ńňđ. 4
(âńĺăî 23)



Parliament. Such an Act is an ‘exercise of arbitrary sovereign power’ but its enactment by a
parliamentary assembly ‘maintains in no small degree the real no less than the apparent
supremacy of law’.

sliding scale of due process or legality which the plurality in Hamdi with
Souter and Ginsburg contemplated. Their approach, which seems closer to
my Legality model, seeks to preserve the rule of law. But it risks preserving the
rule of law in a way that, like Thomas’s dissent, legitimates the government
while it does more or less what it likes. However, the risk of this result is much
greater for the plurality than it is for Souter’s and Ginsburg’s opinion.
My main point is only that the world of legality is more messy than the
Extra-Legal Measures model supposes, even abstracting from the situation of
responses after 9/11. In ordinary administrative law, there are issues about
how best to protect liberty interests when decisions are made by adminis-
trative bodies like parole boards, prison disciplinary panels, and tribunals
which make mental health determinations. And I offered SIAC as an example
of a body which could successfully preserve legality in an area where other-
wise none might exist. With the caveat mentioned above about the category
of the unlegalizable, I do think it is worthwhile running the risk of preserving
legality through institutional experiments. However, judges must insist, as
Souter and Ginsberg did in Hamdi, that there is both an absolutely explicit
legislative mandate for such experiments and that the experiments be con-
ducted in accordance with the Legality model.
Further, when it comes to the category of the unlegalizable, I think it would
be helpful to revisit the idea of an Act or Bill of Attainder. As the author of the
Note in 1962 in the Yale Law Journal explains, the term Act or Bill of
Attainder comes from the practice in sixteenth, seventeenth and eighteenth
century England of using statutes to sentence ‘to death, without a conviction
in the ordinary course of judicial trial, named or described persons or
groups’.64 In addition, the term came to be used for ‘bills of pains and
penalties’, statutes that imposed sanctions less than capital.65 Both sorts of
statute were aimed at revolutionaries and were considered offensive to the
spirit of the common law because they attempted to bypass the courts by
establishing a system of either legislative or administrative conviction and
One way of understanding the offence is in terms of an idea of the
separation of powers, where the judiciary has the role of determining in an
open trial both guilt and appropriate punishment. As T. R. S Allan argues in
the leading theoretical treatment of the rule of law, the substance of the
argument against bills of attainder pertains to the fact that the statute in
issue offends the constitutional guarantee, written or unwritten, of an inde-
pendent judiciary presiding in open court over determinations of guilt and
punishment. A bill of attainder, he says, is just ‘the paradigmatic example of
legislation whose violation of the principles of equality and due process

64 65
Note, 72 Yale Law Journal (1962) 330. Ibid., 331.

contravenes the rule of law’.66 The repugnance of the common law tradition
to such statutes is born of the idea that while the legislature can enact into law
its understandings of subversion and other offences, the rule of law requires
both that that offence be framed generally and that anyone accused of such an
offence be tried in a court of law. The argument is a deeply normative one.
But it is not so much about the separation of powers, as about the reasons for
the separation of powers – the constitutional role of the judges is to see to it
that the fundamental values of legal order are preserved, by whatever means
are most appropriate. When statutes seek to legalize activity contrary to these
fundamental values, to legalize the unlegalisable, they lack legal authority.
Judges may then refuse to enforce such statutes even without explicit con-
stitutional authority to do so.
Those with authority, including those at the very top of the hierarchy of
legal order, must understand these values, so that they can take part in the
common project of their realization. On this view, the separation of powers is
instrumental to realizing the legal order’s ideals. If to ensure the integrity of
legal order, it is necessary to imagine institutions for the enforcement
of legality that go against the grain of received views about the separation of
powers, one should not let those views stand in the way of enforcing legality.
Moreover, this claim applies to all societies that assert they are governed by
the rule of law. Earlier, I mentioned the distinction between liberal demo-
cratic societies and well-ordered societies, the point being that well-ordered
societies, while neither fully liberal nor democratic, share with liberal demo-
cratic societies the commitment to the rule of law. I suggested further that
Gross’s Extra-Legal Measures model had no place in a well ordered society
because its legitimacy seems to depend on protecting liberal democracy. In
contrast, the model I have sketched in this chapter applies to such a society as
long as its assertion has some basis to it, that is, as long as it does not pay mere
lip service to the rule of law. As long as there is such a basis, those subject to
the law will be able to hold public officials to account, their accountability not
being just to the positive law but also to the values of legality.67

T. R. S. Allan, Constitutional Justice: A Liberal Theory of the Rule of Law (Oxford University
Press, 2001), 148.
At the symposium, several people expressed concern that while an institution such as
SIAC might work well in the United Kingdom, it has the potential to be a mere fig-leaf for
the executive in well ordered societies. Everything depends, they suggested, on the
character of those who staff the institution. My response, which owes much to conversa-
tions with Kristen Rundle, is that one should not neglect the way in which the design of
institutions shapes perception of agency. To appoint a person to an internal, advisory
committee with the task of hearing appeals against detention orders is to ensure that those
on the committee are likely to rubber stamp executive decisions. But give those same
people the powers of SIAC and they might be forced into an understanding of their role far
different from the rubber stamp.

It is important to stress this last point lest one think that my claim that the
separation of powers has to be understood as instrumental to the substantive
values of legality means instrumental to whatever values the powerful decide
to implement through law. If that thought were right, then, for example, the
legislature could simply decree that communal values are the values to which
judges should have regard when it came to imposing the rule of law on public
officials. The thought is not right, however.
There are other ways of attempting to place officials in a black hole than the
legislature certifying the blank cheque to officials or indemnifying them after
the fact. One is the general privative clause which seeks to exclude judicial
review. Another is what I like to call a substantive privative clause, a provision
which does not exclude review in general but review on particular grounds,
for example, natural justice or reasonableness. I am not here concerned with
the appropriate judicial response to these provisions, only with the charac-
terization of them – that they place officials in an extra-legal space or black
hole. When it comes to well-ordered societies, which might have a strong
orientation to communal values, my claim is that their commitment to the
rule of law will be tested by the extent to which they are prepared to refrain
from allowing communal values to take precedence over the values of legality.
Extra-legal space can also be created by permitting certain communal values,
for example Schmittian values to do with the alleged moral identity of the
nation, a large role in interpreting the law. In this case, officials become
accountable to those values, which means really to their sense of what the
values require. In other words, they are no longer accountable. Thus, a
commitment to the rule of law is also a commitment to abiding by values
of fairness, reasonableness and equality before the law. Abiding by those
values, the values of the Legality model, provides a basis for a society to
claim some legitimacy even if one supposes that to be fully legitimate, a
society should also be both democratic and liberal. But to have that legiti-
macy, the society must not opt for the Extra-Legal Measures model in its
dealings with either emergencies or terror.

Stability and flexibility: a Dicey business

The chief restraint upon those who command the physical forces of the
country . . . must be their responsibility to the political judgments of their
contemporaries and to the moral judgments of history.1

I. Introduction
A significant part of the life of the law has been attempts to balance the
competing values of stability and flexibility. In some areas greater weight may
be accorded to flexibility while in others stability is particularly valued. In
some areas the balance between stability and flexibility will be stable, in others
it will require frequent re-calibration.
The rule of law has long been identified as one of the most fundamental
tenets of a democratic regime, as ‘the soul of the modern state’.2 The terms
‘rule of law’, ‘constitutionalism’, ‘individual rights’, ‘democracy’ and ‘liberal-
ism’ are frequently mentioned as integral parts of a unified whole. Despite
lack of consensus as to its precise content and scope, the rule of law has been
connected to notions of generality, clarity, certainty, predictability and sta-
bility of rules. At the same time, general legal rules must also be flexible
enough to adapt to unforeseen circumstances and developments. When such
developments take place over time there may be a sufficient lag to allow for
changing the rules so as to accommodate the new realities. Constitutional
amendment provisions offer one example. A power to amend may be neces-
sary to allow adaptations to sustained pressures for change. At the same time,
the tension between the demands of stability and flexibility becomes almost
unbearable when there is not enough time to adapt the laws to the changing
circumstances and when immediate ‘specific’ action is deemed necessary.
Emergencies present the challenge of enabling government to confront the
crisis by, if necessary, using special emergency powers and greater flexibility
of operation while, at the same time, ensuring that such powers and flexibility
do not get out of control and allow government to impose long-term

Korematsu v. United States, 323 US 214, 248 (1944) (Jackson J, dissenting).
Roberto Mangabeira Unger, Law in Modern Society 192 (New York, Free Press, 1976).


limitations on individual rights and liberties or modify the nature of the
relevant constitutional regime. How then should the legal system deal with
the need to allow the authorities to respond effectively to a given crisis while
minimizing the risks of overreaction both in the short and the long term?
David Dyzenhaus begins his excellent contribution to this volume by
reference to A. V. Dicey’s treatment of the challenge of balancing stability
and flexibility in the context of emergencies and grave crises. Dyzenhaus
reads Dicey as suggesting that ‘it would be far better if Parliament could
have delegated in advance the resources to public officials to deal with the
emergency’ than leave those officials to act outside the law with the possibility
of a subsequent Act of Indemnity. Put differently:
Dicey thus distinguishes very clearly between two legal responses to an
emergency situation. In the first, the response is the after-the-fact recogni-
tion that officials made an excusable decision to act outside of the law
because it was necessary that they act and the law did not provide them with
the resources they needed. In the second, Parliament in advance gives to
officials resources to deal with emergencies in accordance with the rule of
law. Dicey prefers the second option . . . one that relies on legislative
solutions that preserve the rule of law.3

Dyzenhaus takes a similar position in order to develop his own model of
legality which is based on ‘experiments in institutional design’.4 He argues
that such experiments carry the promise of dealing with crises and exigencies
in a manner that is fully consistent with the rule of law, i.e. without forcing
public officials to act outside the law.
This chapter will not attempt to address all of the issues raised by
Dyzenhaus’s sophisticated argument. Rather, I will focus on his critique of
my Extra-Legal Measures model and leave a fuller discussion of his own
model of legality for another day. As noted above, Dyzenhaus invokes
Dicey’s discussion of the legal responses to ‘times of tumult or invasion’ in
building his model of legality and argues that Dicey’s approach to the
fundamental dilemma of stability and flexibility in times of crisis was
right.5 Part I follows Dyzenhaus’s lead and examines Dicey’s solutions to
the problem of emergencies. I suggest that Dicey does present two ways
of responding to emergency situations, but that he sees both as two sides of
the same coin. Thus, the two solutions are complementary allowing the use
of one when the other may be unavailable or undesirable. I then tie Dicey’s
analysis with John Locke’s theory of the prerogative, suggesting that
Dicey’s recognition of the possibility of public officials acting outside the law
answers a significant problem with Locke’s theory. Part II focuses on a closer

David Dyzenhaus, Chapter 4, in this volume.
Ibid. at 67. 5 Ibid. at 65.

examination of the ex post ratification component of the Extra-Legal
Measures model. Once again, it does so by using Dicey’s discussion of the
Act of Indemnity, which is a particular case of ex post ratification. This Part
seeks to demonstrate that Dyzenhaus’s critique of the Extra-Legal Measures
model as placing public officials in a ‘legal black hole . . . a zone uncontrolled
by law’ misses some of the essential components of the Extra-Legal Measures
A brief summary of the Extra-Legal Measures model may be in order.6 The
basic assertion of the model is that there may be circumstances where the
appropriate method of tackling grave dangers and threats may entail going
outside the constitutional order, at times even violating otherwise accepted
constitutional principles, rules and norms. The model is built around three
essential components: official disobedience, disclosure and ex post ratifica-
tion. The model calls upon public officials having to deal with catastrophic
cases to consider the possibility of acting outside the legal order while openly
acknowledging their actions and the extra-legal nature of such actions. If a
public official determines that a particular case necessitates her deviation
from a relevant legal rule, she may choose to depart from the rule. At the same
time, not only does the basic rule continue to apply to other situations (that
is, it is not cancelled or terminated), it is not even overridden in the concrete
case at hand. Thus, rule departure constitutes, under all circumstances and all
conditions, a violation of the relevant legal rule. However, whether the actor
will be punished for her violation remains a separate question. Society retains
the role of making the final determination whether the actor ought to be
punished and rebuked, or rewarded and commended for her actions. It
should be up to society as a whole, ‘the people’, to decide how to respond
ex post to extra-legal actions taken by government officials in response to
extreme exigencies. The people may decide to hold the actor accountable for
the wrongfulness of her actions, or may approve them retrospectively. Until
the extra-legal action is ratified ex post, and, as explained below, even after it is
so ratified, the official does not know what the personal consequences of
violating the rule are going to be.
A final point regarding the ‘generality’ of the model seems to be in need of
some clarification. Dyzenhaus questions the extent to which the Extra-Legal
Measures model is designed as a general model for dealing with states of
emergency.7 In Chaos and Rules, I suggested that,
The proposed Extra-Legal Measures model does not seek to do away with
the traditional discourse over emergency powers. It does not claim to

For a fuller discussion of the model see Oren Gross, ‘Chaos and Rules: Should Responses to
Violent Crises Always Be Constitutional?’ (2003) 112 Yale Law Journal 1011, 1096–133.
Dyzenhaus, Chapter 4.

exclude the constitutional models of emergency powers. It is a model for
truly extraordinary occasions. There may be circumstances when it would
be appropriate to go outside the legal order, at times even violating other-
wise accepted constitutional dictates, when responding to emergency situa-
tions. Yet, even in circumstances where use of the model is inappropriate,
and where the constitutional models may supply an answer to the parti-
cular predicament, we must recognize the limitations of each of these
alternatives and its long-term implications.8

Part I starts by suggesting that Dicey would agree with this statement.

II. Dicey’s ‘spirit of legality’
A. ‘Aid from Parliament’
In his celebrated Introduction to the Study of the Law of the Constitution,9
A. V. Dicey addresses the challenge of balancing stability and flexibility. His
basic approach views executive discretion with much suspicion treating it as
leading to use of arbitrary powers. Rather than face the prerogative of the
Crown, Dicey reminds us that ‘the supremacy of the law of the land both calls
forth the exertion of Parliamentary sovereignty, and leads to its being exer-
cised in a spirit of legality’.10 At the same time, he acknowledges that ‘the
rigidity of the law constantly hampers (and sometimes with great injury to
the public) the action of the executive’.11 Specifically, he concedes that ‘under
the complex conditions of modern life no government can in times of
disorder, or of war, keep the peace at home, or perform its duties towards
foreign powers, without occasional use of arbitrary authority’.12 How are we
supposed, then, to meet the challenge of the need to authorize occasional use
of arbitrary (i.e. discretionary) authority while maintaining limitations and
checks on the use of such power and ensuring that it is confined in some
sense? Dicey offers us two complementary solutions. First is for the executive
to obtain from Parliament ‘the discretionary authority which is denied to the
Crown by the law of the land’, i.e. by recourse to ‘exceptional legislation’.13
While exigencies call for the exercise by the executive of discretionary power
such power must be governed by statute. The executive must obtain ‘aid from
Parliament’ in fashioning the discretionary powers with which to meet, and
successfully repel, crises and emergencies.14 The fact that executive emer-
gency powers are derived from, and based on, statutes reaffirms parliamen-
tary supremacy even in times of grave threats to the nation, while at the same

Gross, ‘Chaos and Rules’, at 1134.
A. V. Dicey, Introduction to the Study of the Law of the Constitution (8th edn., Liberty
Classics reprint, 1982). Throughout this chapter I refer to the 8th edition of Dicey’s work
since that was the last edition that he himself prepared.
Ibid. at 273. 11 Ibid. at 271. 12 Ibid. 13 Ibid. 14 Ibid. at 272.

time, it acts to limit and confine the scope of such powers. They may be
discretionary. They may be extraordinary. But they are never unlimited and
as such they are always open to review by the courts.15
What form should ‘aid from Parliament’ take? The obvious response
suggested by Dicey is the passage of ‘exceptional legislation’ that would
enable the executive to exercise discretionary powers.16 Such legislation
may seek to adapt existing laws so as to make them more sensitive to the
needs of dealing with crises,17 or it may, but need not, take the format of
stand-alone legislation: emergency provisions may be included in specific
‘emergency’ legislation, but they may also be incorporated into an ordinary
piece of legislation while retaining their specific emergency features.18 The
special legislation may introduce institutional and structural modifications
to the existing legal system that are deemed essential for crisis management.
Most significantly for the purposes of this chapter, such exceptional legisla-
tion will be enacted ex ante, i.e. prior to the exercise of the relevant powers by
the executive. It may be introduced either on an ad hoc (i.e. to meet a
concrete exigency) or on a permanent (e.g., as a constitutional scheme set
for dealing with ‘emergencies’) basis.19 Once put in place it serves as the legal
background against which executive emergency powers will be exercised, and
against which their legality and eventual legitimacy will be measured and
Yet, Dicey also recognizes that ex ante special legislation does not ‘exhaust
the instances in which the rigidity of the law necessitates the intervention of
Parliament’.20 Rather, ‘there are times of tumult or invasion when for the sake
of legality itself the rules of law must be broken. The course which the
government must then take is clear. The Ministry must break the law and
trust for protection to an Act of Indemnity.’21 By enacting such an Act of
Indemnity Parliament ‘legalizes illegality’ and asserts its sovereignty and
supremacy.22 Here, again, the executive must obtain aid from Parliament.

Ibid. at 273.
Dicey gives here the example of an Alien Act, Foreign Enlistment Act and Extradition Acts.
Ibid. at 272.
See, e.g., William J. Stuntz, ‘Local Policing After the Terror’ (2002) 111 Yale Law Journal
2137, 2139 (noting that some anti-terrorism legislation is not targeted, in that additional
governmental powers are not limited to the fight against terrorism but are rather general);
ibid. at 2162.
Gross, ‘Chaos and Rules’ at 1065–6.
See Oren Gross, ‘Providing for the Unexpected: Constitutional Emergency Provisions’
(2003) 33 Isr. YB Hum. Rts. 13.
Dicey, Law of the Constitution at 272.
Ibid. at 272 (emphasis added).
Ibid. at 10–11 (arguing that Acts of Indemnity ‘being as it were the legalisation of illegality
are the highest exertion and crowning proof of sovereign power [by Parliament]’); see also
ibid. at 142.

But whereas the exceptional legislation discussed above calls for such legis-
lative aid to be accorded ex ante, legislative Acts of Indemnity give an after-
the-fact, retrospective, ex post aid to the executive.
What are we to make of Dicey’s ‘aid of Parliament’ approach? He would
certainly reject claims of inherent executive powers to deal with emergencies,
regarding them as certain to undermine parliamentary sovereignty and
supremacy, by leading to practically unfettered discretion and authority in
the hands of the executive branch of government.23
Significantly, Dicey also rejects what he calls ‘the doctrine of political
expediency’, namely the view that ‘during an invasion, a general, a mayor, a
magistrate, or indeed any loyal citizen, is legally justified in doing any act,
even though prima facie a tort or a crime, as to which he can prove to the
satisfaction of a jury that he did it for the public service in good faith, and for
reasonable and probable cause’.24 For Dicey, then, necessity does not, in and
of itself, make legal that which in other circumstances would have been illegal.
Even where public officials perform illegal actions to preserve and protect the
nation, that alone does not make their actions legal. This is a significant point
that must be considered in light of the tradition of the prerogative power
which is most famously associated with John Locke.

B. John Locke’s theory of the prerogative power
Locke’s theory of the prerogative power is significant not only because of its
intellectual potency, but also because it has greatly influenced many of the
Founding Fathers of the United States and their contemporaries. For Locke,
the prerogative power vested in the executive branch of government25 is
‘nothing but the power of doing public good without a rule’.26 It is the power
‘to act according to discretion, for the public good, without the prescription of
the law, and sometimes even against it’.27 Put somewhat differently ‘preroga-
tive can be nothing, but the people’s permitting their rulers, to do several
things of their own free choice, where the law was silent, and sometimes too
against the direct letter of the law, for the public good; and their acquiescing in

On the concept of inherent executive powers under the US Constitution see, for example,
Laurence H. Tribe, American Constitutional Law (2nd ed., New York, Foundation Press,
1988), 676; William C. Banks and Alejandro D. Carrio, ‘Presidential Systems in Stress:
Emergency Powers in Argentina and the United States’, (1993) 15 Mich. J. Int’l L. 1, 42–6;
Henry P. Monaghan, ‘The Protective Power of the Presidency’, (1993) 93 Colum. L. Rev. 1.
Dicey, Law of the Constitution, at 412.
‘Where the legislative and executive power are in distinct hands . . . there the good of the
society requires, that several things should be left to the discretion of him, that has the
executive power’. John Locke, The Second Treatise of Government in Mark Goldie (ed.),
Two Treatises of Government (London, Everyman, 1993) (first pub. 1690), para 159.
Ibid. para 166 (emphasis added). 27 Ibid. para 160 (emphasis added).

it when so done’.28 Locke considers such power as necessary in order to deal
with situations when strict and rigid observation of the laws may lead to grave
social harm.29
Locke offers a functional litmus test for evaluating whether the prerogative
power has been appropriately used in any given case. He focuses on the
purpose behind the exercise of the prerogative, i.e. whether it was directed
at promoting the public good.30 Government cannot have any legitimate ends
apart from promoting the good of the community. Governmental power
used for any purpose other than the public good is properly regarded as
tyrannical31 and may justify, under certain circumstances, an uprising to
restore the people’s rights and to limit the government’s resort to such
arbitrary power.32
Locke puts much faith in human reason and rationality as mitigating and
limiting factors on the exercise of prerogative power.33 His theory of the
prerogative reveals a substantial degree of trust in government in general, and
particularly in times of emergency. He gives the executive the benefit of the
doubt: if there are allegations that the ruler’s use of the prerogative power has
not been for the purpose of promoting the public good, but rather was in the
service of the ruler’s own interests and purposes the people have no remedy
available from any ‘judge on earth’. Their sole recourse is ‘to appeal to
heaven’ or, when the majority of the people feels wronged, to revolt against
the oppressive ruler.34 This is a tall order indeed.35 On the other hand, if it

Ibid. para 164 (emphasis added).
Ibid. paras 159–60. Locke suggests that it is ‘fit that the laws themselves should in some
cases give way to the executive power, or rather to this fundamental law of nature and
government, viz. that as much as may be, all the members of the society are to be
preserved’. Ibid. para 159. Explaining his reasons for vesting the prerogative power with
the executive, Locke argues that the legislature cannot anticipate in advance and regulate
by statute all that may be, at any point in the future, beneficial to society, and that
lawmaking power may be too slow to adapt adequately to exigencies and necessities of
the times.
Ibid. para 161 (‘But if there comes to be a question between the executive power and the
people, about a thing claimed as a prerogative; the tendency of the exercise of such
prerogative to the good or hurt of the people, will easily decide that question’); see also
ibid. paras 163, 164, 168.
Ibid. para 199 (‘tyranny is the exercise of power beyond right, which nobody can have a right
to. And this is making use of the power any one has in his hands; not for the good of those,
who are under it, but for his own private separate advantage’); see also ibid. 202
(‘Wherever law ends, tyranny begins, if the Law be transgressed to another’s harm’).
Ibid. paras 203–209. 33 Ibid. paras 163–64. 34 Ibid. para 168.
Locke recognizes that the right of the people to revolt against a ruler who abuses her
powers will likely be exercised on rare occasions, for ‘revolutions happen not upon every
little mismanagement in public affairs. Great mistakes in the ruling part, many wrong and
inconvenient laws, and all the slips of human frailty will be born by the people, without
mutiny or murmur.’ Ibid. para 225; see also ibid. paras 223, 230.

appears that the ruler used his prerogative power in an appropriate manner
the evaluation of his actions is straightforward: ‘prerogative can be nothing,
but the people’s permitting their rulers, to do several things of their own free
choice, where the law was silent, and sometimes too against the direct letter of
the law, for the public good; and their acquiescing in it when so done’.36 In
other words, when the ruler applies his prerogative power for the public
good, such action is considered the right thing to do. An appropriate exercise
of the prerogative power is legitimate per se and ex ante due to the implicit
acquiescence of the public to any such exercise (albeit not necessarily to the
specific use of the prerogative power in the circumstances of any particular
crisis). There is no need for any further public involvement.
This is where Dicey’s discussion of the Act of Indemnity departs from, and
improves on, the Lockean model of the prerogative. Whereas Locke seems to put
his trust in an implicit, general, ex ante public acquiescence in the exercise of an
executive power to act outside the law, Dicey, as we already saw, insists that an
explicit, particular, ex post legislative ratification of the same must be awarded.
While the Extra-Legal Measures model takes a further step by not pegging
ratification to an act of the legislature, it retains Dicey’s insistence that extra-
legal actions cannot be justified merely by reference to the motives of the actors,
laudable as these may be. Rather a separate and independent ex post ratification
must take place in order for the extra-legal action to be justified or excused.
There may be another important distinction between Locke’s discussion of
the prerogative and Dicey’s recognition of the possibility of an executive
power to act outside the law. One possible reading of Locke draws distinc-
tions between what is ‘extra-legal’ and that which is ‘extra-constitutional.’
According to this reading, while Locke may be willing to recognize govern-
mental actions that run contrary to positive law, i.e. extra-legal actions, he
does not consider the prerogative to be an extra-constitutional power. Rather,
he sees it as an integral part of the broader constitutional scheme.37 That
constitutional order, which enjoys a higher normative value than any parti-
cular set of positive laws, acknowledges the possibility of extra-legal govern-
mental action in times of emergency and necessity. Thus, the prerogative
power ought to be considered as a concrete (extreme to be sure) example of
the ‘political power’ – that power ‘which every man, having in the state of
nature, has given up into the hands of the society, and therein to the
governors, whom the society hath set over itself, with this express or tacit
trust, that it shall be employed for their good, and the preservation of their

Ibid. para 164 (emphasis added).
See, e.g., Carl J. Friedrich, Constitutional Reason of State: The Survival of the Constitutional
Order (Providence, Brown University Press, 1957) 110, 111 (1957) (Locke’s theory of
prerogative power not as an extra-legal power but rather as a power inherent in the
constitutional order, i.e. as a legal, albeit exceptional, power).

property’38 – as distinguished from ‘despotical power’.39 Locke’s warning
that exercise of the prerogative power by the good, even ‘godlike’, prince will
serve as a precedent for the application of similar powers by a lesser ruler
‘managing the government with different thoughts’,40 may also suggest that
he considers the prerogative power to be functioning within the boundaries
of the constitutional system rather than outside these boundaries. If it were
not so, could the implementation of extra-constitutional powers establish a
precedent for the future and coat with a cover of legality and legitimacy the
actions of a less worthy prince?41
This understanding of the prerogative power leads, in turn, to an apparent
dilemma: is there any point in identifying the prerogative power as a con-
stitutional power, when the substantive content of that power permits viola-
tion of constitutional norms?42 Specifically, if the power of prerogative
permits actions that are against the direct letter of the law, what prevents
the wielder of that power from exercising it in violation of the prescribed legal
limitations on the use of that very power, turning it into an unlimited power,
constrained neither by legal norms nor by principles and rules of the con-
stitutional order?43

III. Acts of indemnity and ex post ratification of extra-legal
actions by public officials
An Act of Indemnity may be one method of ex post ratification. More
generally, such ratification may be formal or informal, legal as well as social
or political.44 Legal modes of ratification include, for example, the exercise of
prosecutorial discretion not to bring criminal charges against the public
officials who acted in violation of the law, jury nullification where criminal
charges are brought, executive pardoning or clemency where criminal pro-
ceedings result in conviction, and governmental indemnification of state
agents who are found liable for damages. Political and social ratification is
also possible.

Locke, The Second Treatise of Government, para 171. 39 Ibid. para 172.
Ibid. para 166. Locke warns that so perilous may such consequences be that upon this is
founded that saying, that ‘the reigns of good princes have been always most dangerous to
the liberties of their people’. Ibid.
Gross, ‘Chaos and Rules’, at 1094–5.
Friedrich, Constitutional Reason of State at 83.
See, e.g., Sotirios A. Barber, On What the Constitution Means (Baltimore, Johns Hopkins
University Press, 1984), 188–90; Joseph M. Bessette and Jeffrey Tulis, ‘The Constitution,
Politics, and the Presidency’ in Bessette and Tulis (eds.), The Presidency in the
Constitutional Order (Baton Rouge, Louisiana State University Press, 1981), 3, 24–5.
Oren Gross, ‘Are Torture Warrants Warranted? Pragmatic Absolutism and Official
Disobedience’ (2004) 88 Minn. L. Rev. 1481; Gross, ‘Chaos and Rules’, at 1111–15.

Does ex post ratification render legal that which previously had been illegal,
or does it excuse the acting official from liability for her extra-legal actions
without making such actions legal? Much would depend on the nature of the
ratification. The answer to such questions would be made on a case-by-case
basis. Sticking with Dicey, this Part focuses on Acts of Indemnity as a method
of ex post ratification. It should be noted, however, that the Extra-Legal
Measures model goes beyond Dicey’s approach in so far as it does not
necessitate the ‘aid of Parliament’ to the executive. Whereas Dicey discusses
the Act of Indemnity as reflecting, ex post, the supremacy and sovereignty of
Parliament, the Extra-Legal Measures model is more open to the possibility of
ex post ratification taking place outside the legislature.
An Act of Indemnity, Dicey suggests, ‘legalises illegality’.45 Dyzenhaus
argues that Dicey’s notion of the Act of Indemnity places public officials in
a ‘legal black hole’ in as much as it allows ‘political power [to be exercised] in
a brute fashion (the last and supreme exercise of Parliamentary sovereignty)
permitting those who wield it to break free of the constraints of constitution-
ality and legality’.46 Rather than make what the acting officials did legal, the
Act of Indemnity places them in a ‘zone uncontrolled by law’.47 Compelling
as this statement may seem at first blush, it is problematic on several levels.
First, Acts of Indemnity may do one of two things: they may shelter the
acting public official from civil or criminal responsibility for her violations of
the law while holding that her actions were, and remain, illegal. Alternatively
they may seek to exculpate the actor from any legal responsibility for her
actions by making such actions, retrospectively, lawful. Dyzenhaus seems to
suggest that Dicey opts for the former interpretation, i.e. that Acts of
Indemnity deny otherwise available legal remedies for certain violations of
the law. That is certainly a plausible understanding of the nature and function
of the Acts, but it is not Dicey’s. In explaining his own view of Acts of
Indemnity and their effects, Dicey is quite clear: ‘Acts of Indemnity . . . are
retrospective statutes which free persons who have broken the law from
responsibility for its breach, and thus make lawful acts which when they were
committed were unlawful’.48 He rejects Frederick Pollock’s suggestion that an
Act of Indemnity is merely ‘a measure of prudence and grace. Its office is not
to justify unlawful acts ex post facto, but to quiet doubts, to provide com-
pensation for innocent persons in respect of damage inevitably caused by

Dicey, Law of the Constitution, at 10–11, 142.
Dyzenhaus, Chapter 4 at 65 (emphasis added).
Dicey, Law of the Constitution, at 142 (emphasis added). See also ibid. at 10 (‘An Act of
Indemnity is a statute, the object of which is to make legal transactions which when they
took place were illegal, or to free individuals to whom the statute applies from liability for
having broken the law’).

justifiable acts which would not have supported a legal claim’,49 calling it a
‘very inadequate description of an Act of Indemnity ’.50 This surely fits with
Dicey’s overall argument for the supremacy of law since an ‘Act of Indemnity,
again, though it is the legalisation of illegality, is also . . . itself a law . . . It is
no doubt an exercise of arbitrary sovereign power; but where the legal
sovereign is a Parliamentary assembly, even acts of state assume the form of
regular legislation.’51 Thus, by ‘making lawful acts which when they were
committed were unlawful’, the Act of Indemnity ensures that all actions by
public officials are done under a legislative framework and are not, at the end
of the day, to be found outside the law.
Second, if by ‘it’ (as in ‘those who wield it’) Dyzenhaus refers to political
power then it seems to me that he fails to distinguish between the actions of
public officials that are taken outside the law and the subsequent acts of
Parliament. For the exercise of the political power in a way that amounts to
‘the last and supreme exercise of Parliamentary sovereignty’, i.e. the enact-
ment of an Act of Indemnity, is carried out by Parliament whereas those who
are permitted ‘to break free of the constraints of constitutionality and legality’
are surely those public officials who may find themselves compelled to act
extra-legally when dealing with an emergency. But then again those public
officials are not those who wield ‘it’, i.e. the political power to pass an Act of
Indemnity and thus express the ‘last and supreme exercise of Parliamentary
sovereignty’. This may seem as no more than a hair-splitting exercise and an
overly pedantic reading. However, I suggest that the view that Dyzenhaus
expresses in the passage quoted above – namely that ex post ratification of
extra-legal acts of public officials, such as done by way of Acts of Indemnity,
empowers public officials to ‘break free of the constraints of constitutionality
and legality’ – misses some of the essential components of the Extra-Legal
Measures model precisely because it does not pay sufficient attention to the
distinction between the extra-legal action (the act of official disobedience)
and the possibility (but not certainty) of a subsequent, ex post, ratification of
such actions by the legislature or, under the Extra-Legal Measures model, the
One of several themes running through Dyzenhaus’s argument is that the
Extra-Legal Measures model does not offer any meaningful method to main-
tain constitutional and legal constraints over public officials. Does not that
model merely amount (or, perhaps more appropriately, descend) to legit-
imating the exercise of ‘brute’ power uncontrolled by law? If we accept the
possibility, in extreme cases, of certain governmental actions that are extra-
legal, can there be any constitutional or legal limitations on such govern-
mental exercise of power? Extra-legal power can only mean an unlimited

Frederick Pollock, ‘What is Martial Law?’ (1902) 70 L. Q. Rev. 152, 157.
Dicey, Law of the Constitution at 414. 51 Ibid. at 145.

power, constrained neither by any legal norms nor by principles and rules of
the constitutional order!
I have defended the Extra-Legal Measures model elsewhere.52 What I wish
to do here is to address the charge regarding the perceived freedom of the
constraints of constitutionality and legality, which the model seems to offer
to public officials. I do so by using as my starting base Dicey’s own arguments
in support of the use of Acts of Indemnity.
To make the argument clearer let me rephrase the challenge to the Extra-
Legal Measures model in the concrete context of Acts of Indemnity: if the
effect of Acts of Indemnity is, indeed, to legalize illegality and if, as Dicey
openly acknowledges, the expectation of the executive that such Acts will be
passed by Parliament ‘has not been disappointed’ as a matter of history and
experience,53 what limits remain on the exercise of powers by the government
and its agents? After all, public officials will always, so it seems, be able to act
outside the law, without legal and constitutional shackles on their actions,
and be secure in the knowledge that an Act of Parliament shielding them from
all responsibility and making their actions lawful is forthcoming.
Dicey is aware of this critical challenge. ‘Still,’ he argues, ‘there are one or
two considerations which limit the practical importance that can fairly be
given to an expected Act of Indemnity. The relief to be obtained from it is
prospective and uncertain.’54 I suggest that the uncertainty of the prospective
relief may not only slow down the rush to act extra-legally in the first place,
but it also facilitates meaningful limitations on such actions once they are
taken and provide important benchmarks against which to evaluate such
actions ex post. To acknowledge the possibility of extra-legal action is not the
same thing as accepting willy-nilly limitless powers and authority in the
hands of state agents.
In a democratic society, where values of constitutionalism, accountability,
the rule of law, and individual rights, freedoms and liberties are firmly
entrenched and traditionally respected, we can expect that the public would
be circumspect about governmental attempts to justify or excuse illegal
actions even if taken, arguably, to promote the general good. That being the
case, ‘any suspicion on the part of the public, that officials had grossly abused
their powers, might make it difficult to obtain a Parliamentary indemnity for
things done’.55 The Act of Indemnity may never be enacted.

Gross, ‘Chaos and Rules’, at 1118–34.
Dicey, Law of the Constitution, at 144 (noting that an Act suspending the Habeas Corpus
Act ‘has constantly been followed by an Act of Indemnity’); see also Mitchell v. Clark, 110
US 633, 640 (1884) (noting that Acts of Indemnity are ‘passed by all governments when
the occasion requires it’).
Dicey, Law of the Constitution, at 144–5 (emphasis added).
Ibid. at 145.

The separation between extra-legal actions taken by public officials in
catastrophic cases and subsequent public ratification – for example by way
of the passage of an Act of Indemnity – creates a ‘prudent obfuscation’56 that
introduces a significant element of uncertainty to the decision-making
calculus of public officials and raises both the individual and national costs
of pursuing an extra-legal course of action.
With the need to obtain ex post ratification, the public official who decides
to act extra-legally undertakes a significant risk because of the uncertain
prospects for subsequent ratification. The public may, for example, disagree
after the fact with the acting official’s assessment of the situation and the need
to act extra-legally. Ratification would be sought ex post, when more infor-
mation about the particular case at hand may be available to the public, and
possibly after the particular danger has been eliminated. Under such circum-
stances, it is possible that calm and rationality, rather than heightened emo-
tions, would govern public discourse, emphasizing the risk for the official in
acting first and seeking approval later. The public may also determine that the
extra-legal actions violated values and principles that are too important to be
encroached upon, as a matter of principle or in the circumstances of the
particular case. The greater the moral and legal interests and values infringed
upon, the less certain the actor can be of securing ratification.
What, then, about Dicey’s, and Dyzenhaus’s,57 observation that as a matter
of practice the expectation of the executive that Parliament will pass an Act of
Indemnity ‘has not been disappointed’? Does not that eliminate any uncer-
tainty on the part of the public officials or at least create a significant risk of
Even if we accept that there exists a good chance that ex post ratification
will be forthcoming, there are still significant costs to acting extra-legally. For
starters, there is still a certain degree of anxiety that ratification will not, in
fact, follow. More significantly, ratification may not be comprehensive or
fully corrective. As Dicey notes: ‘As regards . . . the protection to be derived
from the Act by men who have been guilty of irregular, illegal, oppressive, or
cruel conduct, everything depends on the terms of the Act of Indemnity.’58

Dan M. Kahan, ‘Ignorance of Law Is an Excuse – But Only for the Virtuous’ (1997) 96
Mich. L. Rev. 127, 139–41 (discussing ‘prudent obfuscation’ as a means to respond to the
penal law’s persistent incompleteness). Kahan discusses the use of vague terms in criminal
laws, giving courts ‘the flexibility to adapt the law to innovative forms of crime ex post’.
Ibid. at 139.
Dyzenhaus, Chapter 4, at 72–3 (‘If the Extra-Legal Measures model were public, as it must
be if it is to promote deliberation, the expectation would be generated of after-the-fact
validation of illegal official acts. In an atmosphere of fear that expectation would likely be
met rather easily, especially when the threat is, or is claimed to be, a constant one and the
government successfully manipulates public opinion.’)
Dicey, Law of the Constitution, at 145 (emphasis added).

Subsequent ratification may shield the actor against criminal charges, but not
bar the possibility of civil proceedings. It may also not shield the actor from
liability for all of her actions.59
In addition, if an Act of Indemnity is enacted, then
The fact that the most arbitrary powers of the English executive must
always be exercised under Act of Parliament places the government, even
when armed with the widest authority, under the supervision, so to speak,
of the courts. Powers, however extraordinary . . . are never really unlim-
ited, for they are confined by the words of the Act itself, and, what is more,
by the interpretation put upon the statute by the judges.60

The fact that the courts will be called to deal with an ex post ratification in the
form of an Act of Indemnity may also make judicial supervision over the
exercise of emergency powers by government officials more meaningful and
robust. In times of emergency courts, both domestic and international,
assume a highly deferential attitude when called upon to review govern-
mental actions and decisions.61 The courts’ apparent inability to protect

Ibid. at 145. Similarly, when ratification comes in the form of an executive pardon or
clemency, it eliminates the criminal penalty that was imposed on the individual actor, but
it neither removes the ordeal of criminal prosecution nor the condemnation associated
with criminal conviction. See Leon Sheleff, ‘On Criminal Homicide and Legal Self
Defense’ (1997) 6 PLILIM 89, 111–12; Yale Kamisar, ‘Physician Assisted Suicide: The
Problems Presented by the Compelling, Heartwrenching Case’ (1998) 88 J. Crim. L. &
Criminology 1121, 1143–4 (reliance on mitigation of sentence fails to mitigate the ‘ordeal
of a criminal prosecution or the stigma of a conviction’).
Dicey, Law of the Constitution at 273.
See, e.g., Thomas M. Franck, Political Questions/Judicial Answers: Does the Rule of Law
Apply to Foreign Affairs? (Princeton University Press, 1992), 10–30; Harold Hongju Koh,
The National Security Constitution: Sharing Power After the Iran-Contra Affair (New
Haven, Yale University Press, 1990), 134–49; Christopher N. May, In the Name of War:
Judicial Review and the War Powers Since 1918, (Cambridge, Harvard University Press,
1989), at 261–4 (speaking of ‘ritualistic approval’ by courts of governmental emergency
measures); William H. Rehnquist, All the Laws but One (New York, Knopf, 1998), 221–2;
Michal R. Belknap, ‘The Warren Court and the Vietnam War: The Limits of Legal
Liberalism’ (1998) 33 Ga. L. Rev. 65, 66–7; Anne-Marie Slaughter Burley, ‘Are Foreign
Affairs Different?’ (1993) 106 Harvard Law Review 1980, 1991–5; John Hart Ely, War and
Responsibility: Constitutional Lessons of Vietnam and Its Aftermath (Princeton University
Press, 1993), 54–60; Laurence Lustgarten and Ian Leigh, In from the Cold: National Security
and Parliamentary Democracy (Oxford, Clarendon, 1994), 320–59; George J. Alexander,
‘The Illusory Protection of Human Rights by National Courts During Periods of
Emergency’ (1984) 5 Hum. Rts. L. J. 1, 15–27. For the argument that international and
regional judicial bodies are not necessarily more effective in dealing with the concept of
‘emergency’ than are domestic courts, see Fionnuala Ni Aolain, ‘The Emergence of
Diversity: Differences in Human Rights Jurisprudence’ (1995) 19 Fordham Int’l L. J.
101; Oren Gross, ‘Once More unto the Breach: The Systemic Failure of Applying the
European Convention on Human Rights to Entrenched Emergencies’ (1998) 23 Yale J.
Int’l L. 437, 490–500.

individual rights while extreme violence is raging around them, compared
with their greater willingness to resume their role as guardians of human
rights and civil liberties once the crisis is over, may mean that judicial review
of Acts of Indemnity may be relatively more meaningful.62
When we consider international legal rules and norms, the costs and
uncertainties that are involved in acting extra-legally are increased further.
Thus, even if a particular extra-legal act is domestically ratified ex post, it may
be subject to a different judgment on the international plane. This may have
significant consequences both for the individual public official and her
government. Acting officials may potentially still be subject to criminal and
civil proceedings in jurisdictions other than their own. Moreover, to the
extent that the relevant extra-legal action violates the nation’s international
legal obligations, especially its obligations and undertakings under the major
international human rights conventions, and is not covered by an appro-
priate derogation (or, indeed, it violates a non-derogable right), state agents
who engage in such acts expose their government to a range of possible
remedies under the relevant international legal instruments. Indeed by recog-
nizing what has occurred, a domestic Act of Indemnity may facilitate inter-
national litigation.
The Extra-Legal Measures model imposes significant burdens on public
officials. They must act in the face of great uncertainty. They may still decide
to act extra-legally ‘for the public good’ and expect to be protected subse-
quently by a form of ex post ratification. At the same time, the model makes it
extremely costly to resort to such drastic measures, limiting their use to
exceptional exigencies. As Sanford Kadish notes, ‘Would not the burden on
the official be so great that it would require circumstances of a perfectly
extraordinary character to induce the individual to take the risk of acting?
The answer is of course yes, that’s the point.’63 The more uncertain it is that
ratification will be forthcoming, the more uncertain its potential scope, and
the greater the personal risk involved in wrongly interpreting either of those
is, the greater the incentive for individual actors to conform their action to
the existing legal rules and norms and not risk acting outside them. The
burden lies squarely on the shoulders of the public officials who must act,

See, e.g., May, In the Name of War at 268 (suggesting that, in light of judicial practice of
abdicating review of executive activities during an emergency, ‘courts should steer a
middle course and defer review until the emergency has abated’). Chief Justice
Rehnquist also noted: ‘If, in fact, courts are more prone to uphold wartime claims of
civil liberties after the war is over, may it not actually be desirable to avoid decision on
such claims during the war? . . . While the body of case law might benefit from such
abstention, those who are actually deprived of their civil liberties would not. But a decision
in favor of civil liberty will stand as a precedent to regulate future actions of Congress and
the Executive branch in future wars.’ Rehnquist, All the Laws but One at 222.
Sanford H. Kadish, ‘Torture, the State and the Individual’ (1989) 23 Isr. L. Rev. 345, 355.

sometimes extra-legally, without the benefit of legal pre-approval of their
actions by the courts or the legislature. Public officials have no one to hide
behind. They must put themselves on the front line and act at their own peril.
Their burden is composed not merely of political norms of accountability,
transparency and democratic deliberation, but also of legal norms. The argu-
ment that such an approach puts public officials in a ‘zone uncontrolled by
law’ underestimates the significance of such disincentives to step outside the
legal framework and of the possibilities for external supervision, both by the
public and by other branches of government, which this approach has to offer.
The limitations on the resort to extra-legal actions that the Extra-Legal
Measures model offers are further strengthened by the fact that we can and
should expect public officials to feel quite uneasy about possible resort to
extra-legal measures, even when such actions are deemed to be for the
public’s benefit. The knowledge that acting in a certain way means acting
unlawfully is, in and of itself, going to have a restraining effect on government
agents, even while the threat of catastrophe persists. Furthermore, in order to
enjoy ex post ratification some disclosure as to the extra-legal actions that
were taken during the crisis and some justification for taking such actions in
the first place are likely to be demanded (and, as noted above, disclosure and
justification form critical components of the Extra-Legal Measures model).
The need to give reasons ex post – to publicly justify or excuse, not merely
explain, one’s actions after the fact – may limit the government’s choice of
measures ex ante, adding another layer of restraint on governmental action.64
In any event it emphasizes the accountability of government agents. It will
also contribute to open and reasoned discourse and dialogue between the
different branches of government, between the government and its domestic
constituency, between the government and other governments, and between
the government and non-governmental or international organizations.65
Consider, once again, the issue of Acts of Indemnity. The legislative process
that leads to the enactment of such Acts presents the legislature with a unique
opportunity to review the actions of the executive branch and assess them
ex post, relieved from the pressures of the crisis, before deciding whether to
ratify them. Furthermore, the legislative process may also invoke public
deliberation and force the legislative branch to take an affirmative stand on
issues connected with the emergency. This may counteract the reluctance of
legislatures to assume responsibility or supervise the executive in times of

See Frederick Schauer, ‘Giving Reasons’ (1995) 47 Stan. L. Rev. 633, 656–7; see also
Bessette and Tulis ‘Constitution, Politics’, 10.
The reasons put forward by a state to justify its actions may be subject to scrutiny not only
by other governments and non-governmental organizations, but also by judicial and
quasi-judicial bodies such as the European Court of Human Rights, the Inter-American
Court of Human Rights and the United Nations Human Rights Commission.

emergency.66 In addition, the judiciary has a role to play in interpreting the
scope of the Act of Indemnity.
I cannot end this discussion without briefly commenting on Dyzenhaus’s
most provocative example of the ‘all bets are off’ critique of the Extra-Legal
Measures model, namely the story of the ‘Night of the Long Knives’.67 The
gist of his critique is that the model cannot impose any limitations on actions
such as those taken by Hitler and his henchmen since those were publicly
acknowledged (in fact, boasted about) and received a subsequent statutory
ratification through a special Act promulgated by the German Cabinet, acting
under the authority of the Enabling Act of 1933 that vested the German
government with legislative powers.
Yet, there is very little, if any, point in trying to bring rational discussion –
to talk about the rule of law, legality and legitimacy, rights, and limitations on
powers – to the experience of Nazi Germany. Dyzenhaus is absolutely correct
in attributing to the Extra-Legal Measures model a substantive precondition,
namely that it is applied and used in a community that is ‘worth saving’. But
that, surely, is a condition that ought to underlie any meaningful discussion
of emergency powers.
Writing during the early days of the Cold War, Carl Friedrich, a Harvard
University professor of political science, described the tension between
national security and civil rights and liberties as arising ‘wherever a constitu-
tional order of the libertarian kind has been confronted with the Communist
challenge, and with the Fascist response to that challenge’.68 In other words,
to what extent, if any, can violations of liberal democratic values be justified
in the name of the survival of the democratic, constitutional order itself; and
if they can be so justified, to what extent can a democratic, constitutional
government defend the state without transforming itself into an authoritar-
ian regime? The tension between self-preservation and defending the ‘inner-
most self’ of the democratic regime – those attributes that make the regime
worth defending – presents decision-makers with tragic choices.69 However,
this tension, which is at the heart of all discussions of emergency powers, can
only be captured by those who share the belief in the viability and desirability
of a constitutional, liberal, democratic regime while taking cognizance of the
fact that emergencies require special treatment that may deviate from the
ordinary norms.

Koh, The National Security Constitution, at 117–33.
Dyzenhaus, Chapter 4, at 70.
Friedrich, Constitutional Reason of State, at 13.
Ibid. at 13 (noting that the survival of a constitutional order involves more than self-
preservation due to the rational and spiritual content of this order); see also Pnina Lahav,
‘A Barrel Without Hoops: The Impact of Counterterrorism on Israel’s Legal Culture’
(1998) 10 Cardozo L. Rev. 529, 531 (noting the ‘tragic dimensions of the tension between
terrorism, counterterrorism, and justice in any democratic society’).

Terrorism, risk perception and judicial review

I. Introduction
The legal response to the threat of terrorism has been driven, in large part, by
public fear about future attacks and worst-case scenarios. There are, of course,
many risks that ought to be taken seriously by governments. But all too often
policy responses are motivated by a widespread public misperception of risk
and a heightened collective sense of fear and vulnerability that call into
question our ability to think clearly about policy options. In this chapter,
I reconsider the role of the legislative, executive and judicial branches of
government in an emergency, first by considering how misperception of risk
and public fear influence policy-makers, and then by examining the role that
judicial review can play in times of crisis.
One response to public fear is to respond legislatively to popular opinion,
enacting strict anti-terrorism measures. As democratic as this option might
first seem, it is problematic because, as empirical research shows us, social
forces amplify and distort our judgments about risk, particularly in emotion-
ally charged situations. Only on a thin, populist conception, could democracy
be seen simply as an aggregating mechanism for mere popular opinion,
rather than as a sophisticated system to promote public deliberation and
ensure that public decisions are fair and informed ones. Another approach,
one that recognizes the technocratic nature of lawmaking in the regulatory
state and relies on the integrity of the government and its expert advisors, is to
defer to the executive on matters of national security. The problem with this
option is its assumption that the executive would be competent to assess
the risks in question, and that it would be trustworthy and unbiased. All of
these assumptions are generally suspect, perhaps even more so in an
The approach that I defend in this chapter vests policy-making power in a
well-informed executive that is fully accountable to the courts (or, at the very

I am grateful to Alan Khee-Jin Tan and David Dyzenhaus for their comments on a draft and
to the participants in the Comparative Anti-Terrorism Law & Policy Symposium for their
provocative comments.


least, a specialized, independent tribunal), whose decisions are in turn subject
to public scrutiny and debate. This option allows for a careful consideration
of risks and responses in a forum in which limitations on liberty are given
their due and where the effectiveness of the policy can be assessed against
limitations on rights, ideally in full public view. Although this judicial review
option is largely reactive, it is a critical institutional safeguard against policy-
making motivated primarily by fear and, more broadly, contributes to an
inter-branch and broader societal dialogue on anti-terrorism measures. It
also positions the courts as an integral institution in a democracy, ensuring
that collective decision-making takes place in a setting in which assumptions
about risks are laid bare and substantive values are fully considered.
The next part of this chapter draws on the contemporary literature on
perception of risk to show how misperception often leads to laws that are ill-
advised from a risk-reduction perspective. In this part, I explain why populist
and supposedly democratic responses are inadequate. I then consider the
executive option, with particular emphasis on claims that the executive has a
particular expertise on matters of national security and that, particularly in an
emergency, trust in and deference to the executive are imperative. Finally,
I defend judicial review against objections that the courts have neither the
institutional competence nor the capacity to deal with sensitive intelligence
information. In particular, I argue that both the courts and, in some cases,
specialized administrative tribunals, provide a crucial institutional safeguard
against policies motivated primarily by fear.

II. Risk perception and the limits of democratic populism
How much are we willing to sacrifice to ensure that something catastrophic
doesn’t happen? As tragic as the September 11 terrorist attacks in New York
and Washington – and subsequently in places as diverse as Indonesia,
Pakistan, Kenya, Turkey, Spain and Saudi Arabia – have been, the haunting
thought that seems to be driving legislative responses to terrorism is that
something worse might be in store.1 And if something worse is in store, why
shouldn’t we do everything in our power to stop it, even if it means sacrificing
our fundamental values and principles?
This way of thinking draws its persuasive power by its appeal to fear and to
the way in which we perceive risk. Research on the perception of risk

See, for example, Nicholas D. Kristol, ‘Risks of a nuclear Sept. 11 are increasing’
International Herald Tribune (11 March 2004), at 7; ‘Global terrorism must be tackled,
whatever the cost’ The Straits Times (Singapore) (6 March 2004), at 22 (reporting British
Prime Minister Tony Blair’s view ‘that governments could not ‘‘err on the side of caution’’
when dealing with the threat of terrorism and weapons of mass destruction’ and Foreign
Secretary Jack Straw’s view that ‘international terrorism represented a new scale of threat’).

describes a broad range of phenomena which influence perception of risk.2
I want to focus on two of these, specifically, the psychological factors, such as
probability neglect, that bear on risk perception, and the social amplification
of risk, both of which have direct implications for anti-terrorism law and

A. The psychology of risk perception
In his recent work on risk perception in the context of terrorism, Cass
Sunstein, drawing on a vast body of multi-disciplinary literature, discusses
three psychological responses to acts of terrorism: the use of heuristics, the
tendency to show a disproportionate fear of unfamiliar risks, and the ten-
dency, when strong emotions are involved, to focus on the bad outcome
rather than on the probability that it will occur (‘probability neglect’).3
Sunstein focuses on the last phenomenon, but all three have important
implications for anti-terrorism law and policy.
Consider first the use of heuristics. Sunstein explains: ‘In the face of
ignorance, people assess probabilities through the use of various heuristics,
most notably the availability heuristic, in accordance with which probability
is measured by asking whether a readily available example comes to mind.’4
As we observed in the aftermath of the September 11 attacks, people tend to
think that another such attack is especially likely, whether or not it is in fact,5
as is shown by the dramatic drop in air travel in the months following the
attacks. Second, people ‘show a disproportionate fear of risks that seem
unfamiliar and hard to control’.6 As Sunstein explains, a terrorist attack can
cause significant changes in private and public behaviour ‘even if the magni-
tude of the risk does not justify those changes, and even if statistically
equivalent risks occasion little or no concern’.7 A third response is that
when strong emotions are involved, people focus on ‘the bad outcome itself,
and they are inattentive to the fact that it is unlikely to occur’.8 Sunstein calls
this phenomenon ‘probability neglect’ and argues that it is highly likely to
occur in the aftermath of a terrorist attack.9
These three responses to unfortunate and rare events such as terrorist
attacks suggest that people make significant mistakes about risk, particularly
in situations where strong emotions are involved. Indeed, some experts on
risk perception argue more generally that our emotional responses to an
activity influence our perception of the risk associated with that activity.

Paul Slovic, ‘Introduction and Overview,’ in Paul Slovic (ed.), The Perception of Risk
(London and Sterling, Earthscan Publications, 2000).
Cass Sunstein, ‘Terrorism and Probability Neglect’ (2003) 26 The Journal of Risk and
Uncertainty 121.
Ibid. at 121. 5 Ibid. 6 Ibid. 7 Ibid. at 122. 8 Ibid. 9 Ibid.

Paul Slovic uses the term ‘affect heuristic’ to refer to the mental short-cuts
people use when they draw on their affective or emotional responses when
making risk-benefit judgments and suggests that, particularly when emotions
are involved, people are unlikely to find an activity to be both risky and highly
beneficial.10 While the research in this area raises important questions about
the nature of emotions and their relation to cognition,11 it supports the more
general point that ‘people’s reactions to risks are often based mostly on the
harms caused by the potential outcome and the vividness of that outcome
rather than on the probability of its occurrence’.12

B. The social amplification of risk
Not only do terrorist acts have a direct impact on the way individuals think
about risk, the way that information about a terrorist attack is transmitted
can amplify the perception or risk on a social level. The social amplification of
risk refers to the ‘social structures and processes of risk experience, the
resulting repercussions on individual and group perceptions, and the effects
of those responses on community, society and economy’.13 The basic claim
made by risk researchers is that an unfortunate event such as an accident,
discovery of pollution, sabotage, or product tampering, interacts ‘with psy-
chological, social, and cultural processes in ways that can heighten or attenu-
ate public perception of risk and related risk behaviour’.14 In particular, the
flow of information, particularly through the media, has an important effect
on our perceptions of risk and is ‘a major agent of amplification’.15 Social
amplification is thus said to be influenced by the ‘volume [of information],
the degree to which the information is disputed, the extent of dramatization
and the symbolic connotations of the information’.16 Writing before 9/11, risk
researchers used events such as the nuclear reactor accident at Chernobyl in
1986 to illustrate how the dramatization of the event, together with sensa-
tional media headlines, increased the ‘memorability of that accident and the
perceived catastrophic potential of nuclear power’.17 These lessons are even
more poignant after September 11.
Drawing on this research and other recent work on information cascades,
Sunstein demonstrates very clearly the social origins of individual belief and the

Slovic, The Perception of Risk, xxxi, and in the same volume, Melissa L. Finucane, ‘The
Affect Heuristic in Judgments of Risks and Benefits’, 413–29.
Cass Sunstein, ‘The Laws of Fear’ (2002) 115 Harvard Law Review 1119 at 1140.
Ibid. at 1141.
Roger E. Kasperson et al., ‘The Social Amplification of Risk: A Conceptual Framework’, in
Slovic, The Perception of Risk, 232–45.
Ibid. at 234. 15 Kasperson, ‘Social Amplification’, at 241.
Ibid. 17 Ibid. at 242.

way in which individuals ‘contribute to the intensity of the very forces by which
they are influenced’, particularly when individuals have little first-hand informa-
tion. He uses the following stylized example to illustrate his point:
Ann is unsure whether global warming is a serious problem, but Bob,
whom Ann trusts, believes that it is. Influenced by Bob’s views, Ann
concluded that global warming is indeed a serious problem. Carl is
inclined, on his own, to discount the risk; but confronted with inform-
ational signals given by the shared views of Ann and Bob, Carl might well
come to believe that global warming is indeed a serious problem. Deborah,
a skeptic about global warming, would need a great deal of confidence in
the correctness of her view to reject the shared belief of Ann, Bob, and Carl.
The members of this little community will come to share the belief that
global warming is a matter of considerable concern.18
Sunstein concludes that most of us ‘think and fear what we do because of
what we think other people think and fear’.19 He goes on to show that
reputational influences might also contribute to the social amplification of
risk. In the aftermath of 9/11, he argues, a safety expert may well choose to
remain silent rather than pointing out, to the detriment of his or her reputa-
tion, that new security measures at airports, by making air travel less con-
venient, will force more people to drive, causing a net loss of lives.20 What is
clear from the literature on the social amplification of risk, however, is that
the mistakes that ordinary people are prone to make in their judgments about
risks can be compounded by a range of social influences which, in the case of
terrorism, may well heighten the public perception of risk.

C. Ordinary judgments about risk
Does this mean that ordinary public judgments about risk are generally
misguided? Some risk experts, notably Slovic, resist this conclusion, arguing
that risk ‘does not exist ‘‘out there,’’ independent of our minds and cultures’21
and that ‘there is wisdom as well as error in public attitudes and perceptions’
since the layperson’s ‘basic conceptualization of risk is much richer than that
of the experts and reflects legitimate concerns that are typically omitted from
expert risk assessments’.22 Slovic therefore concludes that risk-management
efforts should involve the public and that each side, the experts and the
public, ‘must respect the insights and intelligence of the other’.23

Sunstein, ‘The Laws of Fear’, at 1132.
Ibid. at 1132–3. It may be, then, that the misperception of risk is compounded in that not
only do we tend to misperceive risk, but we misperceive what others perceive as risks. I am
grateful to David Dyzenhaus for alerting me to this point.
Ibid. at 1134. 21 Slovic, Perception of Risk, at xxxvi.
‘Perception of Risk’ in Slovic, Perception of Risk, at 231. 23 Ibid.

Sunstein challenges both of these claims. As far as the subjectivity of risk is
concerned, he concedes that there is uncertainty and normativity in science
and in the way that risks are presented, but argues that such uncertainty and
normativity does not support Slovic’s broader claim that there is no objective
risk.24 We all recognize the dangers of smoking, driving under the influence
of alcohol, and crossing a busy road against the traffic lights, however difficult
the corresponding risks might be to quantify.25 More important for our
purposes, though, is Sunstein’s claim that ordinary people do not display
a ‘rival rationality’ and that experts are, in their area of expertise, more likely
to be right.26
Sunstein highlights two problems with Slovic’s argument. First, he chal-
lenges the methodology of Slovic’s empirical studies which purport to
demonstrate that ordinary people have a richer rationality. He argues, for
instance, that these studies do not disprove competing and equally plausible
explanations of their judgments of risk and he challenges some of the quali-
tative factors (such as whether a risk involves ‘dread’), purportedly used by
ordinary people in making judgments about risk, which Sunstein says need
conceptual unpacking.27 Second, Sunstein concedes Slovic’s general point,
that qualitative factors (for example, whether children are at risk and the
extent to which the activity is involuntary) are important in assessing risk.
But his more general argument, largely confirmed by Slovic’s own work, is
that ordinary people tend to err on the basic factual question of how large a
risk is.28 The use of heuristics, the tendency disproportionately to fear
unfamiliar risks, and the phenomenon of probability neglect all support
this general conclusion.29
If we are apt to make mistakes in our judgments about risk in everyday life,
then we are all the more likely to make mistakes about the risks associated
with terrorism. This is not to say that there is no risk associated with terrorist
acts, but only to underscore that in the aftermath of high-profile terrorist
attacks, the public response and level of fear will generally be heightened and
our perception of the risks associated with such events will be greatly dis-
torted. A corollary point, which has become even clearer after Richard A.
Clarke’s testimony before the 9/11 commission, is that where a potential

Sunstein, ‘The Laws of Fear’, at 1146–7.
Sunstein argues that even Slovic would have to agree that there is a ‘real risk’ in smoking
three packs of cigarettes a day over a period of years or a small risk of getting killed by a
shark while swimming at the beach in Marblehead, Massachusetts, even if these risks
cannot be quantified with precision (Sunstein, ‘The Laws of Fear’, at 1147). Sunstein’s
position here is not a metaphysical argument about the objectivity of risk, but his basic
position is convincing. There are, as even Slovic admits, dangers and uncertainties in life,
and, argues Sunstein, ‘if dangers are real, so are the risks’ (ibid.).
Ibid. at 1147ff. 27 Ibid. 1147–50, 1152–5. 28 Ibid. at 1150.
Ibid. at 1150ff and text accompanying note 3, above.

danger is well out of public view, both the public and their representatives in
governments are not likely to take it as seriously.30 This tendency on the part
of the public to see high-profile events as greater risks and latent, low-profile
dangers as minimal risks has profound implications for policy-making in a

D. Anti-terrorism law and policy in a democracy
If a high-profile event, such as a terrorist attack, is likely to lead to a
heightened public fear of terrorism, then in a populist democracy, a high-
profile attack will inevitably be followed by a spate of anti-terrorism laws,31
hastily enacted.32 This tendency to look to the law as the solution to distress-
ing events is not unique to the terrorism context, and has been described by
criminologists as ‘govern[ing] through crime’,33 as politicians have an incen-
tive to seek heavier penalties in the wake of horrific crimes.34 The reaction of
politicians to public emotion and fear exposes a deeper flaw in the workings
of modern democracies because politicians can activate, as Philip Pettit puts
it, ‘a politics of passion in which they appear as the only individual or the only
group really concerned about the sort of horrible crime in question’ and can
thus let loose ‘a rule of knee-jerk emotional politics that works systematically
against the common good’.35
Recognition of the dysfunctional nature of populist democracy is, of
course, not new. Political and legal theorists have long recognized and
attempted to address the dangers associated with the tyranny of the majority,
while some contemporary proponents of deliberative democracy attempt
to conceive of democracy ‘as a system for empowering the public reasons
recognized among a people, rather than the will of that people considered
as a collective agent’.36 These efforts have in common their rejection of
popular sovereignty and populism as the essence of democracy. They attempt

‘Excerpts of Testimony from Richard Clarke, Former National Coordinator for
Counterterrorism’ in Steven Strasser (ed.), The 9/11 Investigations (New York, Public
Affairs, 2004), 174–6.
On the anti-terrorism legislation enacted post-September 11, see generally ‘Special
Feature: Terrorism, Security and Rights: Anti-Terrorism Legislation in Singapore,
Britain, Malaysia, South Africa, Canada, Australia, and the United States’ (2002)
Singapore Journal of Legal Studies 1–270.
Philip A. Thomas, ‘9/11: USA and UK’ (2003) 26 Fordham International Law Journal 1193.
Roach, Chapter 7, in this volume, and his September 11: Consequences for Canada
(Montreal and Kingston, McGill-Queen’s University Press, 2003), at 24, referring to
Jonathan Simon, ‘Governing Through Crime’ in L. Friedman and G. Fischer (eds.), The
Crime Conundrum (New York, Westview Press, 1997), at 174.
Philip Pettit, ‘Depoliticizing Democracy’ (2003) 7 Associations: Journal for Legal and Social
Theory 23 at 26.
Ibid. 36 Ibid. at 32–3.

instead to develop a more substantive vision of democracy, which includes
mechanisms from constitutional constraints to other sorts of institutional
‘deliberation filters’,37 such as consultative procedures and arms-length
bodies, to sift out policies that are inconsistent with more enduring public
The literature on perception of risk and the public response to terrorist
acts gives us yet another reason to be suspicious of a populist approach to
policy-making in the anti-terrorism context. It suggests that lay persons
are likely to overreact in the immediate aftermath of a terrorist attack and
that the legislative branch, as the institutional embodiment of popular
sovereignty, is ill-suited to take the lead in policy-making in response.
Whatever attempts are made to refine democracy to get past these difficulties,
they necessarily involve a radical departure from populism, a rejection of the
view that fact of widespread public fear is sufficient justification for anti-
terrorism laws, and a search for mechanisms to transform popular opinion
into informed and fully considered judgments.39
While the literature on perception of risk, and particularly Sunstein’s work
in this area, is unsupportive of a populist response to acts of terrorism, it
leaves open another possibility, namely that anti-terrorism policy ought to be
made largely by technocrats with the requisite expertise. It suggests deference
to those who have access to the relevant information and the expertise to
interpret it. In short, it leaves open the view that there ought to be an increase
in and deference to the executive in times of crisis.

III. National security and executive power
Conferral of greater power on the executive can take a number of forms. For
instance, in many modern legal systems, vast amounts of power have been

Ibid. at 33.
Ibid. at 33, 35. See generally, David Estlund, ‘Introduction’ in David Estlund (ed.),
Democracy (Malden, Blackwell, 2002), 1–28. The National Commission on Terrorist
Attacks Upon the United States, as ‘an independent, bipartisan commission created by
congressional legislation and the signature of President . . . chartered to prepare a full and
complete account of the circumstances surrounding the 11 September 2001 terrorist
attacks, including preparedness for and the immediate response to the attacks [and]
mandated to provide recommendations designed to guard against future attacks’
(http://www.9-11commission.gov/) might be one example of such an arms-length delibe-
rative institution. See, National Commission on Terrorist Attacks Upon the United States,
The 9/11 Commission Report (New York, Norton, 2004).
I do not dispute that the right set of institutional mechanisms might yield legislative
results that better reflect informed, considered judgments rather than misperception and
fear; the point I defend later in this chapter, however, is that judicial review plays an
important institutional role in this regard.

conferred on the executive through delegated legislation. In the particular
context of anti-terrorism policy post-9/11, an increasing amount of power
has been concentrated in the executive,40 such as the power in Canada to
designate groups as ‘terrorist groups’ for the purpose of criminal proceed-
ings41 and to detain and remove non-citizens ‘on the basis of secret evidence
not disclosed to the deportee’.42 But perhaps the most controversial power is
that of preventive detention, when this power is available without a formal
trial or is not subject to full judicial review. Such is the case under internal
security legislation in Singapore and Malaysia,43 and initially under military
tribunals established in the United States post-9/11.44
This section considers three typical arguments in support of increased
executive power: the argument (implicit in Sunstein’s work) that the execu-
tive is in a better position to assess the risks associated with terrorism;
the argument, sometimes based on grounds of cultural particularism, that
a government that has earned the trust of the people ought to be affor-
ded more latitude; and the argument that extraordinary circumstances
call for extraordinary powers, including, in some instances, the power to
act extra-legally. It will be shown that these arguments are, at the very
least, overstated, and to the extent that they do justify greater executive
powers, such powers ought to be accompanied by corresponding checks
and balances.

A. Arguments from executive expertise
Sunstein’s work on risk perception and policy-making suggests a greater role
for and deference to experts in policy-making. It can be argued that when it
comes to threats to national security, the executive, with the advice of the



ńňđ. 4
(âńĺăî 23)