-
1
The Rise of Weak-Form Judicial Review
Mark Tushnet1
I Introduction
Constitutional review, known in the United States as judicial
review, originated
in the United States. It took the form of a practice in which
upon finding a statute
inconsistent with constitutional norms a court gave the statute
no legal effect in the case
at hand and, because of the operation of rules of stare decisis
in US law, other courts
subordinate to the deciding court would similarly give the
statute no legal effect in cases
coming before them.2 For more than a century this form of
review, to which I have given
the label strong-form, provided the only model for
constitutional review. As
parliaments increasingly rested on majoritarian theories of
democracy and as
constitutionalism spread the tensions between the two were
alleviated by accounts of
parliamentary supremacy that stressed legislative responsibility
for adherence to
constitutional norms. That responsibility could be enforced by
political norms
regulating legislators actions or more effectively by
legislators electoral responsiveness
to constituents who themselves cared about adherence to
constitutional norms.
Proponents of strong-form constitutional review sometimes
contend that that
practice is either compelled or strongly supported by the
existence of a written
constitution. Yet even written constitutions can accommodate
pure legislative
responsibility for adherence to constitutional norms, either in
the large or with respect to
specific provisions. The constitution of the Netherlands
specifically states that none of its
provisions are judicially enforceable, although the impact of
that limitation is qualified by
-
2
another provision requiring judicial enforcement of
international treaties to which the
Netherlands is a party. The Irish constitution of 1937 contains
a set of directive
principles of public policy, which are expressly committed
solely to the care of the
legislature and unenforceable by the courts.3 The Swedish
constitution, which provides
for what appears to be strong-form review, constrains the courts
through a limitation that
allows them to declare a statute unconstitutional only if the
inconsistency between the
statute and constitutional norms is (in one translation)
manifest. Even in the United
States, the home of strong-form review, the political question
doctrine is best
understood as identifying constitutional norms whose enforcement
is left in the hands of
elected representatives.4
Systems of complete parliamentary supremacy took strong,
sometimes
debilitating blows during the twentieth century, when experience
was thought to show
that purely political constraints on legislatures were
insufficient to avoid a descent into
totalitarianism or more modest but still troubling forms of
authoritarian rule. Sometimes
urged by U.S. advisers as in Germany and Japan and sometimes
emulating the German
model for designing a constitutional court, constitution
designers gravitated toward the
only available model of constitutional review, the U.S.-style
strong-form system, as a
remedy.5 Defenders of parliamentary supremacy and
constitutionalism argued that
strong-form review raised persistent questions about the
relation between on the one hand
the right of the sovereign people to govern themselves,
proximately through legislation
adopted by their representatives or by direct legislation the
people themselves enacted
and ultimately through their constitution, and on the other the
power of constitutional
courts to refuse to treat as legally effective statutes they
determined were inconsistent
-
3
with the constitutions higher norms. Constitution designers came
up with a number of
mechanisms to address these questions and at least weaken their
sting, including the
creation of specialized constitutional courts selected by
mechanisms different from those
used to select ordinary judges and restrictions on the
preclusive effect of determinations
of unconstitutionality.
In the late twentieth century constitution designers developed
another response,
which I have called weak-form constitutional review.6 Weak-form
review combines
some sort of power in courts to find legislation inconsistent
with constitutional norms
with some mechanism whereby the enacting legislature can respond
to a court decision to
that effect. Weak-form systems vary with respect to both the
nature of the judicial
power, which can be merely declaratory or provisionally
suspensive, and the form of
legislative response, which can be reenactment or slight
modification of the impugned
legislation. After describing several of the weak-forms design
features, this Chapter
examines the development of weak-form constitutional review in
Canada, New Zealand,
and Great Britain. It concludes with some speculations about the
future of weak-form
constitutional review.
II Constitutional review, reasonable disagreement and the
amendment process
Two features of modern constitutions lie at the foundation of
the development of
weak-form constitutional review. Every constitutions terms are
subject to reasonable
disagreement among interpreters and specifically between
legislators and constitutional
court judges. Weak-form review is a mechanism for dealing with
such disagreement in a
way that preserves parliamentary supremacy to as large extent as
possible. In addition
-
4
every constitution has some provision for its own amendment. The
ease or difficulty of
amending a constitution should affect the choice between
strong-form and weak-form
constitutional review.
Put most generally constitutionalism requires some limitations
on
parliamentary/legislative sovereignty. Those limitations can be
enforced culturally and
politically. A culture of legality among political elites might
induce legislators to
deliberate seriously about whether their preferred policy
proposals are consistent with
constitutional limitations on their power. A culture of legality
in the public might lead
voters to remove from office legislators who in the voters
judgment have failed to
comply with constitutional limitations.
The US model of constitutionalism seeks institutional mechanisms
for enforcing
constitutional limitations on parliamentary sovereignty, and the
judgment that such
mechanisms are a desirable feature of constitutional design has
now become nearly
universal. These mechanisms typically involve some determination
by an institution
independent of the legislature on the question of a proposals
consistency with
constitutional limitations. The institution might be a body that
screens proposals before
they are submitted for legislative consideration. Sweden and
Japan have such
institutions, and some observers contend that the near absence
in those nations of judicial
determinations that statutes are unconstitutional results from
the effectiveness of the
screening process in those institutions.7 A screening body might
have the power to
prevent legislative consideration. More typically these
rights-vetting institutions offer
their views as advice.8
-
5
Strong-form constitutional review places the power to determine
the consistency
of legislation with constitutional norms in a court authorized
to deny legal effect to
statutes it concludes are inconsistent with those norms. The
phrase it concludes is
crucial to understanding the persistent nervousness in
constitutional theory in the United
States and elsewhere about strong-form constitutional review. At
least occasionally and
to many observers rather often there can be reasonable
disagreement over the proposition
that a statute is inconsistent with constitutional norms.9 It is
hardly clear that
constitutionalism requires that the reasonable views of the
courts or other reviewing
institution prevail over the reasonable views of the enacting
legislature.
Mechanisms of weak-form constitutional review address the
nervousness about
giving reviewing courts the final and unrevisable word on such
questions. In their most
general form such systems continue to authorize independent
courts to analyze and
express their judgments about legislations consistency with
constitutional norms, but
allow legislatures to respond if legislators conclude that the
legislation is consistent with
constitutional norms reasonably understood. Of course even
strong-form systems of
constitutional review provide at least one mechanism for
responding to what seem to
some mistaken decisions to deny legal effect to a statute. That
mechanism is the process
for amending the nations constitution. Faced with a strong-form
courts decision that a
statute is inconsistent with constitutional norms, a nations
people can revise those
norms.10 The easier it is to amend the constitution, the smaller
the distinction between
strong-form and weak-form review. Put differently, strong-form
review raises questions
about the relation between courts and parliaments only in
systems where constitutional
amendment is significantly more difficult than is the enactment
of ordinary legislation.11
-
6
III Types of weak-form constitutional review
Weak-form review is a relatively recent development. It seems
likely that the
design possibilities extend beyond those already in force.
Enumerating some
mechanisms of weak-form review may nonetheless bring to the fore
several important
characteristics of these mechanisms.
(1) The weakest form of weak-form review may perhaps not deserve
the name of
constitutional review at all. It is the pure interpretive
mandate, which can best be
understood against the background of purely statutory bills of
rights. In a statutory bill of
rights a legislature enacts a list of rights as ordinary
legislation. That bill of rights is not
entrenched against repeal or subject to any special rules
regarding its later amendment.
How should courts treat the statutory bill of rights? The answer
is relatively
straightforward with respect to laws on the books at the time
the statutory bill of rights is
enacted. Courts should treat the statutory bill of rights as
amending all existing
legislation, which thereafter should be interpreted to be
consistent with the statutory bill
of rights. The situation with respect to later-enacted laws is
more complex because there
are two equally plausible approaches to interpreting them. A
court might assume that
subsequent legislatures wish to comply with the statutory bill
of rights and therefore
interpret later-enacted statutes to be consistent with it.
Alternatively a court might treat
later-enacted statutes as raising the possibility that they are
pro tanto subject-specific
amendments to the statutory bill of rights, which would lead to
interpreting them without
giving any special regard to their relation to the statutory
bill of rights.
-
7
The pure interpretive mandate is a legislative directive that
courts should where
fairly possible interpret legislation, both existing and
later-enacted, to be consistent with
some list of protected rights. The interpretive mandate purports
to rule out the choice of
the second approach to statutory bills of rights. The primary
question that arises in
connection with interpretive mandates is their relation to
ordinary principles of statutory
interpretation. Does the direction to interpret a statute to be
consistent with the protected
rights operate merely as a tie-breaker? If so a court would use
the ordinary rules of
statutory interpretation to determine the statutes meaning and
invoke the protected rights
only if the court was genuinely uncertain after doing so what
the statute meant.
Proponents of interpretive mandates typically argue that the
mandate must do more than
that, largely because situations of genuine uncertainty are
rare. They argue that courts
should use the list of protected rights as a new canon of
statutory interpretation. The
fact that a statute interpreted in the ordinary way has some
adverse effect on a protected
right is in their view a reason for interpreting the statute
otherwise. At the extreme this
approach to statutory interpretation can give the protected
rights overriding weight,
displacing even the plain meaning of a statutes words.12
Defenders of parliamentary sovereignty reject such an aggressive
use of an
interpretive mandate. For them the fundamental questions go not
to how courts should
respond to interpretive mandates but to their implications for
subsequent legislatures. An
interpretive mandate purports to place some burden of special
clarity on later legislatures
that seek to enact statutes that can be challenged as
rights-violating. But as a matter of
constitutional theory the power of one legislature so to bind a
later one by ordinary
legislation is quite unclear.
-
8
(2) Less weak is a form of review in which a court is empowered
to declare a
statute incompatible with fundamental rights, typically after
determining that it is unable
to interpret the statute so that it does not infringe such
rights. The circumstances under
which such a declaration will make a difference might be limited
to situations in which
the rights-violation was inadvertent. It might be that a
legislature simply did not
understand that its legislation would have an adverse impact on
rights, perhaps because
the impact was buried in the details of a complex statutory
scheme. Ordinarily courts
should be able to deal with such difficulties by applying the
interpretive mandate, but
occasionally they will be unable to escape a statutes clear
wording. The more likely
scenario is one in which the statutory language is both clear
and obvious. Here the
purported rights-violation may well have been precisely what the
legislature intended.
Again the existence of reasonable disagreement about whether
rights are violated matters.
To oversimplify somewhat either the legislature intended to
violate rights or it held a
view of rights under which the statute did not actually violate
rights. In either case the
legislature is unlikely to do anything in response to the
judicial declaration.
This conclusion should be qualified to take account of the
informational effect of
a judicial declaration of incompatibility. That effect may make
no difference where there
is strong party government because almost necessarily the party
leadership knew what it
was doing when it promoted the impugned legislation. Where party
discipline is weaker
some members of the majority party understood as a coalition may
have harbored
reservations about the legislation that they subordinated to
party loyalty but to which they
would give effect once informed that their reservations were
well founded. A public
informed by a declaration of incompatibility may respond at the
next election, and the
-
9
anticipation of such a response might induce the leadership even
in a strong party system
to alter its legislation.
(3) The strongest of the weak-form mechanisms gives the court
power to suspend
the legal effect of a statute pending a legislative response
through ordinary legislation
rather than constitutional amendment. Canadas so-called
notwithstanding clause,
Section 33 of the Charter of Rights, is the primary example of
such a mechanism.
(Another description of the provision is that it allows a
legislature to override
constitutional provisions.) The clause authorizes the national
legislature or a provincial
one with respect to legislation within their respective
jurisdictions to make a statute
legally effective notwithstanding enumerated constitutional
provisions dealing with
individual rights.13 Statutes protected by a notwithstanding
declaration expire after five
years, but the declaration is renewable. The five-year sunset
period ensures that an
election intervene between initial enactment and renewal,
thereby increasing the
likelihood that legislative responsibility will be enforced
through political accountability.
The presence of a notwithstanding or similar clause in a
constitution presents
some important interpretive questions. When a legislature
invokes such a clause must it
identify precisely which constitutional provisions with which
the statute might be
inconsistent, or may it invoke the clause in general terms, such
as notwithstanding any
provision of this constitution? May a legislature invoke such a
clause prior to any
judicial suspension of an enactments legal effect, or may it do
so only after a court has
found a statute inconsistent with constitutional norms? These
questions are not entirely
unrelated. After a court finds a statute inconsistent with a
specific constitutional norm,
-
10
the legislature can make its statute legally effective by
referring to that constitutional
norm in its override.14
These questions help bring out some general features of
weak-form constitutional
review.15 Proponents of weak-form review note that legislators
may simply overlook
constitutional questions implicated in their policy proposals. A
statute might not in fact
embody a reasoned legislative determination that the statute is
consistent with
constitutional norms. Weak-form review offers the legislature an
opportunity to focus on
that question without necessarily precluding the legislature
from reenacting the impugned
statute. In addition and perhaps related legislators are not
specialists in constitutional
law. Weak-form review in contrast is performed by such
specialists. Weak-form review
may provide legislators with information that they lacked at the
enactment stage.
Specialization can of course distort as well as inform judgment.
Distracted by their
attention to nuances of constitutional law constitutional court
judges might undervalue
the contributions a statute makes to overall social wellbeing.
Again weak-form review
has the advantage of allowing legislatures to take these and
similar distortions into
account as they respond to the courts decision. These processes
have been described
through the metaphor that weak-form review creates a dialogue
between courts and
legislatures.16
Requiring an override to be cast in specific terms substantially
reduces the
possibility that the legislatures action in adversely affecting
constitutional rights, at least
in the eyes of some, is inadvertent, and may increase the
possibility that voters will be
sufficiently informed as to hold legislators responsible for
their decision. Anticipatory
and specific overrides, that is, those made before a court finds
a statute inconsistent with
-
11
constitutional norms identified in the overriding legislation
itself, run the risk that the
legislature will identify only one constitutional problem with
its enactment while others
are severe enough to lead the courts to invalidate the
legislation anyway. Anticipatory
overrides also cannot capitalize on the information that would
be provided by a judicial
invalidation, although the relevant information may be widely
enough available, for
example by inferences from prior court decisions, to reduce the
severity of this concern.
In addition anticipatory overrides cannot contribute to a
dialogue between courts and
legislatures on question of constitutional interpretation.
IV Weak-form review and the idea of judicial-legislative
dialogue
Weak-form review can be quite weak, as the examples of the pure
interpretive
mandate and the declaratory form of review show. What benefits
might there be from
such an institution in a world where courts and legislatures can
reasonably disagree on
the question of a statutes consistency with constitutional
norms? As noted above, one
benefit is dialogue between courts and legislatures. The tension
between strong-form
constitutional review and parliamentary sovereignty is strongest
when fully informed
legislatures simply but reasonably disagree with the
interpretation a court gives to a
constitutional limitation on legislative power. Dialogue allows
courts to inform a
legislature of the courts understanding of the constitutional
provision while allowing the
legislature to respond and take conclusive action based on its
own understanding.
Two episodes in Canadian law illustrate different forms of
dialogue. (1) In R. v.
OConnor ([1995] 4 SCR 411), a sharply divided Supreme Court
developed the common
law in light of the Charter to allow relatively generous
opportunities for defense counsel
-
12
to gain access to the medical and counseling records of a
complainant in a prosecution for
sexual assault. Parliament responded with a statute that
essentially adopted the
dissenters views on when such access should be granted. The
Supreme Court then
upheld the statute against constitutional challenge, observing
that it did not hold a
monopoly on Charter interpretation (R. v. Mills, [1999] 3 SCR
668). (2) In 1989
Parliament adopted a comprehensive and rather strict system
regulating advertising of
tobacco products. The Supreme Court found that the system
violated constitutional
guarantees of free expression because each component of the
system failed under one or
more of the tests for proportionality (RJR-MacDonald Inc v.
Canada (Attorney General),
[1995] 3 SCR 199). Its proportionality analysis delineated the
outlines of constitutionally
permissible legislation, and Parliament responded by enacting a
statute that conformed to
those outlines.
Responses can take several forms depending on the precise
contours of the weak-
form mechanism. Consider first a weak-form mechanism authorizing
courts to do no
more than declare a statute incompatible with constitutional
norms, the declaration
having no immediate legal consequences. A legislature need do
nothing to make its
statute legally effective after such a declaration. It may of
course face political pressure
or social expectations that it do something in response, but the
simple declaration of
incompatibility puts the legislature in no different legal
position than it was in before the
declaration. Next consider a weak-form mechanism that suspends a
statutes legal
effectiveness subject to later legislative action. Importantly
one possible response is
doing nothing because the legislature informed by the courts
decision concludes that
constitutional norms do indeed preclude it from accomplishing
its policy goal. A
-
13
legislature that on reflection wishes to make its statute
legally effective does have to
respond. Depending on the design of the weak-form mechanism, it
can do so by
reenacting the statute either by the ordinary process of
legislation or under some
modestly enhanced procedural requirements, such as a qualified
majority rule requiring a
majority of the legislature as a whole rather than simply a
majority of legislators present
and voting, or a modest supermajority requirement.17
The responses described so far reflect a legislatures judgment
that its own
reasonable view that the statute is consistent with
constitutional norms should prevail
over the reviewing courts view. The Canadian experience
illustrated by the dialogue
on tobacco advertising suggests that the most common legislative
response will be
tinkering with the statute to ensure that it conforms to the
reviewing courts view. This
response has two notable characteristics. It is one in which the
legislature accepts rather
than rejects the reviewing courts constitutional interpretation,
perhaps because the
legislature initially overlooked constitutional concerns or
perhaps because the
legislatures reflective reconsideration led to a change in a
majoritys views about
constitutionality. In addition the tinkering will almost
certainly come at some cost to the
legislatures initial non-constitutional policy goals. Modest
adjustments in procedures
will increase implementation costs, which in turn will reduce
the number of occasions on
which the statute will be invoked. Tinkering as a response may
reflect a legislative
judgment that the reduction in the statutes effectiveness is not
large enough to justify
incurring the political costs of directly disagreeing with the
courts.
V The development of weak-form constitutional review
-
14
By the late twentieth century parliamentary supremacy was the
reigning
constitutional theory in only a handful of nations, mostly in
the British Commonwealth.
Their legal elites came to believe that some mechanism for
overseeing legislative
compliance with constitutional norms was desirable.18 Canada
adopted the strongest of
the weak-form systems, followed by New Zealands choice of the
weakest, and then the
United Kingdoms choice of the intermediate variant.19
(1) After beginning his career as a constitutional lawyer,
Canadas minister of
justice and prime minister Pierre Trudeau came to the view that
Canadas persistent
division between Quebec and the rest of Canada over national
identity could be solved by
patriating the Canadian constitution. Doing so, Trudeau
believed, would provide a basis
for reconfiguring national identity around the constitution as a
charter of rights and
around multiculturalism rather than bilingualism. A 1981
advisory opinion from the
Canadian Supreme Court held that patriation required extensive
consultation with the
provinces and substantial agreement among them over the
constitutions terms. Meetings
with the provinces first ministers produced agreement on many
terms, but division
persisted over the creation of an enforceable charter of rights,
with leaders of social
democratic parties skeptical about US-style constitutional
review because of experience
in the 1930s in the United States and Canada when courts
obstructed economic recovery
programs. Trudeau was committed to such a charter as an
essential component of
constitutional reform. The ministers comprised by inserting two
provisions in the
proposed constitution, and the Canadian constitution was
patriated in 1982 with an
accompanying Charter of Rights enforceable in the courts.
-
15
The first provision affecting constitutional review was the
notwithstanding clause,
a real innovation in constitutional design.20 It authorized
legislative responses to judicial
invalidation in the form of straightforward reenactment of the
impugned legislation.
Quebecs political leadership had not accepted patriation, and
immediately after the
constitution became effective invoked the notwithstanding clause
to protect all of the
provinces statutes against constitutional challenge. The most
controversial of those
statutes was a bill requiring that all commercial signage in
Quebec be in French. Several
businesses challenged the sign law as a violation of their
Charter rights and the use of the
notwithstanding clause as improper. The Canadian Supreme Court
held that it could
examine only whether Quebecs use of the notwithstanding clause
was formally valid
(Ford v. Quebec (Attorney General), [1988] 2 SCR 712).21 Despite
or perhaps because
of this holding the notwithstanding clause has not been used
since 1988 to insulate any
significant legislation against constitutional challenge.
The second provision was a general limitations clause applicable
to the rights
enumerated in the charter. Modeled on language in existing
international human rights
documents, where the terms were used to identify permissible
limitations on particular
rights, the clause provides that all the rights are subject only
to such reasonable limits
prescribed by law as can be demonstrably justified in a free and
democratic society. The
general limitations clause invited a different type of
legislative response. As Canadian
constitutional doctrine has evolved courts make two
determinations. They ask first
whether the challenged legislation infringes on a constitutional
right. The standard for
determining whether a right is infringed is in general easy to
satisfy. Once it is the
second, more consequential determination must be made. The court
asks whether the
-
16
infringement is demonstrably justified. Constitutional doctrine
has converted this into a
multi-stage proportionality test.
The language of the general limitations clause suggests one form
that challenges
to legislation can take. Litigants might argue that the statute
limits a constitutional right,
and that the limitation is not demonstrably justified. If the
courts agree the legislature can
respond by providing a stronger justification than it had
earlier, for example by compiling
a more complete legislative record of the statutes effectiveness
at addressing a social evil
and the relative ineffectiveness of alternatives. The
proportionality doctrine has
components that invite cognate responses. To satisfy the
proportionality requirement
legislation must impair as little as possible the right, and
there must be a
proportionality between the effects of the measures and the
objective of the statute (R
v Oakes, [1986] 1 SCR 103). Legislation that fails the
proportionality test in either of
these ways might be reenacted with a demonstration that no
alternatives achieve the
legislations goals with a smaller effect on rights. More
commonly the legislature
responds with slight alterations in the legislation to reduce
the adverse effect on rights
while preserving the statutes main features.
(2) In New Zealand as in Canada a law professor politician was
the moving force
behind the adoption of a bill of rights and as in Canada
questions about national identity
played a role. Having received a law degree and taught at law
schools in the United
States, Geoffrey Palmer had a principled commitment to a fully
enforceable bill of rights,
and worked for its adoption as minister of justice and then
prime minister. He saw such a
bill of rights as part of a package of constitutional reforms,
which included proportional
representation and the recognition of constitutional status for
the Treaty of Waitangi
-
17
(1840) between New Zealands indigenous Maori people and its
British settlers.
Palmers advocacy of an entrenched bill of rights failed in the
face of strong assertions of
a national commitment to parliamentary supremacy. Shortly before
the Labour Party led
by Palmer lost its majority in 1990 the New Zealand parliament
adopted a statutory Bill
of Rights listing protected rights and containing an
interpretive mandate.
New Zealands courts have been rather cautious but not completely
inactive in
their use of the interpretive mandate. Some judges have invoked
theoretical concerns
about the ability of an earlier legislature to bind a later one
to justify their caution. Yet in
one notable case the nations high court held that the Bill of
Rights Act authorized the
courts to remedy some violations with a damages remedy despite
the fact that they have
no authority even to declare legislation incompatible with the
Act (Baigents Case,
[1994] 3 NZLR 667). More recently the court overturned the
conviction of a
schoolteacher and political activist for dishonouring the
nations flag by burning it in a
political protest, invoking the interpretive mandate as
requiring that the word dishonour
be given as narrow an interpretation as possible to ensure that
the statute not infringe free
speech rights (Hopkinson v. Police, [2004] 3 NZLR 704).
(3) The British Human Rights Act 1998 was the culmination of a
campaign by
the Labour Party to bring rights home, as one of the Partys
documents put it.
Historically the Labour Party had been quite hostile to the
judiciary. The Taff Vale
decision (1901), which imposed monetary liability on a labor
union for actions taken in a
strike that was held to violate common law rules against
interference with contractual
relations, held an iconic and notorious place in Labourites
understanding of courts and
the law. Labour leaders believed that judges as members of the
ruling class and typically
-
18
of upper class origins themselves were inevitably biased against
the working class, which
cautioned strongly against developing policies that required
judicial support to be
effective. The US experience in the first decades of the
twentieth century with the
enforcement of notions of due process to limit social welfare
legislation was taken as
exemplary. Further, the Labour Party was committed to
parliamentary sovereignty with
widespread public participation in elections and a strong party
system, which its leaders
believed would eventually guarantee their programs long-term
success.
The Labour Partys position changed after Margaret Thatcher
became prime
minister in 1979. Leading a strong and cohesive Conservative
Party government,
Thatcher was able, as Labour leaders saw it, to force through
legislation inconsistent with
long-standing British traditions, which they believed would have
been blocked by courts
empowered to enforce human rights. Other developments
contributed to changing
attitudes within the Labour Party about judicial protection of
rights. The democratization
of the legal profession promised to transform the judiciary, at
least in the long run.
Constitutional changes occurred in other components of the
British system of
parliamentary supremacy, weakening the case against
constitutional review. Practices of
ministerial and cabinet responsibility changed dramatically, for
example.22 The United
Kingdoms accession to the European Community and to the human
rights regime of the
European Convention on Human Rights had theoretical and
practical effects as well.
Constitutional theorists struggled to reconcile the doctrine of
parliamentary supremacy
with the doctrine that European Community law had direct effect
within the United
Kingdom. No solution was entirely satisfactory, but one
promising route was a
modification of the doctrine that no earlier parliament could
bind a later one. That
-
19
modification would provide support for the adoption of an
entrenched bill of rights.
Perhaps more important, British politicians were embarrassed by
the rate at which the
European Court of Human Rights found that the United Kingdom had
violated rights
protected by the Convention.
Labour Party leader John Smith put together a package of
constitutional reforms
to incorporate in the Partys platform, including devolution of
power to Scotland and
Wales, and his successor Tony Blair implemented many of the
reforms. One was the
Human Rights Act 1998. The HRA made most of the rights protected
by the European
Convention enforceable as domestic law through an interpretive
mandate coupled with a
power in higher courts to declare primary legislation
incompatible with the enumerated
rights.23 Such a declaration would have no immediate legal
effect. The Labour
government stated its expectation that declarations of
incompatibility would promptly
produce the introduction of legislation to eliminate the
incompatibility. In addition the
HRA authorized fast-track procedures for such remedial
legislation, and even authorized
the minister with jurisdiction over the matter to modify the
impugned statute by executive
order pending prompt submission to parliament of remedial
legislation.
The House of Lords has been reasonably aggressive in its use of
the interpretive
mandate, with some decisions appearing to invoke human rights
concerns to override the
plain meaning of statutory language. An anti-terrorism case
known as the Belmarsh
decision is the most celebrated and controversial use of the
declaration of incompatibility
(A v Secretary of State, [2004] UKHL 56). Several persons
suspected of terrorism
offenses were detained at Belmarsh Prison without charge, for an
indefinite period.
Justifiably concerned that indefinite detention might be held to
violate the detainees
-
20
rights under the European Convention, the government filed a
derogation from the
Convention with respect to the detentions. The House of Lords
held that the derogation
was legally ineffective and declared that the detentions were
incompatible with the HRA
because the standards for detention improperly discriminated
between citizens and non-
citizens equally suspected of terrorism-related offenses, and
because indefinite detention
was a disproportionate response to the threat the detainees
posed.24 Subsequent decisions
have upheld many provisions of terrorism-related legislation, to
the point where some of
those who strongly supported the adoption of the HRA have
expressed disappointment
with its impact.
VI Speculations on the future of weak-form constitutional
review
Strong-form review has existed for two hundred years in the
United States and for
many years in other jurisdictions, providing a rich empirical
basis for evaluating it. The
much shorter experience with weak-form review means that
evaluating it empirically
requires a great deal of speculation. The primary empirical
question about weak-form
review is whether the mechanism in any of its forms is likely to
be stable over time.
Instability might occur in both directions. Review might be so
weak that the system
degenerates into parliamentary supremacy. Alternatively, review
might degenerate
into strong-form review if legislatures never take a stand
against judicial invalidations.
Whether the latter should be regarded as degeneration is
unclear. Suppose courts
regularly declare statutes unconstitutional and legislatures
regularly fail to respond by
overriding the courts decisions, or respond by enacting new
statutes that conform to the
courts requirements. That course of conduct might reflect a
considered judgment by the
-
21
legislature that strong-form review is in practice better than
weak-form review, and that it
can embed a strong-form system simply by regularly complying
with judicial
declarations. Another way of putting what amounts to the same
point is that legislatures
might regularly find that the political costs of overriding
judicial decisions exceed the
policy and political benefits of doing so.
What limited experience there is suggests that sustaining a
weak-form mechanism
is difficult. The notwithstanding clause in Canada has fallen
into desuetude, largely
discredited by its use to insulate French-language favoring
legislation in Quebec from
review under the national Charter of Rights and by occasional
threats, never effectively
carried out, to use the clause to protect legislation
discriminating against gays and
lesbians. The clause remains an important symbol of
parliamentary supremacy as the
governing theory of Canadian constitutionalism: A party leaders
suggestion that the
clause be eliminated to perfect Canadian constitutionalism met a
strong and highly
critical response. The general-limitations clause in Canada has
become the textual basis
for testing all legislation against a requirement of
proportionality, inviting tinkering
responses that accept the courts supremacy in constitutional
interpretation. Evaluations
of the vitality of the judicial-legislative dialogue reach mixed
conclusions, depending on
whether modifying a statute to conform to judicial requirements
is treated as an example
of productive dialogue or as an example of judicial
supremacy.
In Great Britain the Labour government routinely modified
legislation in response
to declarations of incompatibility. Most such declarations
involved provisions that were
probably inadvertent rights violations even from the governments
point of view, but the
most prominent decisions involved aspects of anti-terrorism
legislation and crime control
-
22
laws. The Labour government grudgingly accepted those decisions.
It responded by
developing a discussion paper on rights and responsibilities,
which asserted that the
government was proud that it introduced the Human Rights Act and
it will not resile
from it nor repeal it.25 Yet uneasiness with expansive
interpretations of the protected
rights continues in Great Britain. Conservative Party leaders
have been uncomfortable
with those interpretations, and at this writing it is unclear
whether the Human Rights Act
will retain its current form. Interpretation of statutes to
ensure conformity with the
Human Rights Act placed some strain on traditional theories of
statutory interpretation,
although not to the breaking point.
In New Zealand too occasional expansive uses of the interpretive
mandate have
generated concern that the mandate will create a system of
strong-form review. More
commonly observers assert that the government pays too little
attention to human rights
issues when it introduces legislation, citing the rather large
number of instances in which
the Attorney General, an official with a fair amount of
independence from the
government, has reported that government legislation is
inconsistent with the bill of
rights.
Weak-form constitutional review is one of only a handful of
inventions in
designing constitutional courts to occur over the past two
centuries. Whatever its
ultimate fate, its mechanism for reducing the tension between
parliamentary supremacy
and constitutional limitations on that supremacy provides the
ground for serious
reflection on fundamental features of constitutional design in
modern democracies.
References
-
23
Dixon, Rosalind. A Democratic Theory of Constitutional
Comparison, 46 American
Journal of Comparative Law 947-997 (2008).
Gardbaum, Stephen. The New Commonwealth Model of
Constitutionalism, 49
American Journal of Comparative Law 707-60 (2001).
Gzler, Kemal. Judicial Review of Constitutional Amendments: A
Comparative Study
(Bursa: Ekin Press, 2008).
Hogg, Peter & Allison Bushell, The Charter Dialogue Between
Courts and Legislatures,
35 Osgoode Hall Law Journal 75-124 (1995).
Jowell, Jeffrey & Dawn Oliver, eds. The Changing
Constitution (Oxford: Oxford
University Press, 1985).
Law, David S. The Anatomy of a Conservative Court: Judicial
Review in Japan, 87
Texas Law Review 1545-93 (2009).
Oliver, Dawn. Constitutional Reform in the United Kingdom
(Oxford: Oxford
University Press, 2003).
Tushnet, Mark. Weak Courts, Strong Rights: Judicial Review and
Social Welfare Rights
in Comparative Constitutional Law (Princeton: Princeton
University Press, 2007).
Waldron, Jeremy. The Core of the Case Against Judicial Review,
115 Yale Law
Journal 1346-406 (2006).
1 William Nelson Cromwell Professor of Law, Harvard Law
School.
2 This formulation captures the strong-form character of
constitutional review in the
United States, the focus of this Chapter. For a discussion of
other features in the design
of constitutional review, see Victor Cornella, The Rise of
Specialized Constitutional
Courts, [this collection].
-
24
3 The Indian constitution of 1950 contains a similar section,
modeled on the Irish one.
The Indian Supreme Court has referred to the directive
principles in its interpretations of
constitutional provisions that are judicially enforceable. The
Irish Supreme Court has
gestured in that direction as well, but has resisted the
temptation more successfully.
4 In Luther v. Borden, 48 US (7 How) 1 (1848), the Supreme Court
held that whether a
states system of government satisfied the constitutional
requirement that the United
States guarantee that each state have a republican form of
government was a political
question, meaning that it was to be determined by the
legislative and executive branches.
5 This Chapter deals only with weak- and strong-form
constitutional review, and not
questions about whether review is dispersed among a number of
courts or centralized in a
single one (either a specialized constitutional court or a
generalist court). Nor does it
consider the suggestion made to me by Vicki Jackson that
constitutional courts whose
members serve for short terms might come to resemble weak-form
courts, with quick
replacement of judges who render decisions politicians disagree
with.
6 The first scholar to identify weak-form constitutional review
as a distinctive and general
feature of constitutional design was Gardbaum (2001). For my
contribution, see Tushnet
(2007).
7 On Japan, see Law (2009). Law addresses but rejects the
argument that Japans
screening mechanism accounts for the low level of judicial
invalidations.
8 For a discussion of these institutions, see Janet Hiebert,
Legislative Alternatives to
Judicial Review, [this collection].
-
25
9 Jeremy Waldron has been the legal theorist most insistent on
this claim and its
implications for constitutional review. For his most careful
formulation, see Waldron
(2006).
10 Subject to whatever limitations the system places on the
ability of a people to amend
their constitution. For a survey of the issue of unamendable
constitutional provisions, see
Gzler (2008).
11 I think it important to note here that those who think about
constitution design should
not assume that the enactment of ordinary legislation is a
simple majoritarian process. It
might be, as it tends to be in pure parliamentary systems with
strong political parties, but
it need not be. Enactment is not a simple majoritarian process
in separation of powers
systems or even in parliamentary systems with weak party
discipline.
12 The issue has been mooted in a number of British cases, with
some judges defending
this approach in principle though never actually applying
it.
13 The clause does not apply to voting and language rights.
14 Subject to the possibility that the court ruled on only one
constitutional challenge in a
situation where others might remain available.
15 For a discussion bearing on these responses, see Dixon
(2008).
16 The metaphor was introduced in Hogg & Bushell (1995), and
has spread widely.
17 The only limitation on these procedural enhancements is that
they cannot be as
stringent as those required for constitutional amendment.
18 The major hold-out among nations committed to parliamentary
supremacy was France,
where the Constitutional Council created in 1957 was initially
understood as a
mechanism to adjudicate disagreements over the allocation of
law-making authority
-
26
between the president and the parliament and only gradually
became a mechanism for
general constitutional review. The last step appears to have
been taken with a
constitutional amendment adopted in 2008 that authorizes the
Constitutional Council to
determine the constitutionality of statutes after enactment,
although that recent
development has not taken effect as of this writing and its
impact therefore cannot be
determined.
19 Designers of constitutional courts in central and eastern
Europe in the 1990s chose
strong-form review, almost certainly under the influence of the
German model and so
probably independent of their views on the relation between
parliamentary supremacy
and constitutionalism.
20 There is some dispute over who first devised this
solution.
21 For complex reasons the court was able to find that the sign
law violated Quebecs
provincial bill of rights, and the broader sign law was replaced
by one requiring that
signage be predominantly but not exclusive in French.
22 For a discussion of these and other constitutional changes,
see Oliver (2003). For a
survey of changes in their incipiency, see Jowell & Oliver
(1985).
23 The Convention right to effective remedies for rights
violations was excepted from the
list, out of concern that incorporating such a right would be
inconsistent with the limited
declaratory remedy created by the HRA.
24 The government responded by creating a system of fairly
stringent house arrest rather
than detention in a prison-like facility.
25 Ministry of Justice, Rights and Responsibilities: developing
our constitutional
framework (March 2009).
-
27