broader anti-
terrorism policy can be taken from the fields of public health and science and
technology.49 In order to assess a variety of countermeasures that could reduce
death and injury from traffic accidents, epidemiologist William Haddon con-
structed a matrix evaluating countermeasures that could be taken to minimize
harm before, during and after the accident. Haddon also distinguished between


ńňđ. 5
(âńĺăî 23)