43A1A8E4-5F61-084271.

DOC 12/15/2005 9:33 AM

Jeremy Wa ldron

The Core of the Case Against Judicial Review

abst ract. This Essay states the general case against judicial review of
legislation clearly and in a way that is uncluttered by discussions of particular
decisions or the history of its emergence in particular systems of constitutional
law. The Essay criticizes judicial review on two main grounds. First, it argues that
there is no reason to suppose that rights are better protected by this practice than
they would be by democratic legislatures. Second, it argues that, quite apart from
the outcomes it generates, judicial review is democratically illegitimate. The
second argument is familiar; the first argument less so.
However, the case against judicial review is not absolute or unconditional. In
this Essay, it is premised on a number of conditions, including that the society in
question has good working democratic institutions and that most of its citizens
take rights seriously (even if they may disagree about what rights they have). The
Essay ends by considering what follows from the failure of these conditions.
[ME: format abstract]

aut hor. University Professor in the School of Law, Columbia University.

Earlier versions of this Essay were presented at the Colloquium in Legal and
Social Philosophy at University College London, at a Law Faculty workshop at the
Hebrew University of Jerusalem, and at a Constitutional Law Conference at
Harvard Law School. I am particularly grateful to Ronald Dworkin, Ruth Gavison,
and Seana Shiffrin for their formal comments on those occasions and also to
Aharon Barak, Richard Bellamy, Aileen Cavanagh, Arthur Chaskalson, Michael
Dorf, Richard Fallon, Charles Fried, Stephen Guest, Ian Haney-Lopez, Alon
Harel, David Heyd, Sam Issacharoff, Elena Kagan, Kenneth Keith, Michael
Klarman, John Manning, Andrei Marmor, Frank Michelman, Henry Monaghan,
Véronique Munoz-Dardé, John Morley, Matthew Palmer, Richard Pildes, Joseph
Raz, Carol Sanger, David Wiggins and Jo Wolff for their suggestions and
criticisms. Hundreds of others have argued with me about this issue over the
years: This Essay is dedicated to all of them, collegially and with thanks.

100
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

ess ay co ntent s

introduction 102

i. definition of judicial review 107

ii. four assumptions 112

A. Democratic Institutions 114
B. Judicial Institutions 116
C. A Commitment to Rights 117
D. Disagreement About Rights 119

iii. the form of the argument 122

iv. outcome-related reasons 128

A. Orientation to Particular Cases 132
B. Orientation to the Text of a Bill of Rights 133
C. Stating Reasons 135

v. process-related reasons 138

vi. the tyranny of the majority 147

vii. non-core cases 153

conclusion 157

101
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

introduct io n

Should judges have the authority to strike down legislation when they are
convinced that it violates individual rights? In many countries they do. The best
known example is the United States. In November 2003, the Supreme Judicial
Court of Massachusetts ruled that the state’s marriage licensing laws violated
state constitutional rights to due process and equal protection by implicitly
limiting marriage to a union between a man and a woman.1 The decision
heartened many people who felt that their rights were unrecognized and that, as
gay men and women, they were treated as second-class citizens under the existing
marriage law.2 Even if the decision is eventually overturned, the plaintiffs and
their supporters can feel that at least the issue of rights is now being confronted
directly. A good decision and a process in which claims of rights are steadily and
seriously considered3—for many people these are reasons for cherishing the
institution of judicial review. They acknowledge that judicial review sometimes
leads to bad decisions—such as the striking down of 170 labor statutes by state
and federal courts in the Lochner era4—and they acknowledge that the practice
suffers from some sort of democratic deficit. But, they say, these costs are often
exaggerated or mischaracterized. The democratic process is hardly perfect and, in
any case, the democratic objection is itself problematic when what is at stake is
the tyranny of the majority. We can, they argue, put up with an occasional bad
outcome as the price of a practice that has given us decisions like Brown, Roe,
and Lawrence,5 which upheld our society’s commitment to individual rights in the
face of prejudiced majorities.
That is almost the last good thing I shall say about judicial review. (I wanted
to acknowledge up front the value of many of the decisions it has given us and the
complexity of the procedural issues.) This Essay will argue that judicial review of
legislation is inappropriate as a mode of final decisionmaking in a free and
democratic society.

1. Goodridge v. Dep’t of Pub. Health, 798 N.E.2d. 941 (Mass. 2003).

2. See [SC2: author for this article?] Landmark Ruling: The Victors, BOSTON HERALD, Nov. 19,
2003, at 5.
3. This adapts a phrase of Ronald Dworkin’s, from RONALD DWORKIN, Political Judges and the
Rule of Law, in A MATTER OF PRINCIPLE 9, 32 (1985) [SC2: is this a series of essays, or just a
book with chapters?].
4. Lochner v. New York, 198 U.S. 45 (1905). The calculation of the overall number of cases in
which state or federal statutes on labor relations and labor conditions were struck down in the
period 1880-1935 is based on lists given in WILLIAM E. FORBATH, LAW AND THE SHAPING OF THE
AMERICAN LABOR MOVEMENT apps. at 177-92, 199-203 (1991).
5. Lawrence v. Texas, 539 U.S. 558 (2003); Roe v. Wade, 410 U.S. 113 (1973); Brown v. Bd. of
Educ., 347 U.S. 483 (1954).

102
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

Arguments to this effect have been heard before, and often. They arise
naturally in regard to a practice of this kind. In liberal political theory, legislative
supremacy is often associated with popular self-government,6 and democratic
ideals are bound to stand in an uneasy relation to any practice that says elected
legislatures are to operate only on the sufferance of unelected judges. Alexander
Bickel summed up the issue in the well-known phrase, “the counter-majoritarian
difficulty.”7 We can try to mitigate this difficulty, Bickel said, by showing that
existing legislative procedures do not perfectly represent the popular or the
majority will. But, he continued,
nothing in the further complexities and perplexities of the system, which
modern political science has explored with admirable and ingenious
industry, and some of which it has tended to multiply with a fertility that
passes the mere zeal of the discoverer—nothing in these complexities can
alter the essential reality that judicial review is a deviant institution in the
American democracy.8

In countries that do not allow legislation to be invalidated in this way, the

people themselves can decide finally, by ordinary legislative procedures, whether
they want to permit abortion, affirmative action, school vouchers, or gay marriage.
They can decide among themselves whether to have laws punishing the public
expression of racial hatred or restricting candidates’ spending in elections. If they
disagree about any of these matters, they can elect representatives to deliberate
and settle the issue by voting in the legislature. That is what happened, for
example, in Britain in the 1960s, when Parliament debated the liberalization of
abortion law, the legalization of homosexual conduct among consenting adults,
and the abolition of capital punishment.9 On each issue, wide-ranging public
deliberation was mirrored in serious debate in the House of Commons. The
quality of those debates (and similar debates in Canada, Australia, New Zealand,
and elsewhere) make nonsense of the claim that legislators are incapable of
addressing such issues responsibly—just as the liberal outcomes of those

6. The locus classicus for this concept is JOHN LOCKE, TWO TREATISES OF GOVERNMENT 366-67
(Peter Laslett ed., Cambridge Univ. Press 1988) (1690). [SC2: this cite must be to one of the
two treatises. Please cite as shorter work in a collection by a single author.]
7. ALEXANDER BICKEL, THE LEAST DANGEROUS BRANCH 16-17 (1986) [SC2: 1986 is not the date
of the first edition. If this is a revised ed, cite accordingly. LE: check precedent on this one]
(“[J]udicial review is a counter-majoritarian force in our system. . . . [W]hen the Supreme Court
declares unconstitutional a legislative act . . . it thwarts the will of representatives of the actual
people of the here and now. . . .”).
8. Id. at 17-18.
9. Medical Termination of Pregnancy Act of 1966, [sc2: missing at sc1]; Sexual Offences Act,
1967, c. 60; Murder Act, 1965, c. 71.

103
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

proceedings cast doubt on the familiar proposition that popular majorities will not
uphold the rights of minorities.
By contrast, in the United States the people or their representatives in state
and federal legislatures can address these questions if they like, but they have no
certainty that their decisions will prevail. If someone who disagrees with the
legislative resolution decides to bring the matter before a court, the view that
finally prevails will be that of the judges. As Ronald Dworkin puts it—and he is a
defender of judicial review—on “intractable, controversial, and profound
questions of political morality that philosophers, statesmen, and citizens have
debated for many centuries,” the people and their representatives simply have to
“accept the deliverances of a majority of the justices, whose insight into these
great issues is not spectacularly special.”10
In recent years, a number of books have appeared attacking judicial review in
America.11 For years, support for the practice has come from liberals, and
opposition from conservative opponents of the rights that liberal courts have
upheld. In recent years, however, we have seen the growth of liberal opposition to
judicial review has grown, as the Rehnquist Court struck down some significant
achievements of liberal legislative policy.12 But there have been spirited defenses
of the practice as well.13 The two-hundredth anniversary of Marbury v. Madison
elicited numerous discussions of its origins and original legitimacy, and the
fiftieth anniversary of Brown v. Board of Education provided a timely reminder of
the service that the nation’s courts performed in the mid-twentieth century by
spearheading the attack on segregation and other racist laws.
So the battle lines are drawn, the maneuvering is familiar, and the positions
on both sides are well understood. What is the point of this present intervention? I
have written plenty about this myself already.14 Why another article attacking
judicial review?

10. RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE A MERICAN CONSTITUTION 74
(1996).
11. See LARRY D. KRAMER, THE PEOPLE THEMSELVES: POPULAR CONSTITUTIONALISM AND JUDICIAL
REVIEW (2004); MARK TUSHNET, TAKING THE CONSTITUTION AWAY FROM THE COURTS (1999).
12. See, e.g., United States v. Morrison, 529 U.S. 598 (2000) (striking down part of the Violence
Against Women Act); United States v. Lopez, 514 U.S. 549 (1995) (holding that Congress has
no authority to legislate a prohibition on the possession of guns within a certain distance from a
school); see also Mark Tushnet, Alarmism Versus Moderation in Responding to the Rehnquist
Court, 78 IND. L.J. 47 (2003).
13. See, e.g., DWORKIN, supra note 10; CHRISTOPHER L. EISGRUBER, CONSTITUTIONAL SELF-
GOVERNMENT (2000); LAWRENCE G. SAGER, JUSTICE IN PLAINCLOTHES: A THEORY OF AMERICAN
CONSTITUTIONAL PRACTICE (2004).
14. See, e.g., JEREMY WALDRON, LAW AND DISAGREEMENT 10-17, 211-312 (1999); Jeremy Waldron,
Deliberation, Disagreement, and Voting, in DELIBERATIVE DEMOCRACY AND HUMAN RIGHTS 210
(Harold [SC2: middle name given?] Koh and Ron Slye eds., 1999) [hereinafter Waldron,

104
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

What I want to do is identify a core argument against judicial review that is

independent of both its historical manifestations and questions about its
particular effects—the decisions (good and bad) that it has yielded, the
heartbreaks and affirmations it has handed down. I want to focus on aspects of the
case against judicial review that stand apart from arguments about the way judges
exercise their powers and the spirit (deferential or activist) in which they
approach the legislation brought before them for their approval. Recent books by
Mark Tushnet and Larry Kramer entangle a theoretical critique of the practice
with discussions of its historical origins and their vision of what a less judicialized
American Constitution would involve.15 This is not a criticism of Tushnet and
Kramer. Their books are valuable in large part because of the richness and color
they bring to the theoretical controversy. As Frank Michelman says in his blurb
on the back cover of The People Themselves, “Kramer’s history . . . puts flesh on
the bones of debates over judicial review and popular constitutionalism.”16 And
so it does. But I want to take off some of the flesh and boil down the normative
argument to its bare bones so that we can look directly at judicial review and see
what it is premised on.
Charles Black once remarked that in practice opposition to judicial review
tends to be “a sometime thing,” with people supporting it for the few cases they
cherish (like Brown or Roe) and opposing it only when it leads to outcomes they
deplore.17 In politics, support for judicial review is sometimes intensely embroiled
in support for particular decisions. This is most notably true in the debate over
abortion rights, in which there is a panic-stricken refusal among pro-choice
advocates to even consider the case against judicial review for fear this will give
comfort and encouragement to those who regard Roe v. Wade as an unwarranted
intrusion on the rights of conservative legislators. I hope that setting out the core
case against judicial review in abstraction from its particular consequences can
help overcome some of this panic. It may still be the case that judicial review is
necessary as a protective against legislative pathologies relating to sex, race, or
religion in particular countries. But even if that is so, it is worth figuring out
whether that sort of defense goes to the heart of the matter, or whether it should be
regarded instead as an exceptional reason to refrain from following the tendency

Deliberation, Disagreement, and Voting]; Jeremy Waldron, Judicial Power and Popular
Sovereignty, in MARBURY VERSUS MADISON: DOCUMENTS AND COMMENTARY 181 (Mark Graber &
Michael Perhac eds., 2002) [hereinafter Waldron, Judicial Power and Popular Sovereignty];
Jeremy Waldron, A Right-Based Critique of Constitutional Rights, 13 OXFORD J. LEGAL STUD.
18 (1993) [hereinafter Waldron, A Right-Based Critique].
15. See KRAMER, supra note 11; TUSHNET, supra note 11. [AU: do you have a particular part of
either of these books in mind? If so, please provide citations. (NO – JW)]
16. Frank Michelman, Jacket Blurb on KRAMER, supra note 11.
17. CHARLES L. BLACK, JR., A NEW BIRTH OF FREEDOM: HUMAN RIGHTS, NAMED AND UNNAMED 109
(1997).

105
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

of what, in most circumstances, would be a compelling normative argument

against the practice.
A connected reason for boiling the flesh off the bones of the theoretical
critique is that judicial review is an issue for other countries that have a different
history, a different judicial culture, and different experience with legislative
institutions than the United States has had. For example, when the British debate
the relatively limited powers their judges have to review legislation, they are not
particularly interested in what the Republicans said to the Federalists in 1805 or
in the legacy of Brown v. Board of Education. What is needed is some general
understanding, uncontaminated by the cultural, historical, and political
preoccupations of each society.18
My own writing on this has been more abstract than most. But I have managed
to discuss judicial review in a way that embroils it with other issues in
jurisprudence and political philosophy.19 I am not satisfied that I have stated in a
clear and uncluttered way what the basic objection is, nor do I think I have given
satisfactory answers to those who have criticized the arguments I presented in
Law and Disagreement and elsewhere.
In this Essay, I shall argue that judicial review is vulnerable to attack on two
fronts. It does not, as is often claimed, provide a way for a society to focus clearly
on the real issues at stake when citizens disagree about rights; on the contrary it
distracts them with side-issues about precedent, texts, and interpretation. And it

18. Again, this is not to dismiss the more fleshed-out accounts. The idea is that we take a clear
view of the theoretical argument and put it alongside our richer understanding of the way the
debate unfolds in, to name a few examples, Britain, the United States, Canada, and South
Africa.
19. I have asked whether the very idea of individual rights commits us to judicial review in
Waldron, A Right-Based Critique, supra note 14. I have considered its relation to civic
republican ideas in Jeremy Waldron, Judicial Review and Republican Government, in THAT
EMINENT TRIBUNAL: JUDICIAL SUPREMACY AND THE CONSTITUTION 159 (Christopher Wolfe ed.,
2004), its relation to the difference between Benthamite and Rousseauian conceptions of
democracy in Jeremy Waldron, Rights and Majorities: Rousseau Revisited, in NOMOS XXXII:
MAJORITIES AND MINORITIES 44 (John W. Chapman & Alan Wertheimer eds., 1990) [hereinafter
Waldron, Rights and Majorities], and its relation to Continental theories of popular sovereignty
in Waldron, Judicial Power and Popular Sovereignty, supra note 14. I have considered the
relation of the judicial review controversy to debates in meta-ethics about realism and the
objectivity of values. See Jeremy Waldron, The Irrelevance of Moral Objectivity, in NATURAL
LAW THEORY 158 (Robert P. George ed., 1992) [hereinafter Waldron, The Irrelevance of Moral
Objectivity]; Jeremy Waldron, Moral Truth and Judicial Review, 43 AM. J. JURIS. 75 (1998). I
have responded to various defenses of judicial review, ranging from the precommitment case,
see Jeremy Waldron, Precommitment and Disagreement, in CONSTITUTIONALISM: PHILOSOPHICAL
FOUNDATIONS 271 (Larry Alexander ed., 1998) [hereinafter Precommitment and Disagreement],
to the particular argument that Ronald Dworkin makes in Freedom’s Law about its ultimate
compatibility with democracy, see Jeremy Waldron, Judicial Review and the Conditions of
Democracy, 6 J. POL. PHIL. 335 (1998).

106
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

is politically illegitimate, so far as democratic values are concerned: by

privileging majority voting among a small number of unelected and unaccountable
judges, it disenfranchises ordinary citizens and brushes aside cherished
principles of representation and political equality in the final resolution of issue
about rights.

i. definit io n o f j udicial review

I begin with a brief account of what I mean by judicial review. This is an

Essay about judicial review of legislation, not judicial review of executive action
or administrative decisionmaking.20 The question I want to address concerns
primary legislation enacted by the elected legislature of a polity. It might be
thought that some of the same arguments apply to executive action as well: After
all, the executive has some elective credentials of its own with which to oppose
decisionmaking by judges. But it is almost universally accepted that the
executive’s elective credentials are subject to the principle of the rule of law, and,
as a result, that officials may properly be required by courts to act in accordance
with legal authorization.21 The equivalent proposition for legislators has been
propounded too: judicial review just is the subjection of the legislature to the rule
of law. But in the case of the legislature, it is not uncontested; indeed that is
precisely the contestation we are concerned with here.
There are a variety of practices all over the world that could be grouped under
the general heading of judicial review of legislation. They may be distinguished
along several dimensions. The most important difference is between what I shall
call strong judicial review and weak judicial review. My target is strong judicial
review.22

20. Much of what is done by the European Court of Human Rights is judicial review of executive
action. Some of it is judicial review of legislative action, and actually some of it is also judicial
review of judicial action. See Seth F. Kreimer, Exploring the Dark Matter of Judicial Review: A
Constitutional Census of the 1990s, 5 WM. & MARY BILL RTS. J. 427, 458-459 [SC2: check this
pincite; it was added at sc1] (1997), for the claim that the majority of constitutional decisions
by lower federal courts in the United States concern challenges to the actions of low-level
bureaucrats rather than of legislatures.
21. Seana Shiffrin, Richard Pildes, Frank Michelman, and others have urged me to consider how
far my argument against judicial review of legislation might also extend to judicial review of
executive action in the light of statutes enacted long ago or statutes whose provisions require
extensive interpretation by the courts. Clearly more needs to be said about this. Pursuing the
matter in this direction might be considered a reductio ad absurdum of my argument or an
attractive application of it.
22. The distinction between strong and weak judicial review is separate from the question of
judicial supremacy. Judicial supremacy refers to a situation in which (1) the courts settle
important issues for the whole political system, (2) those settlements are treated as absolutely
binding on all other actors in the political system, and (3) the courts do not defer to the

107
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

In a system of strong judicial review, courts have the authority to decline to

apply a statute in a particular case (even though the statute on its own terms
plainly applies in that case) or to modify the effect of a statute to make its
application conform with individual rights (in ways that the statute itself does not
envisage). Moreover, courts in this system have the authority to establish as a
matter of law that a given statute or legislative provision will not be applied, so
that as a result of stare decisis and issue-preclusion a law that they have refused
to apply becomes in effect a dead letter. A form of even stronger judicial review
would empower the courts to actually strike a piece of legislation out of the
statute-book altogether. Some European courts have this authority.23 American
courts do not,24 but the real effect of their authority is not much short of it.25
In a system of weak judicial review, by contrast, courts may scrutinize
legislation for its conformity to individual rights but they may not decline to apply
it (or moderate its application) simply because rights would otherwise be
violated.26 Nevertheless, the scrutiny may have some effect. In the United
Kingdom, the courts may review a statute with a view to issuing a “Declaration of
Incompatibility” in the event that “the court is satisfied that the provision is
incompatible with a Convention right”—i.e. with one of the rights set out in the
European Convention of Human Rights as incorporated into British Law through
the Human Rights Act. The Act provides that such declaration “does not affect
the validity, continuing operation or enforcement of the provision in respect of
which it is given; and . . . is not binding on the parties to the proceedings in

positions taken on these matters in other branches (not even to the extent to which they defer to
their own past decisions under a limited principle of stare decisis). See Barry Friedman, The
History of the Countermajoritarian Difficulty, Part One: The Road to Judicial Supremacy, 73
N.Y.U. L. REV. 333, 352 & n.63 (1998)[SC2: check pincite; SC1 added it]; Jeremy Waldron,
Judicial Power and Popular Sovereignty, supra note 14, at 191-198 [SC2: check pincite; SC1
added it].
23. See Mauro Cappelletti & John Clarke Adams, Comment [SC2: Commentary?], Judicial Review
of Legislation: European Antecedents and Adaptations, 79 HARV. L. REV. 1207, 1222 (1966).
There are further complications in regard to whether the statute declared invalid is deemed to
have been invalid as of the time of its passage.
24. A contrary impression may appear from McCorvey v. Hill, 385 F.3d 846, 849 (5th Cir. 2004), in
which the Fifth Circuit held that the Texas abortion statute at issue in Roe v. Wade, 410 U.S.
113 (1973), must be deemed to have been repealed by implication. A close reading of that
case, however, shows that the implicit repeal was held to have been effected by the Texas
statutes regulating abortion after Roe, not by the decision in Roe itself. (I am grateful to Carol
Sanger for this reference.)
25. See Richard H. Fallon, Jr., As-applied and Facial Challenges and Third-party Standing, 113
HARV. L. REV. 1321, 1339 (2000).
26. See Stephen Gardbaum, The New Commonwealth Model of Constitutionalism, 49 AM. J. COMP.
L. 707 (2001).

108
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

which it is made.”27 But still it has an effect: A minister may use such a
declaration as authorization to initiate a fast-track legislative procedure to remedy
the incompatibility.28 (This is a power the minister would not have but for the
process of judicial review that led to the declaration in the first place.)
A form of even weaker judicial review would give judges not even that much
authority. Like their British counterparts, the New Zealand courts may not decline
to apply legislation when it violates human rights (in New Zealand, the rights set
out in the Bill of Rights Act of 199029); but they may strain to find interpretations
that avoid the violation.30 Although courts there have indicated that they may be
prepared on occasion to issue declarations of incompatibility on their own
initiative, such declarations in New Zealand do not have any legal effect on the
legislative process.31
There are some intermediate cases. In Canada, there is provision for the
review of legislation by courts, and courts there, like their U.S. counterparts, may
decline to apply a national or provincial statute if it violates the provisions of the
Canadian Charter of Rights and Freedoms. But Canadian legislation (provincial or
national) may be couched in a form that insulates it from this scrutiny—Canadian
assemblies may legislate “notwithstanding” the rights in the Charter.32 In
practice, however, the notwithstanding clause is rarely invoked.33 Thus, in what
follows I shall count the Canadian arrangement as a form of strong judicial review,

27. Human Rights Act, 1998, c. 42, §4(2), (6).

28. Id. § 10.
29. New Zealand Bill of Rights Act 1990, 1990 S.N.Z. No. 109, § 4 (“No court shall, in relation to
any enactment (whether passed or made before or after the commencement of this Bill of
Rights), . . . [h]old any provision of the enactment to be impliedly repealed or revoked, or to be
in any way invalid or ineffective; or . . . [d]ecline to apply any provision of the enactment—by
reason only that the provision is inconsistent with any provision of this Bill of Rights.”).
30. Id. § 6 (“Wherever an enactment can be given a meaning that is consistent with the rights and
freedoms contained in this Bill of Rights, that meaning shall be preferred to any other
meaning.”)
31. See Moonen v. Film & Literature Bd. of Review, [2000] 2 N.Z.L.R. 9, 22-3 (C.A.).
32. Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule
B to the Canada Act 1982, ch. 11, § 33(1) (U.K.). The full text of the provision reads:
(1) Parliament or the legislature of a province may expressly declare in an Act of Parliament or of
the legislature, as the case may be, that the Act or a provision thereof shall operate
notwithstanding a provision included in section 2 or sections 7 to 15 of this Charter. (2) An Act
or a provision of an Act in respect of which a declaration made under this section is in effect
shall have such operation as it would have but for the provision of this Charter referred to in the
declaration. [ME: We need to format this as a block quote. Also, each of the numbers is a
separate paragraph in the source]
33. When it has been invoked, it has mostly been in the context of Québécois politics. See Tsvi
Kahana, The Notwithstanding Mechanism and Public Discussion: Lessons from the Ignored
Practice of Section 33 of the Charter, 44 J. INST. PUB. ADMIN. CAN. 255 (2001).

109
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

with its vulnerability to my argument affected only slightly by the formal

availability of the override.34
A second distinction among types of judicial review pays attention to the
place of individual rights in the constitutional system of a society. In the United
States, statutes are scrutinized for their conformity to individual rights as set out
in the Constitution. Rights-oriented judicial review is part and parcel of general
constitutional review, and the courts strike down statutes for violations of
individual rights in exactly the spirit in which they strike down statutes for
violations of federalism or separation of powers principles.35 This gives American
defenses of judicial review a peculiar cast. Though philosophical defenses of the
practice are often couched in terms of the judiciary’s particular adeptness at
dealing with propositions about rights, in reality that argument is subordinate to a

34. Jeffrey Goldsworthy has suggested that the notwithstanding provision provides a sufficient
answer to those of us who worry, on democratic grounds, about the practice of strong judicial
review. Jeffrey Goldsworthy, Judicial Review, Legislative Override, and Democracy, 38 WAKE
FOREST L. REV. 451, 454-59 (2003). It matters not, he says, that the provision is rarely used.
[S]urely that is the electorate’s democratic prerogative, which Waldron would be
bound to respect. It would not be open to him to object that an ingenuous electorate
is likely to be deceived by the specious objectivity of constitutionalised rights, or
dazzled by the mystique of the judiciary—by a naive faith in judges’ expert legal
skills, superior wisdom, and impartiality. That objection would reflect precisely the
same lack of faith in the electorate’s capacity for enlightened self-government that
motivates proponents of constitutionally entrenched rights. [me: is this block quote
formatted correctly?]
Id. at 456-57. I believe that [ME: There are weird left margin issues throughout this entire footnote]
the real problem is that section 33 requires the legislature to misrepresent its position on rights.
To legislate notwithstanding the Charter is a way of saying that you do not think Charter rights
have the importance that the Charter says they have. But the characteristic stand-off between
courts and legislatures does not involve one group of people (judges) who think Charter rights
are important and another group of people (legislators) who do not. What it usually involves is
groups of people (legislative majorities and minorities, and judicial majorities and minorities)
all of whom think Charter rights are important, though they disagree about how the relevant
rights are to be understood. Goldsworthy acknowledges this:
When the judiciary . . . is expected to disagree, with the legislature as to the “true”
meaning and effect of Charter provisions, the legislature cannot ensure that its view
will prevail without appearing to override the Charter itself. And that is vulnerable to
the politically lethal objection that the legislature is openly and self-confessedly
subverting constitutional rights.
Id. at 467-68. However, maybe there is no form of words that can avoid this difficulty. As a matter of
practical politics, the legislature is always somewhat at the mercy of the courts’ public
declarations about the meaning of the society’s Bill or Charter of Rights. I am grateful to John
Morley for this point.
35. The most famous judicial defense of judicial review, Marbury v. Madison, 5 U.S. (1 Cranch)
137 (1803), had nothing to do with individual rights. It was about Congress’s power to appoint
and remove justices of the peace.

110
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

defense of the structural role the courts must play in upholding the rules of the
Constitution. Sometimes these two defenses are consistent; other times, they come
apart. For example, textualism may seem appropriate for structural issues, but it
can easily be made to seem an inappropriate basis for thinking about rights, even
when the rights are embodied in an authoritative text.36 In other countries,
judicial review takes place with regard to a Bill of Rights that is not specifically
designated as part of the (structural) constitution. Weak judicial review in the
United Kingdom on the basis of the Human Rights Act is of this kind. Because
most cases of strong judicial review are associated with constitutional review, I
shall focus on these cases. But it is important to remember not only that an
approach oriented to structural constraints might not be particularly appropriate
as a basis for thinking about rights, but also the additional point that many of the
challenges that are posed to rights-oriented judicial review can be posed to other
forms of constitutional review as well. In recent years, for example, the Supreme
Court of the United States has struck down a number of statutes because they
conflict with the Supreme Court’s vision of federalism.37 Now, everyone concedes
that the country is governed on a quite different basis so far as the relation
between state and central government is concerned than it was at the end of the
eighteenth century, when most of the Constitutional text was ratified, or in the
middle of the nineteenth century, when the text on federal structure was last
modified to any substantial extent. But opinions differ as to what the new basis of
state/federal relations should be. The text of the Constitution does not settle that
matter. So it is settled instead by voting among justices—some voting for one
conception of federalism (which they then read into the Constitution), the others
for another, and whichever side has the most votes on the Court prevails. It is not
clear that this is an appropriate basis for the settlement of structural terms of
association among a free and democratic people.38
A third distinction is between a posteriori review of the American kind, which
takes place in the context of particular legal proceedings, sometimes long after a
statute has been enacted, and ex ante review of legislation by a constitutional
court specifically set up to conduct an abstract assessment of a bill in the final

36. See DWORKIN, supra note 3, 11-18; ANDREI MARMOR, INTERPRETATION AND LEGAL THEORY 156-
57 (rev. 2d ed. 2005).
37. See, e.g., cases cited supra note 12.
38. The need for judicial review for patrolling structural limits on the allocation of authority
between state and federal legislatures is often cited (opportunistically) by defenders of rights-
based limitations on legislatures. People say, “Legislatures are subject to judicial review
anyway, for federalism reasons. So why not exploit that practice to develop rights-based judicial
review as well?” My analysis of the desirability of rights-based judicial review will be pertinent
to this sort of hybrid or opportunistic argument.

111
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

stages of its enactment.39 There are questions about how to understand ex ante
review. Something that amounts in effect to a final stage in a multicameral
legislative process, with the court operating like a traditional senate, is not really
judicial review (though the case against empowering an unelected body in this
way may be similar).40 I shall not say much more about this. For some defenses of
judicial review, the a posteriori character of its exercise—its rootedness in
particular cases41—is important, and I shall concentrate on that.
A fourth distinction is connected with the third. Judicial review can be carried
out by ordinary courts (as in the Massachusetts case we began with) or it can be
carried out by a specialized constitutional court. This may be relevant to an
argument I will make later: The ability of judges in the regular hierarchy of courts
to reason about rights is exaggerated when so much of the ordinary discipline of
judging distracts their attention from direct consideration of moral arguments.
Perhaps a specialist constitutional court can do better, though experience
suggests that it too may become preoccupied with the development of its own
doctrines and precedents in a way that imposes a distorting filter on the rights-
based reasoning it considers.
v

ii. fo ur ass umpt io ns

To focus my argument, and to distinguish the core case in which the objection
to judicial review is at its clearest from non-core cases in which judicial review
might be deemed appropriate as an anomalous provision to deal with special
pathologies, I shall set out some assumptions.42

39. Some systems of the first kind make provision for ex ante advisory opinions in limited
circumstances. For example, in Massachusetts, “[e]ach branch of the legislature . . . [has]
authority to require the opinions of the justices of the supreme judicial court, [SC2: check
source—is Supreme Judicial Court capitalized?]upon important questions of law,” or when
“having some action in view, [it] has serious doubts as to [its] power and authority to take such
action, under the Constitution.” Answer of the Justices, 364 Mass. 838, 844, 302 N.E.2d 565
(1973). [SC2: This quotation and source are messed up. Check the original source] This
procedure was used in the months following the Goodridge decision, discussed at the beginning
of this Essay. In Opinions of the Justices to the Senate, 802 N.E.2d 565 (Mass. 2004), the
Supreme Judicial Court of Massachusetts held that a legislative provision for civil unions for
same-sex couples that also prohibited discrimination against civilly joined spouses would not
be sufficient to avoid the constitutional objection to the ban on same-sex marriages noted in
Goodridge.
40. See Jeremy Waldron, Eisgruber’s House of Lords, 37 U.S.F. L. REV. 89 (2002).
41. See infra Section IV.A.
42. These assumptions are adapted from those set out in Jeremy Waldron, Some Models of Dialogue
Between Judges and Legislators, 23 SUP. CT. L. REV. 2d [SC2: check this “2d” thing] 7, 9-21
(2004).

112
v

43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

Certain of these assumptions may strike some readers as question-begging,

but I am not trying any sort of subterfuge here. The reasons for beginning with
these assumptions will be evident as we go along, and the possibility of non-core
cases, understood as cases in which one or more of these assumptions does not
hold, is freely acknowledged and will be considered in Part VII. In effect, my
contention will be that the argument against judicial review is conditional; if any
of the conditions fail, the argument may not hold.43 Let me add that part of what I
want to combat in this Essay is a certain sort of bottom-line mentality toward the
issue of judicial review.44 I fully expect that some readers will comb quickly
through my assumptions to find some that do not apply, say, to American or
British society as they understand it, leading them to ignore the core argument
altogether. What matters to them is that judicial review be defended and
challenges to it seen off; they don’t particularly care how. That is an unfortunate
approach. It is better to try and understand the basis of the core objection, and to
see whether it is valid on its own terms, before proceeding to examine cases in
which, for some reason, its application may be problematic.
Let me lay out in summary the four assumptions I shall make. We are to
imagine a society with: (1) democratic institutions in reasonably good working
order, including a representative legislature elected on the basis of universal
adult suffrage; (2) a set of judicial institutions, again in reasonably good order, set
up on a nonrepresentative basis to hear individual lawsuits, settle disputes, and
uphold the rule of law; (3) a commitment on the part of most members of the
society and most of its officials to the idea of individual and minority rights; and
(4) persisting, substantial, and good-faith disagreement about rights (i.e., about
what the commitment to rights actually amounts to and what its implications are)
among the members of the society who are committed to the idea of rights.
I shall argue that, relative to these assumptions, the society in question ought
to settle the disagreements about rights that its members have using its legislative
institutions. If these assumptions hold, the case for consigning such
disagreements to judicial tribunals for final settlement is weak and unconvincing,
and there is no need for decisions about rights made by legislatures to be second-
guessed by courts. And I shall argue that allowing decisions by courts to override
legislative decisions on these matters fails to satisfy important criteria of political
legitimacy.
I will proceed as follows. First, in this Part I elaborate each of the four
assumptions noted above. Then, in Part III, I review the general character of the

43. See infra text accompanying note 136.

44. For general critique of the “bottom-line” mentality in political philosophy, see Jeremy
Waldron, What Plato Would Allow, in NOMOS XXXVII: THEORY AND PRACTICE 138 (Ian Shapiro
& Judith Wagner DeCew eds., 1995).

113
v

43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

argument I propose to make. That argument will attend to both outcome-and

process-related reasons, and these will be discussed in Parts IV and V,
respectively. In Part VI, I will expose the fallacy of the most common argument
against allowing representative institutions to prevail: that such a system
inevitably leads to the tyranny of the majority. Finally, in Part VII, I shall say a
little bit about non-core cases—that is, cases in which there is reason to depart
from the assumptions on which the core argument depends.

A. Democratic Institutions v

I assume that the society we are considering is a democratic society and that,
like most in the modern Western world, it has struggled through various forms of
monarchy, tyranny, dictatorship, or colonial domination to a situation where its
laws are made and its public policies are set by the people and their
representatives working through elective institutions. This society has a broadly
democratic political system with universal adult suffrage, and it has a
representative legislature, to which elections are held on a fair and regular
basis.45 I assume that this legislature is a large deliberative body, accustomed to
dealing with difficult issues, including important issues of justice and social
policy. The legislators deliberate and vote on public issues, and the procedures
for lawmaking are elaborate and responsible,46 and incorporate various
safeguards, such as bicameralism,47 robust committee scrutiny, and multiple
levels of consideration, debate, and voting. I assume that these processes connect
both formally (through public hearings and consultation procedures) and
informally with wider debates in the society. Members of the legislature think of
themselves as representatives, in a variety of ways, sometimes making the
interests and opinions of their constituents key to their participation, sometimes
thinking more in terms of virtual representation of interests and opinions
throughout the society as a whole. I assume too that there are political parties, and
that legislators’ party affiliations are key to their taking a view that ranges more
broadly than the interests and opinions of their immediate constituents.

45
Thus, the application of my argument to nondemocratic societies, or societies whose
institutions differ radically from these forms, is not a subject discussed in this Essay.
[ME: format]
46. See Jeremy Waldron, Legislating with Integrity, 72 FORDHAM L. REV. 373 (2003).
47. The assumption of bicameralism might seem problematic. There are in the world a number of
well-functioning unicameral legislatures, most notably in the Scandinavian countries:
Denmark, Norway, and Sweden. But unicameral arrangements can easily exacerbate other
legislative pathologies. For an argument that this has happened in New Zealand to an extent
that may take that country outside the benefit of the argument developed in this Essay, see
Jeremy Waldron, Compared to What?—Judicial Activism and the New Zealand Parliament,
[sc2: volume and page needed. Missing at sc1] N.Z. L.J. (forthcoming December 2005).

114
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

None of this is meant to be controversial; it picks out the way in which

democratic legislatures usually operate. In general, I am assuming that the
democratic institutions are in reasonably good order. They may not be perfect and
there are probably ongoing debates as to how they might be improved. I assume
these debates are informed by a culture of democracy, valuing responsible
deliberation and political equality. The second of these values—political
equality—is worth particular emphasis. I assume that the institutions, procedures,
and practices of legislation are kept under constant review from this perspective,
so that if there are perceived inequities of representation that derogate seriously
from the ideal of political equality, it is understood among all the members of the
society that this is an appropriate criticism to make and that, if need be, the
legislature and the electoral system should be changed to remedy it. And I
assume that the legislature is capable of organizing such change, either on its own
initiative or by referendum.48 v

I belabor these points about a democratic culture and electoral and legislative
institutions in reasonably good working order because they will be key to the
argument that follows. The initial structure of the argument will be to ask the
following question: Once we have posited this first assumption, what reason can
there be for wanting to set up a nonelective process to review and sometimes
override the work that the legislature has done? On the other hand, I do not want
to beg any questions with this initial assumption. I shall balance it immediately
with the assumption that the society we are postulating also has courts in good
working order—this will be the second assumption—doing reasonably well what
courts are good at doing. The society we are contemplating has what it takes to
have a system of judicial review, if judicial review can be shown to be
appropriate.
One note of caution: When I say that the institutions are in good working
order, I am not assuming that the legislation that the reasonably democratic
legislature enacts is by and large good or just, so far as its content is concerned. I
assume some of the legislation is just and some of it unjust (people will disagree
about which is which), and that this is true both of the measures that might
conceivably be subject to judicial review and of the measures that nobody is

48. It is sometimes said that elective institutions are incapable of reforming themselves because
legislators’ have an entrenched interest in the status quo. This may be true of some of the
pathological electoral and legislative arrangements in the United States. (But the issues for
which this is most true in the United States are those on which the courts have scarcely dared
to intervene—consider the disgraceful condition of American redistricting arrangements, for
example.) It is patently false elsewhere. In New Zealand, for example, in 1993 the legislature
enacted statutes changing the system of parliamentary representation from a first-past-the-post
system to a system of proportional representation, in a way that unsettled existing patterns of
incumbency: see Electoral Act 1993 (1993 no. 87) and Electoral Referendum Act 1993 (1993
no. 86).

115
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

proposing to subject to judicial review. All that I have said about the legislative
and electoral arrangements being in good working order goes to process values
rather than outcome values. In Part V, however, I shall say more about the sort of
reasoning that we would expect to see in such a process.

B. Judicial Institutions

I assume that the society we are considering has courts—that is, a well-
established and politically independent judiciary, again in reasonably good
working order, set up to hear lawsuits, settle disputes, and uphold the rule of law.
I assume that these institutions are already authorized to engage in judicial review
of executive actions, testing it against statutory and constitutional law.
I assume that, unlike the institutions referred to in the previous Section, the
courts are mostly not elective or representative institutions. By this I mean not
only that judicial office is not (for the most part) an elective office, but also that
the judiciary is not permeated with an ethos of elections, representation, and
electoral accountability in the way that the legislature is. Many defenders of
judicial review regard this as a huge advantage, because it means courts can
deliberate on issues of principle undistracted by popular pressures and
invulnerable to public anger. Sometimes, however, when it is thought necessary to
rebut the democratic case against judicial review defenders of the practice will
point proudly to states where judges are elected. This happens in some states in
the United States. But even where judges are elected, the business of the courts is
not normally conducted, as the business of the legislature is, in accordance with
an ethos of representation and electoral accountability.
I am going to assume that, in the society we are considering, courts are
capable of performing the functions that would be assigned to them under a
practice of judicial review. They could review legislation; the question is whether
they should, and if so, whether their determinations should be final and binding
on the representative branches of government. I assume, though, that if they are
assigned this function, they will perform it as courts characteristically perform
their functions. There is an immense law review literature on the specific
character of the judicial process and of the tasks for which it does and does not
seem institutionally competent.49 I do not want to delve deeply into that here. As I
indicated above, I will assume that we are dealing with courts that (1) do not act
on their own motion or by abstract reference, but rather respond to particular
claims brought by particular litigants; (2) deal with issues in the context of

49. See, e.g., HENRY M. HART, JR. & ALBERT M. SACKS, THE LEGAL PROCESS: BASIC PROBLEMS IN
THE MAKING AND APPLICATION OF LAW 640-47 (William N. Eskridge, Jr. & Philip P. Frickey
eds., 1994); Lon L. Fuller, Forms and Limits of Adjudication, 92 HARV. L. REV. 353 (1978).

116
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

binary, adversarial presentation; and (3) refer to and elaborate their own past
decisions on matters that seem relevant to the case at hand. I further assume a
familiar hierarchy of courts, with provisions for appeal, and with larger
multimember bodies (perhaps five or nine judges) addressing cases at the highest
level of appeal, with lower courts being required largely to follow the lead of
higher courts in the disposition of the matters that come before them.
In some societies, judges are specially and separately trained; in other
societies, they are chosen from the ranks of eminent lawyers and jurists. In either
case, I assume that they have high status in the political system and a position
that insulates them from specific political pressures. In other regards, I assume
they are typical of the high-status and well-educated members of their society.
This is important for two reasons. First, because the society prides itself on being
largely democratic, I shall assume that the judges share some of that pride and so
are likely to be self-conscious about the legitimacy of their own activity if they
engage in judicial review of legislation. This may affect how they exercise such
authority.50 Second, although judges are likely to be at least as committed to
rights as anyone else in the society, I assume that like other members of the
society judges disagree with one another about the meaning and implications of
individual and minority rights. That is, I assume they are subject to my fourth
assumption about rights-disagreement, and that this too affects how they exercise
powers of judicial review (if they have such powers). Specifically, just like
legislators, modes of decisionmaking have to be developed for multi-judge
tribunals whose members disagree about rights. The decision-procedure most
often used is simple majority-voting. In Part V, I will address the question of
whether this is an appropriate procedure for judges to use.

C. A Commitment to Rights v

I assume that there is a strong commitment on the part of most members of the
society we are contemplating to the idea of individual and minority rights.
Although they believe in the pursuit of the general good under some broad
utilitarian conception, and although they believe in majority-rule as a rough
general principle for politics, they accept that individuals have certain interests
and are entitled to certain liberties that should not be denied simply because it
would be more convenient for most people to deny them. They believe that
minorities are entitled to a degree of support, recognition, and insulation that is
not necessarily guaranteed by their numbers or by their political weight.

50. See JESSE H. CHOPER, JUDICIAL REVIEW AND THE NATIONAL POLITICAL PROCESS (1980)
(discussing the Supreme Court’s legitimacy in this context); see also Planned Parenthood of Se.
Pa. v. Casey, 505 U.S. 833, 864-69 (1992) (same).

117
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

The details of the prevalent theory of rights need not detain us here. I assume
that this society-wide commitment to rights involves an awareness of the
worldwide consensus on human rights and of the history of thinking about rights.51
I assume that this commitment is a living consensus, developing and evolving as
defenders of rights talk to one another about what rights they have and what those
rights imply. I assume that the commitment to rights is not just lip service and
that the members of the society take rights seriously: they care about them, they
keep their own and others’ views on rights under constant consideration and lively
debate, and they are alert to issues of rights in regard to all the social decisions
that are canvassed or discussed in their midst.
No doubt there are skeptics about rights in every society, but I assume that
this position is an outlier. Some reject rights as they reject all political morality;
others reject rights because they hold utilitarian, socialist, or other doctrines that
repudiate them for (what purport to be) good reasons of political morality—e.g.,
rights are too individualistic or their trumping force undermines the rational
pursuit of efficiency or whatever. But I assume that general respect for individual
and minority rights is a serious part of a broad consensus in the society, part of
the most prevalent body of political opinion, and certainly part of the official
ideology.
To make this third assumption more concrete, we may assume also that the
society cherishes rights to an extent that has led to the adoption of an official
written bill or declaration of rights of the familiar kind. I shall refer to this
throughout as “the Bill of Rights” of the society concerned. This is supposed to
correspond to, for example, the rights provisions of the U.S. Constitution and its
amendments, the Canadian Charter of Rights and Freedoms, the European
Convention on Human Rights (as incorporated, say, into British law in the Human
Rights Act), or the New Zealand Bill of Rights Act. Those familiar with the last of
these examples will recognize that I am making no assumption that the “Bill of
Rights” is entrenched or part of a written constitution. I want to leave that open.
All I assume at this stage is that a Bill of Rights has been enacted to embody the
society’s commitment to rights. Thus, it may have been enacted sometime in the
past on the society’s own initiative, or it may be the product of imitation, or it may
be a fulfillment of the country’s external obligations under human rights law.
Readers may be puzzled by these assumptions. On the one hand, I appear to
be arguing against interest, stacking the deck in favor of judicial review by
assuming a Bill of Rights. On the other hand, it may seem that something sneaky

51
This is so even if this awareness does not involve much more than a vague
understanding that human rights conventions have become ascendant in the world since
1945, and that their history reaches back to the sort of conceptions of natural right
alluded to in documents such as the 1776 Declaration of Independence and the 1789
French Declaration of the Rights of Man and the Citizen. [ME: format]

118
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

is in the offing. Readers may be aware that I have argued in the past that judicial
review should not be understood as a confrontation between defenders of rights
and opponents of rights but as a confrontation between one view of rights and
another view of rights.52 What I want to emphasize in response to both these
observations is that there is a distinction both at the cultural and at the
institutional level between a commitment to rights (even a written commitment to
rights) and any particular institutional form (e.g., judicial review of legislation)
that such a commitment may take. I am tired of hearing opponents of judicial
review denigrated as being rights-skeptics. The best response is to erect the case
against judicial review on the ground of a strong and pervasive commitment to
rights.
This third assumption defines as non-core cases societies where the
commitment to rights is tenuous and fragile. It may seem strange or unfair to
proceed in this way, for defenders of judicial review do sometimes argue that we
need the practice to help shore up our commitment to rights, to teach participants
in a new democracy to value rights, or to give guarantees to minorities that might
not be forthcoming in a pure majority-rules system. Such arguments are
interesting, but they do not go to the heart of the case that is made for judicial
review in countries like the United States, Britain, or Canada. In those countries,
we are told that judicial review is an appropriate way of institutionalizing or
administering a society’s existing commitment to rights. These formulations
should be taken at face value, and that is what I am doing with my third
assumption.53

D. Disagreement About Rights

My final and crucial assumption is that the consensus about rights is not
exempt from the incidence of the general disagreement about all major political
issues that we find in modern liberal societies. So I assume that there is
substantial dissensus as to what rights there are and what they amount to. Some of
these disagreements are apparent at a philosophical level (e.g., whether
socioeconomic rights should be included in the Bill of Rights), some become

52. See Waldron, A Right-Based Critique, supra note 14, at 28-31, 34-36. [SC2: check pincite; SC1
added it]
53. My approach here is similar to that of John Rawls. I am using this device of the core case to
define something like a well-ordered society with a publicly accepted theory of justice. See,
e.g., JOHN RAWLS, POLITICAL LIBERALISM 35-36 (1993) [hereinafter RAWLS, POLITICAL
LIBERALISM]. Rawls seems to assume that judicial review of legislation is appropriate for even a
well-ordered society. Id. at 165-66, 233-40 [SC2: Check these pincites]; see also JOHN RAWLS,
A THEORY OF JUSTICE 195-99, 228-31 (1971) [hereinafter JOHN RAWLS, A THEORY OF JUSTICE].
One of my aims is to show that he is wrong about that.

119
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

apparent when we try to relate abstract principles of right to particular legislative

proposals (e.g., whether the free exercise of religion demands exemptions from
otherwise generally applicable laws), and some become apparent only in the
context of hard individual cases (e.g., how much tolerance for dissident speech
there should be in a time of national emergency).
I assume that the rights-disagreements are mostly not issues of interpretation
in a narrow legalistic sense. They may present themselves in the first instance as
issues of interpretation, but they raise questions of considerable practical moment
for the political community. Elsewhere I have referred to these as “watershed”
issues of rights.54 They are major issues of political philosophy with significant
ramifications for the lives of many people. Moreover, I assume that they are not
idiosyncratic to the society in which they arise. They define major choices that
any modern society must face, choices that are reasonably well understood in the
context of existing moral and political debates, choices that are focal points of
moral and political disagreement in many societies. Examples spring quickly to
mind: abortion, affirmative action, the legitimacy of government redistribution or
interference in the marketplace, the rights of criminal suspects, the precise
meaning of religious toleration, minority cultural rights, the regulation of speech
and spending in electoral campaigns, and so on.
As these examples suggest, disagreements about rights are often about central
applications, not just marginal applications. Because I am already assuming a
general commitment to rights, it is tempting to infer that that general commitment
covers the core of each right and that the right only becomes controversial at the
outer reaches of its application. That is a mistake. A commitment to rights can be
wholehearted and sincere even while watershed cases remain controversial. For
example, two people who disagree about whether restrictions on racist hate-
speech are acceptable may both accept that the right to free speech is key to
thinking through the issue and they may both accept also that the case they
disagree about is a central rather than marginal issue relative to that right. What
this shows, perhaps, is that they have different conceptions of the right, 55 but that
is no reason to doubt the sincerity of their adherence to it.
Generally speaking, the fact that people disagree about rights does not mean
that there must be one party to the disagreement who does not take rights
seriously. No doubt some positions are held and defended disingenuously or
ignorantly by scoundrels (who care nothing for rights) or moral illiterates (who
misunderstand their force and importance). But I assume that in most cases

54. See Waldron, Judicial Power and Popular Sovereignty, supra note 14, at 198. [SC2: check
pincite; SC1 added it]
55. For a discussion of the distinction between the concept of a right and various conceptions of it,
see RONALD DWORKIN, TAKING RIGHTS SERIOUSLY 134-36 (1977).

120
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

disagreement is pursued reasonably and in good faith. The issues involved are
serious issues on which it is not reasonable to expect that there would be
consensus. In other words, I assume something like John Rawls’s “burdens of
judgment,” but applied (where Rawls hesitated to apply the doctrine) to issues of
the right as well as issues of the good.56 It is not reasonable to expect that people’s
views on complex and fraught issues of rights will always converge to consensus.
And as Rawls emphasizes, it is “unrealistic . . . to suppose that all our differences
[on these matters] are rooted solely in ignorance and perversity, or else in the
rivalries for power, status, or economic gain.”57
The assumption of disagreement has nothing to do with moral relativism. One
can recognize the existence of disagreement on matters of rights and justice—one
can even acknowledge that such disagreements are, for practical political
purposes, irresolvable—without staking the meta-ethical claim that there is no
fact of the matter about the issue that the participants are disputing. The
recognition of disagreement is perfectly compatible with there being a truth of the
matter about rights and the principles of constitutionalism—assuming that our
condition is not one in which the truth of the matter discloses itself in ways that
are not reasonably deniable.58 ce qui veut dire qu’aucune preuve n’est possible
If there is a Bill of Rights, I assume that it bears on, but does not resolve, the
issues at stake in the disagreements. I mentioned some examples a few
paragraphs back. In the United States, it is indisputable both that the provisions
of the Bill of Rights have a bearing on how each of these issue is to be resolved
and that the provisions of the Bill of Rights do not themselves determine a
resolution of the issue in a way that is beyond reasonable dispute. Thus, I assume
that the extent of these disagreements belies our ingenuity at devising abstract
formulations. Disagreement does not prevent the enactment of a Bill of Rights.59
But the disagreements remain unresolved, leaving us in a situation in which—
when an issue about a possible rights-violation arises—it is beyond dispute that a
Bill of Rights provision bears on the matter, but what its bearing is and whether it

56. See RAWLS, POLITICAL LIBERALISM, supra note 53, at 55-60 (discussing “the burdens of
judgment”). Rawls argues that “many of our most important judgments are made under
conditions where it is not to be expected that conscientious persons with full powers of reason,
even after free discussion, will arrive at the same conclusion.” For an argument applying this to
the right as well as the good, see WALDRON, supra note 14, at Ch. 7 [sc2: missing at sc1. cite
the pages rather than the chapter].
57 [me: format this footnote number (“57”)] RAWLS, POLITICAL LIBERALISM, supra note 51, at 58.
[LE: Check the accuracy]
58. See Waldron, Irrelevance of Moral Objectivity, supra note 19, at 182.
59. See Thomas Christiano, Waldron on Law and Disagreement, 19 L. & PHIL. [SC2: Check the
starting page number for this source ], at 537 (2000).

121
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

prohibits (or should limit the application of) the legislative provision that is called
into question remains a matter of dispute among reasonable people.60
This is not to deny that arguments can be made that seem conclusive—at
least to those who make them—as to the bearing of the Bill of Rights on the issue
in question. If judicial review is set up in the society, then lawyers will argue
about these issues using both the text and the gravitational force of the text of the
Bill of Rights. In fact, lawyers will have a field day. Each side to each of the
disagreements will claim that its position can be read into the bland commitments
of the Bill of Rights if only those texts are read generously (or narrowly) enough.
Neither will be prepared to acknowledge that which I am assuming now will be
obvious: that the bland rhetoric of the Bill of Rights was designed simply to
finesse the real and reasonable disagreements that are inevitable among people
who take rights seriously for long enough to see such a Bill enacted. Instead of
encouraging us to confront these disagreements directly, judicial review is likely
to lead to their being framed as questions of interpretation of those bland
formulations. Whether that is a desirable context in which to deliberate about the
moral issues that they pose is one of the things we shall consider in Part V.

iii. t he for m o f t he arg ument

So these are our assumptions. What do we do with the situation they define?
The members of the community are committed to rights, but they disagree about
rights. Most issues of rights are in need of settlement. We need it not so much to
dispose of the issue—nothing can do that61—but to provide a basis for common
action when action is necessary. Now, there are all sorts of issues on which we do
not need society-wide settlement—transubstantiation, the meaning of Hamlet, the
value of a purely contemplative life—and that is fortunate, because there is little
prospect of agreement in these areas. Unfortunately, on issues of rights, where we
do need settlement, there is little prospect of agreement either. The need for
settlement does not make the fact of disagreement evaporate; rather, it means that
a common basis for action has to be forged in the heat of our disagreements.

60. Once again, I am not saying that the provisions in the Bill of Rights cover the central cases,
with disagreement confined to the margins of their application. The provisions are usually
vague and abstract, leaving open the possibility that even when there are uncontroversial cases,
people still might be using the same abstract formula to cover different substantive approaches
to the right—and we should still say that they both take the right seriously.
61. Cf. JON STEWART ET AL., AMERICA (THE BOOK): A CITIZEN’S GUIDE TO DEMOCRACY INACTION 90
(2004) (discussing Roe v. Wade and noting: “The Court rules that the right to privacy protects a
woman’s decision to have an abortion and the fetus is not a person with constitutional rights,
thus ending all debate on this once-controversial issue.”).

122
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

In the real world, the need for settlement confronts us in the legislative arena.
We legislate in certain areas, and the legislation we enact raises issues of rights.
Those issues may not be facially prominent in the legislation. The legislation may
be on marriage formalities, minimum working hours, campaign finance reform, or
the historic preservation of city centers, but what happens is that somebody
notices that its application happens to raise an issue of rights and it is in
connection with that issue—is the legislation to be applied according to its terms
or not?—that the need for settlement arises.
An argument, which I respect, for some sort of power of judicial review goes
as follows: It may not always be easy for legislators to see what issues of rights are
embedded in a legislative proposal brought before them; it may not always be
easy for them to envisage what issues of rights might arise from its subsequent
application. So it is useful to have a mechanism that allows citizens to bring these
issues to everyone’s attention as and when they arise. But this is an argument for
weak judicial review only, not for a strong form of the practice in which the
abstract question of right that has been identified is settled in the way that a court
deems appropriate. It is an argument for something like the system in the United
Kingdom, in which a court may issue a declaration that there is an important
question of rights at stake.62 Alternatively, it is an argument for the arrangement
we find in systems of even weaker review, whereby the attorney general has the
nonpartisan duty to scrutinize legislative proposals and publicly identify any
issues of rights that they raise.63 Such an arrangement is a kind of
institutionalization of the alertness to issues of rights that was embodied in
assumption three above.
Let us assume, for now, that the legislature is broadly aware of the issues of
rights that a given bill gives rise to and that, having deliberated on the matter, it
resolves—through debate and voting—to settle those issues in a particular way.
The legislature takes sides, as it were, on one or more of the disagreements we
imagined in assumption four. The question we face is whether that resolution of
the legislature should be dispositive or whether there is reason to have it second-
guessed and perhaps overruled by the judiciary.

62. See supra text accompanying notes 26-28.

63. Cf. New Zealand Bill of Rights Act 1990, 1990, S.N.Z. No. 109, § 7 (“Where any Bill is
introduced into the House of Representatives, the Attorney-General shall . . . as soon as
practicable after the introduction of the Bill—bring to the attention of the House of
Representatives any provision in the Bill that appears to be inconsistent with any of the rights
and freedoms contained in this Bill of Rights.”). For a controversial example of the exercise of
this power, see Grant Huscroft, Is the Defeat of Health Warnings a Victory for Human Rights?
The Attorney-General and Pre-legislative Scrutiny for Consistency with the New Zealand Bill of
Rights, PUB. L. REV. 109 (2003). [sc2: missing at sc1—get volume # and check carefully]

123
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

How should we answer this question? I have heard people say that the
decision-rule should be this: the legislature’s decision stands, except when it
violates rights. But clearly this will not do. We are assuming that the members of
the society disagree about whether a given legislative proposal violates rights. We
need a way of resolving that disagreement. The point is as old as Hobbes: We
must set up a decision-procedure whose operation will settle, not reignite, the
controversies whose existence called for a decision-procedure in the first place.64
This means that even though the members of the society we are imagining
disagree about rights, they need to share a theory of legitimacy for the decision-
procedure that is to settle their disagreements. So, in thinking about the reasons
for setting up such a procedure, we should think about reasons that can be
subscribed to by people on both sides of any one of these disagreements.65
I am presenting the need for legitimate decision-procedures as a response to
the problem of moral disagreement. But I have heard philosophers say that
because disagreement is pervasive in politics we should not let it throw us off our
stride. Because we disagree as much about legitimate decision-procedures as we
do about the justification of outcomes, and because (on my own account) it is
plain that we have to take a stand on something—namely, decision-procedures—
despite such disagreement, why can’t we just take a stand on the issue of
substance and be done with it?66 The response to this is that we must go to the
issue of legitimacy whether we are likely to find disagreement there or not. For
one thing, we do need to design a decision-procedure and we need to consider
reasons relevant to that design. For another thing, there are important reasons
relating to legitimacy—e.g., fairness, voice, participation—that arise because of
disagreement and do not arise apart from our addressing the question of decision-
procedures. Even if we disagree about these too, we have no choice but to
consider them. The fact that we will disagree about them is not a proper ground
for pushing them to one side and simply taking a stand on one side or the other in
the prior (or substantive) disagreement.

64. Cf. THOMAS HOBBES, LEVIATHAN 123 (Richard Tuck ed., 1996) [SC2: provide original date of
publication].
65. Another way of saying this is that a normative political theory needs to include more than just a
basis for justifying certain decisions on their merits. It needs to be more than, say, a theory of
justice or a theory of the general good. It also has to address the normative issue of the
legitimacy of the decision-procedures that are used to make political decisions in the face of
disagreement. A normative political theory that does not do that is seriously incomplete.
66. Christiano phrases the point in terms of a regress of procedures: “We can expect disagreement
at every stage, if Waldron is right; so if we must have recourse to a higher order procedure to
resolve each dispute as it arises, then we will be unable to stop the regress of procedures.”
Christiano, supra note 59, at 521. But Christiano makes no attempt to show that this is a vicious
regress. For discussion of the regress, see WALDRON, supra note 14, at 298-301. [sc2: missing
at sc1]

124
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

No decision-procedure will be perfect. Whether it is a process of

unreviewable legislation or whether it is a process of judicial review, it will
sometimes come to the wrong decision, betraying rights rather than upholding
them.67 This is a fact of life in politics. Everyone must concede that there will
sometimes be a dissonance between what they take to be the right choice and
what they take to be the choice yielded by the decision-procedure they regard as
legitimate. Richard Wollheim called this “a paradox in the theory of
democracy,”68 because it allows one and the same citizen to assert that A ought
not to be enacted, where A is the policy he voted against, and A ought to be
enacted, because A is the policy chosen by the majority. But Wollheim was wrong
to ascribe this paradox to democracy. It is a general paradox in the theory of
politics affecting any political theory that complements its account of what ought
to be done with an account of how decisions ought to be made when there is
disagreement about what ought to be done.
With that caution in mind, what are the reasons that need to be taken into
account in designing or evaluating a decision-procedure for settling
disagreements about rights? Two sorts of reasons may be considered. I shall call
them outcome-related and process-related reasons, though they are both relevant to
the issue of decision-procedure.
Process-related reasons are reasons for insisting that some person make, or
participate in making, a given decision that stand independently of considerations
about the appropriate outcome. In personal life, we sometimes say that a parent
has the right to make the decision as to whether her child should be disciplined
for a given infraction: It is not for an interested passer-by to make that decision.
We may say that while reserving judgment on whether the child should be
disciplined. Indeed, we may say that even though we think the passer-by is likely
to make a better decision on this than the parent. In politics, the most familiar
process-related reasons are those based on political equality and the democratic
right to vote, the right to have one’s voice counted even when others disagree with
what one says.
Outcome-related reasons, by contrast, are reasons for designing the decision-
procedure in a way that will ensure the appropriate outcome (i.e., a good, just, or
right decision). Our subject-matter is disagreements about rights. Because rights

67. I have heard people say that the errors are always likely to be worse on the legislative side: The
legislature may actually violate rights, whereas the worst that the courts can do is fail to
interfere to protect them. This is a mistake. Courts exercising the power of judicial review may
sometimes violate rights by striking down a statute that aims to protect them. I will discuss this
further at the end of Part IV.
68. Richard Wollheim, A Paradox in the Theory of Democracy, in PHILOSOPHY, POLITICS AND
SOCIETY 71, [SC2: need first page of the article and pincite for the quote] (Peter Laslett & W.G.
Runciman eds., 2d ser. 1969).

125
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

are important, it is likewise important that we get them right and so we must take
outcome-related reasons very seriously indeed. Wrong answers may be tolerable
in matters of policy; but on matters of principle, if the wrong answer is given, then
rights will be violated. The members of the society we are imagining understand
how important it is to avoid such outcomes or minimize them to the extent they
can.
Of course, it may not be easy to identify outcome-related reasons that people
on opposing sides of rights-disagreements can agree upon. As I said earlier,69 the
design of a decision-procedure must be independent of the particular
disagreement it is supposed to settle; it is no good if it simply reignites it. So we
must avoid outcome-related reasons that aim specifically at particular
controversial outcomes—e.g., favoring a decision-procedure because it is more
likely to generate a pro-choice than a pro-life outcome. A decision-procedure
chosen on this basis will hardly command the allegiance of the pro-life advocates.
Given the disagreement, the whole point here is to set up a procedure for
generating settlements in a way that can be recognized as legitimate on both
sides.
It is possible, however, to garner outcome-related reasons on a more modest
basis. Instead of saying (in a question-begging way) that we should choose those
political procedures that are most likely to yield a particular controversial set of
rights, we might say instead that we should choose political procedures that are
most likely to get at the truth about rights, whatever that truth turns out to be. As
Aileen Kavanagh puts it:
[W]e do not need a precise account of what rights we have and how they
should be interpreted in order to make some instrumentalist [i.e.,
outcome-related] claims. Many instrumentalist arguments are not based
on knowledge of the content of any particular rights. Rather, they are
based on general institutional considerations about the way in which
legislatures make decisions in comparison to judges, the factors which
influence their decision and the ways in which individuals can bring their
claims in either forum.70

Reasons of this kind deserve to be taken seriously. Joseph Raz has gone
further and suggested that these kinds of outcome-related reasons are the only

69. See supra text accompanying note 64.

70. Aileen Kavanagh, Participation and Judicial Review: A Reply to Jeremy Waldron, 22 L. & PHIL.
451, 466 (2003).

126
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

reasons worth considering.71 This dogmatism is based, presumably, on the

importance of the issues at stake. The outcomes of decisions about rights are
important. But there are also all sorts of important reasons that are not outcome-
related that we should not hesitate to also apply to the choices we make about the
design of procedures for the resolution of disagreements about rights. I have given
a few examples already, but here is another one: the principle of self-
determination. There is a reason for having these disagreements be settled for
each society within its own political system, rather than by diktat from outside
(e.g., by a neighboring government or a former colonial power). Some think this is
not a conclusive reason. They say that national self-determination and sovereignty
should sometimes give way to international authority on questions of human
rights.72 But few deny that it has some importance. Raz has paid insufficient
attention to the point that although outcome-related reasons are very important in
this area of decisionmaking about rights, reasons of other kinds may be important
too.
Once we see that there are reasons of all sorts in play, we have to consider
their normative character because this will affect how they relate to one another.
The term “outcome-related” sounds consequentialist. But because the
consequences we are trying to avoid are rights-violations, their avoidance has
some of the deontological urgency associated with rights. They may not be quite
as compelling as the principle that prohibits direct violations: The designers of a
decision-procedure are indirectly, not directly responsible for the violations that
might be involved in an exercise of that procedure. But their responsibility is still
a rights-based responsibility—there is a duty to take care in this regard.73
What about the normative character of the process-related reasons? Process-
related reasons are often matters of deontological urgency also. Ronald Dworkin, I
think, misstates the character of participatory reasons when he refers to them as
“[t]he participatory consequences of a political process.”74 He suggests that
allowing individual citizens the opportunity to play a part in the community’s
political decisionmaking has as a consequence—a good consequence—which is
that it confirms their equal membership or standing in the community. It
reassures them that they are regarded by others as persons whose opinions and
choices have value. Allowing people to participate also has the good consequence

71. Joseph Raz, Disagreement in Politics, 43 AM. J. JURIS. 25, 45-46 (1998); see also JOHN RAWLS,
A THEORY OF JUSTICE, supra note 53, at 230 (“The fundamental criterion for judging any
procedure is the justice of its likely results.”).
72. See, e.g., Louis Henkin, That “S” Word: Sovereignty, and Globalization, and Human Rights, Et
Cetera, 68 FORDHAM L. REV. 1 (1999).
73. For the idea of various waves of duty being generated by a particular right, see Jeremy
Waldron, Rights in Conflict, 99 ETHICS 503, 509-12 (1989).
74. RONALD DWORKIN, SOVEREIGN VIRTUE: THE THEORY AND PRACTICE OF EQUALITY 187 (2000).

127
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

of helping citizens to identify with the results of political decisions and to view
those decisions as in some sense theirs, with good knock-on effects for legitimacy
(in the sociologist’s sense of that word).75 All this is no doubt important. But it has
the flavor of a headmaster noting the advantages that may accrue from giving his
pupils a say in educational affairs through a school council. Dworkin’s account
radically underestimates the notion of a right to participate, the imperative that
one be treated as an equal so far as a society’s decisionmaking is concerned, the
sense of principle that is at stake when someone asks, “Why has my say been
excluded from this decision, which affects me and to which I am subject?”
So, how do we weight these process-related and outcome-related
considerations? We face the familiar problem of trying to maximize the value of
two variables, like asking someone to buy the fastest car at the lowest price. There
are various ways we can set up the question. We could ask: “What method is most
likely to get at the truth about rights, while at the same time adequately
respecting the equal claim to be heard of the voices of those affected?”76 Or we
could ask: “What method best respects the equal claim to be heard of the voices
of those affected, while at the same time being reasonably likely to get at the truth
about rights?” I think I can cut through this Gordian knot. What I will argue, in
Part IV, is that the outcome-related reasons are at best inconclusive. They are
important, but they do not (as is commonly thought) establish anything like a
clear case for judicial review. The process-related reasons, however, are quite
one-sided. They operate mainly to discredit judicial review while leaving
legislative decisionmaking unscathed. Thus, it seems to me the legislative side
wins on either formulation of the question. And that will be the core of the case
against judicial review.

iv. o ut co me- relat ed reaso ns

According to Raz, “[a] natural way to proceed is to assume that the

enforcement of fundamental rights should be entrusted to whichever political
decision-procedure is, in the circumstances of the time and place, most likely to
enforce them well, with the fewest adverse side effects.”77 [SC2: check original
source] I guess the discussion at this point ought to be continuous with the
broader debate about the institutional competence of courts, initiated by the legal

75. These summary formulations of Dworkin’s view are adapted from Kavanagh, supra note 70, at
458-59.
76. This is how the question is stated in FRANK MICHELMAN, BRENNAN AND DEMOCRACY 59-60
(1999). [SC2: missing at sc1]
77. Raz, supra note 71, at 45.

128
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

process school.78 Courts are good at deciding some issues and not others.
Technically, we use the term “rights” to denote the issues that courts
characteristically decide, because a plaintiff has to state a claim of right to be
heard in a court at all. But as Lon Fuller observed, it does not follow that courts
are therefore the appropriate forum for dealing with claims of right in the less
technical sense under consideration here.79 Some claims of right have the
character of the sort of binary issue that courts might be competent to address;
others have a multifaceted character that has usually been regarded as
inappropriate for decision in a judicial structure. This matter bears further
consideration. I will not say much more about it now, but will turn instead to the
more specific claims that are made about the competence of courts and
legislatures on the important moral issues that are the subject of this Essay.
It is tempting to associate outcome-related reasons with the case for judicial
review (and process-related reasons with the case against it). This is a mistake. It
is true that many of the more important process-related reasons are participatory
and therefore favor elective or representative institutions. But it does not follow
that all or most outcome-related reasons argue the other way. Outcome-related
reasons, as we shall see, cut in both directions. There are things about legislatures
that make them vulnerable sometimes to the sorts of pressures that rights are
supposed to guard against; but there are also things about courts that make it
difficult for them to grapple directly with the moral issues that rights-
disagreements present.
Raz acknowledges that outcome-related reasons may weigh on both sides. On
the one hand, he argues in familiar fashion that

[i]n many countries there are ample reasons to suspect that members of
the legislature are moved by sectarian interests to such a degree that they
are not likely even to attempt to establish what rights (some) people have.
. . . We may know that certain factors are likely to cloud people’s
judgments. They may be, for example, liable to be biased in their own
interest. We may therefore prefer a procedure in which those charged with
a decision are not affected, or not directly affected, by their own decision.
There are other factors known to bias judgment, and their nature and
presence can be established even without knowledge of the content of the
rights concerned.80

Now, in considering a charge like this, we have to ask about its compatibility
with our third assumption: Is this sort of sectarian prejudice typical of legislatures

78. See HART & SACKS, supra note 49, at 640-7.

79. Fuller, supra note 49, at 368-70.
80. Raz, supra note 71, at 46.

129
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

in all societies? Or should we associate it with the non-core case of a society

whose members are largely indifferent to rights? I shall say more about this in
Part VII.81 But even taken at face value, Raz’s argument is not univocal in its
tendency. The same sectarian pressures often explain judicial neglect of rights as
well. We have seen this in the United States in cases as diverse as Prigg, Dred
Scott, Schenck, and Korematsu.82 More recently, Laurence Tribe (usually a
stalwart defender of judicial review) observed that in the panic that afflicted
America after 9/11, “it would be a terrible mistake for those who worry about civil
rights and liberties to pin too much hope on the judiciary in times of crisis.”83
In any case, Raz acknowledges that outcome-related reasons also argue in the
opposite direction:
Sometimes . . . there are reasons for thinking that those whose interests
are not going to be affected by a decision are unlikely to try honestly to
find out what is just in the circumstances. Sometimes one may be unable
to appreciate the plight of classes of people unless one belongs to the
same class oneself, and therefore rather than entrusting the decision to
those not affected by it, it should be given to those who are so affected.84

Legislatures are set up with structures of representation precisely in order to

foster this sense of appreciation.
It is sometimes suggested that structures of democratic participation take no
cognizance at all of the independent importance of securing appropriate
outcomes—they just blindly empower the majority. This is nonsense. All
democracies limit the franchise in various ways in order to secure a modicum of
mature judgment at the polls. They exclude children from voting, for example,
even though children are affected by the decisions under consideration. Moreover,

81. See infra text accompanying notes 137-140. This is where I will deal with the claim (for non-
core cases) that judges who sympathize with minority rights are in a better position to resist
popular prejudice than legislators who sympathize with minority rights.
82. Korematsu v. United States, 323 U.S. 214 (1944) (refusing to protect citizens of Japanese
descent from internment during the Second World War); Schenck v. United States, 249 U.S. 47
(1919) (holding that criticizing conscription during the First World War was like shouting fire
in a crowded theater); Dred Scott v. Sanford, 60 U.S. (19 How.) 393, 493 (1857) (Campbell, J.,
concurring) (refusing to recognize that a person of African descent could be a citizen of
Missouri) [SC2/LE: isn’t this idea in the majority opinion as well? If so, cite it there]; Prigg v.
Pennsylvania, 41 U.S. (16 Pet.) 539, 612 (1842) (striking down state legislation that sought to
protect African-Americans from slave-catchers).
83. Laurence Tribe, Trial by Fury: Why Congress Must Curb Bush’s Military Courts, THE NEW
REPUBLIC, Dec. 10, 2001, at 18; Ronald Dworkin, The Threat to Patriotism, N.Y. REV. OF
BOOKS, Feb. 28, 2002, at 44 (noting courts’ past tolerance of rights-violations in times of
crisis).
84. Raz, supra note 71, at 46.

130
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

legislatures are constituted in a way that ensures that information about the
tolerability of various options to different sections of the society is fed into the
decision-process. And decisions are usually made in the context of bicameral
institutions, so that each legislative proposal has to secure majority support in
each of two houses on slightly different elective schedules.85 Furthermore,
systems with weak judicial review or no judicial review sometimes make specific
provision in the legislative process for issues of rights to be highlighted.86 Specific
provision is made in most democracies for carefully orchestrated debate around
election time, as well as a whole array of connections between formal debate in
the legislature and informal debate and accumulation of information outside the
legislature. All these are outcome-related adjustments to democratic procedures
What we see, on the participatory side, is not what Rawls called a claim of pure
procedural justice, but something like imperfect procedural justice.87 In general,
what I notice when I read outcome-related arguments in favor of judicial review is
that people assume that an outcome-related case must be able to be made in favor
of courts, if only because the most familiar arguments against judicial review are
non-outcome-related. People strain to associate outcome-related reasons with the
judiciary and in so doing they often peddle a quite unrealistic picture of what
judicial decisionmaking is like.88 Opponents of judicial review are often accused
of adopting a naively optimistic view of legislatures. But sometimes we do this
deliberately, matching one optimistic picture with another in the face of the
refusal of the defenders of courts to give a realistic account of what happens
there.89
In the remainder of this Part, I want to consider in more detail three outcome-
related advantages that are sometimes claimed for courts: (1) that issues of rights
are presented to courts in the context of specific cases; (2) that courts’ approach
to issues of rights is oriented to the text of a Bill of Rights; and (3) that reasoning

85. Some bicameral systems, like the United Kingdom, have a unelected upper house and
provisions (in the Parliament Acts and in some of the conventions of the British Constitution)
that allow the lower house to prevail (eventually) in the event of conflict.
86. See supra note 63 and accompanying text.
87. See RAWLS, A THEORY OF JUSTICE, supra note 53, at 84-85. We speak of pure procedural
justice when we want to indicate that there is nothing more to the justice of the outcome than
the fact that it was arrived at by scrupulously following a just procedure. We speak of
imperfect procedural justice when we want to convey the point that a given outcome must be
judged on its merits as well as on the basis of the procedure that yielded it.
88. For a general critique of arguments that associate judicial review with careful moral
deliberation among, for example, Justices on the U.S. Supreme Court, see KRAMER, supra note
11, at 240. Kramer gives a fine description of the way in which Justices’ political agendas, and
the phalanxes of ideologically motivated clerks in the various chambers, interfere with anything
that could be recognized as meaningful collegial deliberation.
89. See JEREMY WALDRON, THE DIGNITY OF LEGISLATION 2 (1999).

131
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

and reason-giving play a prominent role in judicial deliberation. These are said to
weigh in favor of judicial review. On all three counts, however, I shall argue that
there are important outcome-related defects in the way courts approach rights and
important outcome-related advantages on the side of legislatures.

A. Orientation to Particular Cases

People sometimes argue that the wonderful thing about judicial reasoning on
rights (as opposed to legislative reasoning on rights) is that issues of rights
present themselves to judges in the form of flesh-and-blood individual situations.
Rights, after all, are individual rights, and it helps focus the mind to see how an
individual is affected by a piece of legislation. As Michael Moore puts the point,
“[J]udges are better positioned for . . . moral insight than are legislatures because
judges have moral thought experiments presented to them everyday [sic] with the
kind of detail and concrete personal involvement needed for moral insight.”90
But this is mostly a myth. By the time cases reach the high appellate levels we
are mostly talking about in our disputes about judicial review, almost all trace of
the original flesh-and-blood right-holders has vanished, and argument such as it
is revolves around the abstract issue of the right in dispute. Plaintiffs or
petitioners are selected by advocacy groups precisely in order to embody the
abstract characteristics that the groups want to emphasize as part of a general
public policy argument. The particular idiosyncrasies of the individual litigants
have usually dropped out of sight by the time the Supreme Court addresses the
issue, and the Court almost always addresses the issue in general terms.91
The process of legislation is open to consideration of individual cases, through
lobbying, in hearings, and in debate. Indeed there is a tendency these days to
initiate legislation on the basis of notorious individual cases—Megan’s Law, for
example.92 Hard cases make bad law, it is sometimes said. To the extent that this
is true, it seems to me that legislatures are much better positioned to mount an

90. Michael Moore, Law as a Functional Kind, in NATURAL LAW THEORY, supra note 19, at 188,
230. For a response, see Waldron, Moral Truth and Judicial Review, supra note 19, at 83-8.
91. See Sarah Weddington, Roe v. Wade: Past and Future, 24 SUFFOLK U. L. REV. 601, 602-603
(1990). [SC2: check pincite; SC1 added it]
92. Megan’s Law (about creating a register of sex offenders) was enacted in New Jersey in 1995
(N.J. Stat. Ann. §§ 2C:7-1-7-11) after Megan Nicole Kanka was raped and murdered by a
convicted sex offender. There is also now a Federal Megan’s Law (42 U.S.C. § 14071 (2000).
For a description of the enactment of this legislation, see Daniel M. Filler, Making The Case
For Megan's Law: A Study In Legislative Rhetoric 76 Ind. L.J. 315 (2001).

132
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

assessment of the significance of an individual case in relation to a general issue

of rights that affects millions and affects them in many different ways.93

B. Orientation to the Text of a Bill of Rights

We are imagining a society with a Bill of Rights, and if there is to be judicial

review of legislation, it will presumably center around the Bill of Rights. The Bill
of Rights, we have assumed, has been adopted in the society pursuant to
members’ shared commitment to the idea of individual and minority rights
notwithstanding the fact that they disagree about what these rights are and what
they entail. Now, when rights-disagreements erupt in regard to legislation, there is
a question about the role that the established Bill of Rights should play in the
decision-process in which the issue is posed. From an outcome-related point of
view, is it a good idea or a bad idea that rights-disagreements be fought out in
relation to the terms of a Bill of Rights?
One reason for thinking it is a good idea is that the written formulations of the
Bill of Rights can help disputants focus on the abstract rights-issues at stake. But
there are powerful reasons on the other side. The forms of words used in the Bill
of Rights will not have been chosen with rights-disagreements in mind. Or, if they
were, they will have been chosen in order to finesse the disagreements about
rights that existed at the time the Bill of Rights was set up. Their platitudes may
be exactly the wrong formulations to focus clear-headed, responsible, and good-
faith explorations of rights-disagreements.
The written formulations of a Bill of Rights also tend to encourage a certain
rigid textual formalism.94 A legal right that finds protection in a Bill of Rights
finds it under the auspices of some canonical form of words in which the
provisions of the Bill are enunciated. One lesson of American constitutional
experience is that the words of each provision tend to take on a life of their own,
becoming the obsessive catch-phrase for expressing everything one might want to
say about the right in question. This may be less of a danger in a system of
legislative supremacy, because legislators can pose the issue for themselves if
they like without reference to the Bill of Rights’ formulations. But it is part of the

93. See also EISGRUBER, supra note 13, at 173 (“Judges take up constitutional issues in the course
of deciding controversies between particular parties. As a result, those issues come to them in a
way that is incomplete. . . . Not all interested parties will have standing to appear before the
court. Judges receive evidence and hear arguments from only a limited number of parties. …
As a result, judges may not have the information necessary to gain a comprehensive perspective
on the fairness of an entire social, political, or economic system.”). Eisgruber concludes from
this that it is probably unwise for judges to attempt to address issues that turn on what he calls
“comprehensive” moral principles. Id. at 165 and 171.
94. This is an argument I developed in Waldron, A Right-Based Critique, supra note 14.

133
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

modus operandi of courts to seek textual havens for their reasoning, and they will
certainly tend to orient themselves to the text of the Bill of Rights in a rather
obsessive way.
At the very least, courts will tend to be distracted in their arguments about
rights by side-arguments about how a text like the Bill of Rights is best
approached by judges. American experience bears this out: The proportion of
argument about theories of interpretation to direct argument about the moral
issues is skewed in most judicial opinions in a way that no one who thinks the
issues themselves are important can possibly regard as satisfactory. This is partly
because the legitimacy of judicial review is itself so problematic. Because judges
(like the rest of us) are concerned about the legitimacy of a process that permits
them to decide these issues, they cling to their authorizing texts and debate their
interpretation rather than venturing out to discuss moral reasons directly.95
One final point. The text of a Bill of Rights may distort judicial reasoning not
only by what it includes but also by what it omits. Suppose the members of a
given society about whether the Bill of Rights should have included positive
(socioeconomic) as well as negative (liberty) rights.96 Those who think positive
rights should have been included may think the present Bill of Rights distorts
moral reasoning by excluding them. A response may be that, at worst, this
omission just leads to a possible failure to review legislation in cases in which
review would be appropriate, but it is not an argument against judicial review as
such. But that’s too simple. A failure to include positive rights may alter (or
distort) judges’ understanding of the rights that are included. Judges may give
more weight to property rights or to freedom of contract, say, than they would if
property and freedom of contract were posited alongside explicit welfare rights.
And giving them greater weight may lead judges to strike down statutes that ought
not to be struck down—statutes that are trying to make up the deficiency and
implement by legislation those rights that failed to register in the formulations of
the Bill of Rights.

95. See also TUSHNET, supra note 11, at 60 (“Courts may design some doctrines to reflect their
sense of their own limited abilities, not to reflect directly substantive constitutional values.”).
96. See Jackson v. City of Joliet, 715 F.2d 1200, 1203-04 (7th Cir. 1983) (Posner, J.) (observing
that the American constitutional scheme “is a charter of negative rather than positive
liberties”); cf. Mark Tushnet, An Essay on Rights, 62 TEX. L. REV. 1363, 1393-94 (1984) (“We
could of course have a different Constitution. . . . One can argue that the party of humanity
ought to struggle to reformulate the rhetoric of rights so that Judge Posner’s description would
no longer seem natural and perhaps would even seem strained.”).

134
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

C. Stating Reasons

It is often thought that the great advantage of judicial decisionmaking on

issues of individual rights is the explicit reasoning and reason-giving associated
with it. Courts give reasons for their decisions, we are told, and this is a token of
taking seriously what is at stake, whereas legislatures do not. In fact, this is a
false contrast. Legislators give reasons for their votes just as judges do. The
reasons are given in what we call debate and they are published in Hansard or the
Congressional Record. The difference is that lawyers are trained to close study of
the reasons that judges give; they are not trained to close study of legislative
reasoning (though they will occasionally ransack it for interpretive purposes).
Perhaps this argument is not really about the presence or absence of reason-
giving, but rather about its quality. In my view, however, the reasons that courts
tend to give when they are exercising powers of judicial review of legislation are
seldom the reasons that would be canvassed in a full deliberative discussion, and
the process of searching for, citing, assessing, and comparing the weight of such
reasons is quite different for courts than for an ideal political deliberator. Partly
this is the point mentioned earlier—that the reasons will be oriented toward the
terminology of the Bill of Rights. If one is lucky enough to have a fine and up-to-
date Bill of Rights, then there may be some congruence between judicial reason-
giving and the reason-giving we would look for in fully rational, moral, or political
deliberation. But if one has an antiquated constitution, two or three hundred years
old, then the alleged reason-giving is likely to be artificial and distorted. In the
United States, what is called “reason-giving” is usually an attempt to connect the
decision the court is facing with some antique piece of ill-thought-through
eighteenth- or nineteenth-century prose. (For example, is an argument about
whether “substantive ‘due process’” is an oxymoron the best framework for
thinking about labor law or, for that matter, abortion rights?)
Courts’ reason-giving also involves attempts to construct desperate analogies
or disanalogies between the present decision they face and other decisions that
happen to have come before them (and in which they were engaged in similar
contortions). There is laborious discussion of precedent, even though it is
acknowledged at the highest levels of adjudication that precedent does not settle
the matter.97 (So there is also laborious discussion of the circumstances in which
precedent should or shouldn’t be overridden.98) And all the time, the real issues
at stake in the good-faith disagreement about rights get pushed to the margins.

97. See, e.g., Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88 COLUM. L.
REV. 723 (1988).
98. See, e.g., Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 854-69 (1992) (discussing the
circumstances in which long-standing constitutional precedents may be overturned).

135
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

They usually take up only a paragraph or two of the twenty pages or more devoted
to an opinion, and even then the issues are seldom addressed directly. In the
Supreme Court’s fifty-page opinion in Roe v. Wade, for example, there are but a
couple of paragraphs dealing with the moral importance of reproductive rights in
relation to privacy, and the few paragraphs addressed to the other moral issue at
stake—the rights-status of the fetus—are mostly taken up with showing the
diversity of opinions on the issue.99 Read those paragraphs: the result may be
appealing, but the “reasoning” is thread-bare.
I actually think there is a good reason for this. Courts are concerned about the
legitimacy of their decisionmaking and so they focus their “reason-giving” on
facts that tend to show that they are legally authorized—by constitution, statute,
or precedent—to make the decision they are proposing to make. This is an
understandable thing to do. But it counts heavily against the courts in the
outcome-related argument about the preferability of judicial review over
legislation.100 Distracted by these issues of legitimacy, courts focus on what other
courts have done, or what the language of the Bill of Rights is, whereas
legislators—for all their vices—tend at least to go directly to the heart of the
matter.101
In this regard, it is striking how rich the reasoning is in legislative debates on
important issues of rights in countries without judicial review.102 I recently read
through the House of Commons debates on the Medical Termination of Pregnancy
Bill from 1966.103 This was a bill proposing to liberalize abortion law. The second
reading debate on that bill is as fine an example of a political institution grappling
with moral issues as you could hope to find. It is a sustained debate—about 100

99. Roe v. Wade, 410 U.S. 113 (1973). There is a tremendous amount of legal and social history in
the opinion, but only a few pages address the actual moral issues at stake. See id. at 153-55
(discussing privacy and the importance of reproductive rights); id. at 159-60 (discussing the
alleged rights or personality of the fetus).
100. Eisgruber seems to concede this, acknowledging that “[t]oo often judges attempt to justify
controversial rulings by citing ambiguous precedents, and . . . veil their true reasons behind
unilluminating formulae and quotations borrowed from previous cases.” EISGRUBER, supra note
13, at 70; see also id. at 135 (“[J]udges . . . often . . . pretend that they are not making political
judgments themselves, and that their decisions were forced upon them by textual details or
historical facts.”).
101. There is an important point here that Mark Tushnet has emphasized. We should not be
criticizing legislators for failing to reason as judges do, for that may not be a smart way to
address the issues at stake. TUSHNET, supra note 11, at 63-65. [SC2: Check the pincite
carefully. SC1 added it.]
102. This is adapted from Waldron, supra note 46, at 373. [SC2: Check the pincite carefully. SC1
added it.]
103. In the British legislature, the second reading debate is when deliberation takes place on the
main principles of the bill.

136
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

pages in Hansard104—and it involved pro-life Labour people and pro-choice

Labour people, pro-life Conservatives and pro-choice Conservatives, talking
through and focusing on all of the questions that need to be addressed when
abortion is being debated. They debated the questions passionately, but also
thoroughly and honorably, with attention to the rights, principles, and pragmatic
issues on both sides. It was a debate that in the end the supporters of the bill won;
the pro-choice faction prevailed.105 One remarkable thing was that everyone who
participated in the debate, even the pro-life MPs (when they saw which way the
vote was going to go), paid tribute to the respectfulness with which their positions
had been listened to and heard in that discussion.106 Think about that: How many
times have we ever heard anybody on the pro-life side pay tribute to the attention
and respectfulness with which their positions were discussed, say, by the
Supreme Court in Roe v. Wade?107
In the United States, we congratulate ourselves on consigning issues of
individual rights to the courts for constitutional adjudication on the ground that
courts may be regarded as forums of principle, to use Ronald Dworkin’s famous
phrase.108 Indeed we sometimes say the British are backward for not doing things
that way.109 But the key difference between the British legislative debate and the
American judicial reasoning is that the latter is mostly concerned with
interpretation and doctrine, while in the former decisionmakers are able to focus
steadfastly on the issue of abortion itself and what it entails—on the ethical status

104. 732 PARL. DEB., H.C. (5th ser.) (1966) 1067, 1067-1166. [SC2: Missing at SC1]
105. The second reading debate was not the end, of course. There was a long committee stage and
then a third reading debate, and then similar debates in the House of Lords. But eventually the
liberalizing legislation was enacted. [SC2: verify that the citation in 104 covers this]
106. See, e.g., 732 PARL. DEB., H.C. (5th ser.) (1966) 1152. [SC2: missing at sc1] Norman St. John-
Stevas, a Catholic MP who voted against the bill, nevertheless began his argument by noting,
“[w]e all agree that this has been a vitally important debate, conducted on a level which is
worthy of the highest traditions of the House.” Id. He then moved on to congratulate the bill’s
sponsor “on the manner in which he introduced the Bill, which he did with extraordinary
moderation and skill.” Id.
107. When I mention this example, my American friends tell me that the British legislature is
organized to make forms of debate possible that are not possible in the United States. Well,
leaving aside the question of whether the United States should be regarded as a pathological
case, this is simply false. The debate I have just referred to worked because the House of
Commons suspended one of its distinguishing features—strong party discipline—for the
purpose of this issue of rights. MPs actually debated the matter much more in the style of their
American counterparts, not necessarily toeing a party line but stating their own opinions clearly
and forcefully.
108. DWORKIN, supra note 3, at 33, 69-71.
109. See Editorial, Half-Measures on British Freedoms, N.Y. TIMES, Nov. 17, 1997, at A22
(criticizing the Human Rights Act for not moving the United Kingdom wholeheartedly to a
system of strong judicial review).

137
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

of the fetus, on the predicament of pregnant women and the importance of their
choices, their freedom, and their privacy, on the moral conflicts and difficulties
that all this involves, and on the pragmatic issues about the role that law should
play in regard to private moral questions. Those are the issues that surely need to
be debated when society is deciding about abortion rights, and those are the
issues that are given most time in the legislative debates and least time in the
judicial deliberations.110
I am sure there is more to be said on the outcome-related question. It is
certainly the case that just as courts address questions of rights in ways that
distort what is really at stake, so too can legislative reasoning also be a disgrace,
as legislative majorities act out of panic, recklessly, or simply parrot popular or
sectarian slogans in their pseudo-debates. The question is this: Which defects in
deliberation should be regarded as normal and which as aberrations in the way
that the respective institutions—courts and legislatures—are supposed to
behave? Despite Dworkin’s rhetoric about “forums of principle,” I think courts
are expected to behave in the ways that I have criticized, focusing on precedent,
text, doctrine, and other legalisms. Our assumption about courts—assumption
two—is about institutions that behave in that way, indeed behave well by those
(legalistic) standards. In the case of legislatures, however, hasty or sectarian
legislating is not part of the normal theory of what legislatures are set up to do. It
is not what we should assume for the core case of legislative decisionmaking in a
society most of whose members respect rights. There may be some countries—
perhaps the United States—in which peculiar legislative pathologies have
developed. If that is so, then Americans should confine their non-core argument
for judicial review to their own exceptional circumstances.

v. pro cess -relat ed reaso ns

Among the reasons we have for setting up decision-procedures one way or

another, some may have little to do with outcomes, either particular outcomes or
outcomes in general. They are concerned instead with voice or fairness or other

110. Elena Kagan and others have suggested to me that this critique of the way courts discuss rights
is predicated on an assumption that what we are aiming to protect are moral rights. If, on the
other hand, what we value is the protection of our legal constitutional rights, then this mode of
discussion is not as inappropriate as my critique suggests. I am not convinced. What we aim to
protect is rights, and the question is what mechanisms available in the modern state are best at
protecting them and facilitating intelligent discussion about them. I do not assume that the
mode of discourse in a moral philosophy seminar is the appropriate one. What I am suggesting
here is that it is important, one way or another, to get at the real issues of human interests and
human liberties that are at stake in our disagreements. A legalistic way of proceeding may or
may not be the best way of doing that, but it would be quite wrong to say that we ought to value
the legalism as an end in itself.

138
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

aspects of the process itself. As I said earlier, it is often assumed that process-
related arguments weigh unequivocally against judicial review. This is not quite
true. Some feeble process-related arguments have been concocted by defenders of
the practice, and I shall review those at the end of this Part. But it is mostly true:
The preponderance of the process-related reasons weigh in favor of legislatures.
The question of the political legitimacy of decision-procedures in the face of
disagreement about outcomes may be posed as follows. (I am afraid this is going
to be quite abstract.)
We imagine a decision being made by a certain process and we imagine a
citizen Cn—who is to be bound or burdened by the decision—disagreeing with
the decision and asking why she should accept, comply, or put up with it. Some of
those who support the decision may try to persuade Cn that it is right in its
substance. But they may fail, not because of any obtuseness on her part, but
simply because Cn continues (not unreasonably) to hold a different view on this
vexed and serious matter. What then is to be said to Cn? A plausible answer may
be offered to her concerning the process by which the decision was reached. Even
though she disagrees with the outcome, she may be able to accept that it was
arrived at fairly. The theory of such a process-based response is the theory of
political legitimacy.
Political decision-procedures usually take the following form. Because there
is disagreement about a given decision, the decision is to be made by a
designated set of individuals {C1, C2, . . . Cm} using some designated decision-
procedure. The burden of legitimacy-theory is to explain why it is appropriate for
these individuals, and not some others, to be privileged to participate in the
decisionmaking. As Cn might put it, “Why them? Why not me?” The theory of
legitimacy will have to provide the basis of an answer to that question. Because
the problem is general—it is not just a matter of Cn’s idiosyncratic perversity—it
will have to give a similar answer to similar questions from Co and Cp and all the
other C’s not included in the set of privileged decisionmakers. But even if this
answer is accepted, the struggle is not over. The theory of legitimacy also has to
provide an answer to an additional question that Cn may pose: “In the decision-
procedure that was used, why wasn’t greater weight given to the views of those
decisionmakers who felt as I do about the matter?” There must be a defense of the
decision-procedure used by {C1, C2, . . . Cm}, not just defense of its membership.
Let us now make this abstract algebra more concrete. Suppose a citizen who
disagrees with a legislative decision about rights poses the two questions I have
envisaged. She asks: (1) “Why should this bunch of roughly five hundred men
and women (the members of the legislature) be privileged to decide a question of
rights affecting me and a quarter billion others?”; and (2) “Even if I accept the
privileging of this five hundred, why wasn’t greater weight given to the views of
those legislators who agreed with me?”

139
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

In democracies, legislatures are set up in ways that provide reasonably

convincing answers to these two questions. The answer to the first question is
provided by the theory of fair elections to the legislature, elections in which
people like Cn were treated equally along with all their fellow citizens in
determining who should be privileged to be among the small number participating
in decisions of this kind. The answer to the second question is given by the well-
known fairness arguments underlying the principle of majority-decision (MD). It
is not my task to defend this here. The fairness/equality defense of the majority-
decision rule is well known.111 Better than any other rule, MD is neutral as
between the contested outcomes, treats participants equally, and gives each
expressed opinion the greatest weight possible compatible with giving equal
weight to all opinions. When we disagree about the desired outcome, when we do
not want to bias the matter up-front one way or another, and when each of the
relevant participants has a moral claim to be treated as an equal in the process,
then MD—or something like it—is the principle to use.112
But what if someone responds as follows: I can see why individual citizens
like Cn have a right to be treated as equals in a decisionmaking process on a
matter that affects them all. But why do the five hundred representatives in the
legislature have a right to be treated as equals in this process? What justifies their
use of MD?
The answer refers to the continuity as between the answers to the first and
second questions in the case of legislatures. For legislatures, we use a version of
MD to choose representatives and we use a version of MD for decisionmaking
among representatives. The theory is that together these provide a reasonable
approximation of the use of MD as a decision-procedure among the citizenry as a
whole (and so a reasonable approximation of the application of the values
underlying MD to the citizenry as a whole).
In general, then, what we are saying to Cn is roughly as follows: You are not
the only one who makes this challenge to the decision-procedures we use. As a

111. For the theorem (in social choice theory) that MD alone satisfies elementary conditions of
fairness, equality, and rationality, see Kenneth D. May, A Set of Independent Necessary and
Sufficient Conditions for Simple Majority Decision, 20 ECONOMETRICA 680 (1952) [LE: check
this article]; and AMARTYA K. SEN, COLLECTIVE CHOICE AND SOCIAL WELFARE 71-74 (1970).
There are also useful discussions in CHARLES R. BEITZ, POLITICAL EQUALITY 58-67 (1989); and
ROBERT A. DAHL, DEMOCRACY AND ITS CRITICS 139-41 (1989).
112. Ronald Dworkin has convinced me, in conversation, that MD is not an appropriate principle to
use in regard to first-order issues of justice. If we were in an overcrowded life-boat and
somebody had to leave, it would not be appropriate to use MD to decide who that should be.
MD is an appropriate principle, however, for choosing among general rules. If someone in the
life-boat proposes that we should draw straws and someone else suggests that the oldest person
should be required to leave the life-boat, then MD seems a fair basis for choosing among these
rules.

140
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

matter of fact, millions of individuals do. And we respond to each of them by

conceding her point and giving her a say in the decision. In fact, we try to give her
as much of a say as we can, though of course it is limited by the fact that we are
trying to respond fairly to the case that can be made along the same lines to take
into account the voice of each individual citizen. We give each person the greatest
say possible compatible with an equal say for each of the others. That is our
principle. And we believe that our complicated electoral and representative
arrangements roughly satisfy that demand for political equality—that is, equal
voice and equal decisional authority.
Of course, in the real world, the realization of political equality through
elections, representation, and legislative process is imperfect. Electoral systems
are often flawed (e.g., by unsatisfactory arrangements for drawing district
boundaries or a lack of proportionality between districts) and so are legislative
procedures (e.g., by a system of seniority that compromises fairness in the
legislature). All this can be acknowledged. But remember our first assumption: a
set of legislative institutions—including a system of elections to the legislature
and a system of decisionmaking within it—that are in reasonably good shape so
far as these democratic values of equality and fairness are concerned. We are
assuming also that the legislators and their constituents keep this system under
review for its conformity to these principles. For example, in many democracies
there are debates about rival systems of proportional representation, districting,
and legislative procedure. Cn may complain that these systems are not perfect and
that they have not been reformed to the extent that they ought to have been. But a
good theory of legitimacy (for real-world polities) will have a certain looseness to
accommodate inevitable defects. It will talk about reasonable fairness, not perfect
fairness. No doubt some electoral and legislative systems fail even these generous
criteria. But our core case is not supposed to address situations in which the
legislative and electoral systems are pathologically or incorrigibly dysfunctional.
Let’s return to our core case and to the confrontation we are imagining with
our recalcitrant citizen Cn. That something along the lines described above can be
said in response to Cn’s complaint about the decision of a reasonably well-
organized legislature is important for legitimacy, but it is not conclusive. For Cn
may envisage a different procedure that is even more legitimate than the
legislative procedure is. Legitimacy is partly comparative.113 Because different
institutions and processes might yield different results, defending the legitimacy
of a given institution or process involves showing that it was or would be fairer
than some other institution or process that was available and might have reached
the contrary decision.114

113. See Waldron supra note 47, at [sc2: missing at sc1]

114. See MICHELMAN, supra note 76, at 57-59. [SC2: missing at sc1]

141
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

So now we imagine—or, in a system like the United States, we observe—

decisions being made not by a legislature but by a court (let’s make it the U.S.
Supreme Court) on a vexed issue of rights on which the citizens disagree. And a
citizen—again we’ll call her Cn—who disagrees with the substance of one of the
court’s decisions complains about it. She asks: (1) why should these nine men
and women determine the matter? and (2) even if they do, why should they make
their decision using the procedure that they use rather than a procedure that gives
more weight to Justices with a view that Cn favors?
These are much tougher questions for the Court to answer than they were for
legislators to answer. We have it on good authority that challenges like these are
often voiced noisily outside the Court and that the Justices are sometimes
distressed by them. Some of them, however, reflect on that distress. (It is time to
roll your eyes now and pay no attention for a few minutes, because I am going to
quote Justice Antonin Scalia and quote him at length.)
In truth, I am as distressed as the Court is . . . about the “political pressure”
directed to the Court: the marches, the mail, the protests aimed at inducing us to
change our opinions. How upsetting it is, that so many of our citizens (good
people, not lawless ones, on both sides of this abortion issue, and on various sides
of other issues as well) think that we Justices should properly take into account
their views, as though we were engaged not in ascertaining an objective law but in
determining some kind of social consensus. The Court would profit, I think, from
giving less attention to the fact of this distressing phenomenon, and more attention
to the cause of it. That cause permeates today’s opinion: a new mode of
constitutional adjudication that relies not upon text and traditional practice to
determine the law, but upon what the Court calls “reasoned judgment,” . . . which
turns out to be nothing but philosophical predilection and moral intuition.115 [ME:
Format this as a block quote]
Justice Scalia continued:
What makes all this relevant to the bothersome application of “political
pressure” against the Court are the twin facts that the American people
love democracy and the American people are not fools. As long as this
Court thought (and the people thought) that we Justices were doing
essentially lawyers’ work up here—reading text and discerning our
society’s traditional understanding of that text—the public pretty much
left us alone. Texts and traditions are facts to study, not convictions to
demonstrate about. But if in reality, our process of constitutional
adjudication consists primarily of making value judgments . . . then a free

115. Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 999-1000 (1992) (Scalia, J., dissenting)
(citation and emphasis omitted). [SC2: please add the emphasis back into this quote and the
next.]

142
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

and intelligent people’s attitude towards us can be expected to be (ought

to be) quite different. The people know that their value judgments are
quite as good as those taught in any law school—maybe better. If, indeed,
the “liberties” protected by the Constitution are, as the Court says,
undefined and unbounded, then the people should demonstrate, to protest
that we do not implement their values instead of ours.116

So, as Scalia says, the legitimacy-questions are front-and-center, and the

defenders of judicial review have to figure out a response.
First, why should these judges and these judges alone decide the matter? One
answer might be that the judges have been appointed and approved by
decisionmakers and decisionmaking bodies (the President and the Senate) who
have certain elective credentials. The President is elected and people often know
what sort of persons he is likely to appoint to the Supreme Court, and the U.S.
Senators who have to approve the appointments are elected also, and their views
on this sort of thing may be known as well. True, the judges are not regularly held
accountable in the way legislators are, but, as we have already remarked, we are
not looking for perfection.
So, the defender of judicial review is not altogether tongue-tied in response to
our citizen’s challenge; there is something to say. Nevertheless, if legitimacy is a
comparative matter, then it is a staggeringly inadequate response. The system of
legislative elections is not perfect either, but it is evidently superior as a matter of
democracy and democratic values to the indirect and limited basis of democratic
legitimacy for the judiciary. Legislators are regularly accountable to their
constituents and they behave as though their electoral credentials were important
in relation to the overall ethos of their participation in political decisionmaking.
None of this is true of judges.
Second, even if we concede that vexed issues of rights should be decided by
these nine men and women, why should they be decided by simple majority-
voting among the justices? Here, the situation gets worse for defenders of judicial
review. I have always been intrigued by the fact that courts make their decisions
by voting, applying the MD principle to their meager numbers. I know they
produce reasons and everything we discussed above. But in the end it comes
down to head-counting: five votes defeat four in the U.S. Supreme Court,
irrespective of the arguments that the Justices have concocted. If MD is
challenged in this context, can we respond to it in roughly the same way that we
imagined a response on behalf legislatures? Actually, no, we cannot. MD is
appropriate for persons who have a moral claim to insist on being regarded as
equals in some decision-process. But I cannot see any moral basis for this claim

116. Id. at 1000-01 (emphasis omitted).

143
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

in the case of Supreme Court Justices. They do not represent anybody. Their
claim to participate is functional, not a matter of entitlement.
I am handicapped here by the more or less complete lack of theoretical
attention to the use of MD in courts.117 Scholars have written some about our
empirical experience of voting and voting strategy on courts, and some have
suggested novel ways of combining judges’ votes on the particular issues involved
in each case, rather than on the overall outcome.118 But I am not aware of any
elementary defense of judicial majoritarianism.119 The usual fairness-and-equality
defense is unavailable. I suspect that if the use of MD by courts were to be
defended, it would be defended either as a simple technical device of decision
with no further theoretical ramifications,120 or on the basis of Condorcet’s jury
theorem (majority-voting by a group of adjudicators arithmetically enhances the
competence of the group beyond the average competence of its members).121 If it
is the latter, then the defense of MD is part of the outcome-related case for
judicial competence, which means that it will have to compete with a similar case
that can be made for the much larger voting bodies in legislatures.122 However
this argument would play out, my point is this: There is no additional fairness
argument for the use of MD by courts, as there is for its use by legislatures.

117. I try to say a little about it in Waldron, Deliberation, Disagreement, and Voting, supra note 14,
at 215-224.
118. See, e.g., Lewis A. Kornhauser & Lawrence G. Sager, The One and the Many: Adjudication in
Collegial Courts, 81 CAL. L. REV. 1 (1993); Lewis A. Kornhauser & Lawrence G. Sager,
Unpacking the Court, 96 YALE L.J. 82 (1986); David Post & Steven C. Salop, Rowing Against
the Tidewater: A Theory of Voting by Multijudge Panels, 80 GEO. L.J. 743 (1992).
119. One reason for this is that defenders of judicial review prefer not to talk about the use of simply
majority-voting among the Justices on issues of rights. They want to be able to condemn
majority-voting on rights as a characteristic of legislatures. If pressed, they will acknowledge
that, of course, judges decide issues by, say, 5-4 or 6-3 majorities on the Supreme Court. But I
have never, ever heard a defender of judicial review introduce this into discussion himself or
herself, let alone undertake to explain why it is a good idea.
120. See HANNAH ARENDT, ON REVOLUTION 163 (photo. reprint 1983) (1963), for the view that “the
principle of majority is inherent in the very process of decision-making” and is “likely to be
adopted almost automatically in all types of deliberative councils and assemblies.”
121. Marie-Jean-Antoine-Nicolas, Marquis de Condorcet, Essay on the Application of Mathematics to
the Theory of Decision-Making, in CONDORCET: SELECTED WRITINGS 33 (Keith Michael Baker
ed. & trans., 1976).
122. The Condorcet theorem holds that the larger the voting group, the greater the enhancement of
group competence above average individual voter competence by majority voting. Of course the
result presupposes that average individual competence is higher than fifty percent. For a
discussion of Condorcet’s doubts about the application of this last condition, see Jeremy
Waldron, Democratic Theory and the Public Interest: Condorcet and Rousseau Revisited, 83 AM.
POL. SCI. REV. 1317, 1322 (1989).

144
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

These last points should remind us that the responses we have been imagining
to Cn’s challenge to legislative and judicial procedures do not stand alone. We
may also make an outcome-related case to respond to her challenge. But I think I
have been able to show in this Part, and the previous Part, that the outcome-
related case is inconclusive (or it argues in favor of legislatures) while the
process-related case is almost wholly on the legislative side. Remember too what
we said at the end of Part III. The reasons on both sides have to do with rights. If
one institution or the other was clearly superior at determining what rights people
really have, then that would weigh very heavily indeed in favor of that institution.
But that is not the case. On the process side, institutions giving final authority on
these matters to judges fail to offer any sort of adequate response to the fairness-
complaint of the ordinary citizen based on the principle—not just the value—of
political equality. That failure might be tolerable if there were a convincing
outcome-based case for judicial decisionmaking. Defenders of judicial review
pretend that there is. But as we saw above, it is just unsupported assertion.
Perhaps aware of all this, defenders of judicial review have tried a number of
last ditch attempts to reconcile their favored institution to democratic values. I
will consider these briefly, because there is not much to them.
First, defenders of judicial review claim that judges do not make their own
decisions about rights; they simply enforce decisions of the people that are
embodied in a Bill of Rights, which itself has democratic credentials, either as
legislation or as part of a constitution. This claim does not undermine the core
case against judicial review. We are assuming that the Bill of Rights does not
settle the disagreements that exist in the society about individual and minority
rights. It bears on them but it does not settle them. At most, the abstract terms of
the Bill of Rights are popularly selected sites for disputes about these issues. The
question we have been considering is who is to settle the issues that are fought out
on those sites.
Second, and in much the same spirit, defenders of judicial review claim that
judges are simply enforcing the society’s own preexisting commitment
(precommitment) to rights. The society has bound itself to the mast on certain
principles of right, and, like Ulysses’ shipmates, the judges are just making sure
the ropes remain tied. This common analogy has been thoroughly discredited in
the literature.123 Briefly, the response is that the society has not committed itself
to any particular view of what a given right entails, so when citizens disagree
about this, it is not clear why giving judges the power to decide should be
understood as upholding a precommitment. If someone insists nevertheless that
society has committed itself to a particular view about the right in question (and

123. See JON ELSTER, ULYSSES UNBOUND 88-96 (2000 ) (casting doubt on some arguments made in
JON ELSTER, ULYSSES AND THE SIRENS: STUDIES IN RATIONALITY AND IRRATIONALITY 93 (1984));
see also WALDRON, supra note 14, at 255-81.

145
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

the judges, by voting among themselves, somehow ascertain that precommitment),

once an alternative understanding of the right is in play, it is not clear why the
existing precommitment should hold. The Ulysses model works only when the
precommitment guards against various aberrations, not when it guards against
changes of mind in relation to genuine disagreement as to what a reasonable
outcome would be.124
Third, defenders of judicial review claim that if legislators disagree with a
judicial decision about rights, they can campaign to amend the Bill of Rights to
explicitly override it. Their failure to do this amounts to a tacit democratic
endorsement. This argument is flawed because it does not defend the baseline
that judicial decisionmaking establishes. Amending a Bill of Rights
characteristically involves a supermajority; or if it is a British- or New Zealand-
style statute, it will have credentials in the political culture that raise the stakes
and increase the burden associated with the amendment effort. If our disgruntled
citizen Cn asks why the deck should be stacked in this way, the only answer we
can give her refers back to judicial decision. And that has already been found
wanting.
Fourth, defenders of judicial review insist that judges do have democratic
credentials: They are nominated and confirmed by elected officials, and the kind
of judicial nominations that a candidate for political office is likely to make
nowadays plays an important role in the candidate’s electoral campaign.125 This is
true; but (as I have already remarked) the issue is comparative and these
credentials are not remotely competitive with the democratic credentials of
elected legislators. Moreover, to the extent that we accept judges because of their
democratic credentials, we undermine the affirmative case that is made in favor of
judicial review as a distinctively valuable form of political decisionmaking.
Fifth and finally, defenders of judicial review claim that the practice may be
justified as an additional mode of access for citizen input into the political system.
Sometimes citizens access the system as voters, sometimes as lobbyists,
sometimes as litigants. They say we should evaluate the legitimacy of the whole
package of various modes of citizen access, not just the democratic credentials of
this particular component. The point is a fair one, as far as it goes. But embedding
judicial review in a wider array of modes of citizen participation does not alter the
fact that this is a mode of citizen involvement that is undisciplined by the
principles of political equality usually thought crucial to democracy. People tend

124. See WALDRON, supra note 14, at 266-70.

125. EISGRUBER, supra note 13, at 4 (“Though the justices are not chosen by direct election, they are
nevertheless selected through a process that is both political and democratic. . . . [T]hey are
chosen by elected officials: they are nominated by the president and confirmed by the Senate. .
. . The justices have . . . a democratic pedigree: they owe their appointments to their political
views and their political connections as much as . . . to their legal skills.”).

146
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

to look to judicial review when they want greater weight for their opinions than
electoral politics would give them. Maybe this mode of access can be made to
seem respectable when other channels of political change are blocked.126 We will
discuss this in Part VII But the attitudes toward one’s fellow citizens that judicial
review conveys are not respectable in the core case we are considering, in which
the legislature and the elective arrangements are in reasonably good shape so far
as democratic values are concerned.

vi. t he tyr anny o f t he maj or ity

I want to give defenders of judicial review—for the core case—one last bite at
the apple. The concern most commonly expressed about the work of a democratic
legislature is that, because they are organized on a majoritarian basis, legislative
procedures may give expression to the “tyranny of the majority.” So widespread is
this fear, so familiar an element is it in our political culture, so easily does the
phrase “tyranny of the majority” roll off our tongues,127 that the need for judicially
patrolled constraints on legislative decisions has become more or less axiomatic.
What other security do minorities have against the tyranny of the majority?
I believe that this common argument is seriously confused. Let us grant, for
now, that tyranny is what happens to someone when their rights are denied. The
first thing to acknowledge is that, according to this definition, tyranny is almost
always going to be at stake in any disagreement about rights. In other words, in
any disagreement about rights, the side in favor of the more expansive
understanding of a given right (or the side that claims to recognize a right that the
other denies) will think that the opposite side’s position is potentially tyrannical.
For example, the peyote smokers will think their subjection of their sacraments to
generally applicable narcotics laws is tyrannical. Opponents of campaign finance
laws will think those laws are tyrannical. But it is an open question whether they
are right. Some of these claims about tyranny are no doubt correct. But they do
not become correct simply because they are asserted. Indeed in some cases, there
will be allegations of tyranny on both sides of a rights issue. Defenders of abortion
rights think the pro-life position would be tyrannical to women; but the pro-life
people think the pro-choice position is tyrannical to another class of persons
(fetuses are persons, on their account). Some think that affirmative action is

126. Cf. JOHN HART ELY, DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW (1980).
127. Mill’s one criticism of Tocqueville’s Democracy in America was that the likely political effect of
his popularizing the phrase, “the tyranny of the majority,” would be to give conservative forces
additional rhetoric with which to oppose progressive legislation. See JOHN STUART MILL, M. de
Tocqueville on Democracy in America (1840), in 2 DISSERTATIONS AND DISCUSSIONS: POLITICAL,
PHILOSOPHICAL, AND HISTORICAL 79, 81 (1882). [SC2: Check to see whether p. 79 supports]

147
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

tyrannical; others think the failure to implement affirmative action programs is

tyrannical. And so on.
Let us grant what we acknowledged in Part III, in our discussion of
Wollheim’s paradox. Democratic institutions will sometimes reach and enforce
the incorrect decision about rights. This means they will sometimes act
tyrannically. But the same is true of any decision process. Courts will sometimes
act tyrannically as well.128 Tyranny, on the definition we are using, is more or less
inevitable. It is just a matter of how much tyranny there is likely to be, which was
the subject of our discussion in Part IV.
Is the tyranny of a political decision aggravated by the fact that it is imposed
by a majority? I leave aside the pedantic point that a court may also reach its
decision by majority-voting. Is tyranny by a popular majority (e.g., a majority of
elected representatives, each supported by a majority of his constituents) a
particularly egregious form of tyranny? I do not see how it could be. Either we say
that tyranny is tyranny irrespective of how (and among whom) the tyrannical
decision is made, or we say—and this is my view—that the majoritarian aspect
actually mitigates the tyranny, because it indicates that there was at least one
non-tyrannical thing about the decision: It was not made in a way that
tyrannically excluded certain people from participation as equals.
That may seem a little flip, so let me address the question less provocatively.
The most commonly expressed misgiving about unrestrained legislative authority
is that minorities or individuals may suffer oppression in relation to the majority.
They may be oppressed, or discriminated against, or their rights denied and
violated compared to those of the majority, or their interests unduly subordinated
to those of members of the majority (for example, harmed or neglected in a way
that justice condemns). In describing these forms of tyranny, oppression, or
injustice, we use the terms “majority” and “minority.” But in this particular
context they are not necessarily terms related to political decision-processes. Let
me explain.
Injustice is what happens when the rights or interests of the minority are
wrongly subordinated to those of the majority. Now, we have conceded that this
may happen as a result of majoritarian political decisionmaking. When it does,
however, we need to distinguish at least in the first instance between the
decisional majority and minority and what I shall call the topical majority and

128. I am not referring to their sins of omission (failing to protect us against certain legislative rights
abuses). For examples of these, see supra note 82. Here, I am referring to their sins of
commission: Sometimes the power of judicial review will be exercised tyrannically to prevent
legislatures from according people (what are in fact) their rights. For reference to some
examples, see supra note 4 and accompanying text.

148
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

minority129—i.e., the majority and minority groups whose rights are at stake in the
decision. In some cases the membership of the decisional majority may be the
same as the membership of the topical majority and the membership of the
decisional minority—those who voted against the injustice—may be the same as
the membership of the topical minority. This is often true in the case of racial
injustice for example: White legislators (decisional majority) vote for white
privilege (topical majority); black legislators lose out in the struggle for equal
rights for blacks. These are the cases, I submit, that we should be particularly
concerned about under the heading of “the tyranny of the majority.”
With this distinction in mind, let us return to cases of rights-disagreement.
Suppose that there is disagreement in a society about what the rights of a topical
minority are. Assuming this disagreement has to be settled, the society will have
to deliberate about it and apply its decision-procedures to the issue. Suppose the
society uses MD to settle this matter, I take part in this decisionmaking, using my
vote, and the side that I vote for loses. I am therefore a member of the decisional
minority on this issue. But so far it has not been shown that anything tyrannical
has happened to me. To show that we would have to show two additional things:
(1) that the decision really was wrong and tyrannical in its implications for the
rights of those affected; and (2) that I was a member of the topical minority whose
rights were adversely affected by this wrong decision.
The point to remember here is that nothing tyrannical happens to me merely
by virtue of the fact that my opinion is not acted upon by a community of which I
am a member. Provided that the opinion that is acted upon takes my interests
properly into account along with everyone else’s, the fact that my opinion did not
prevail is not itself a threat to my rights, or to my freedom, or my well-being. None
of this changes necessarily if I am also a member of the topical minority whose
rights are at issue. People—including members of topical minorities—do not
necessarily have the rights they think they have. They may be wrong about the
rights they have; the majority may be right. Responsible talk about “tyranny of
the majority” will keep these analytic points in mind.
To sum up, tyranny of the majority is possible. But the term should not be
used simply to mark the speaker’s disagreement with the outcome of a majority
decision. The most fruitful way of characterizing tyranny of the majority is to say

129. “Topical,” because their rights and interests are the topic of the decision. The term “topical
minority” is a loose one, and there is always likely to be dispute about whom it comprises (and
the same is true of “topical majority”). But the looseness is not a problem. Even loosely
defined, the distinction between topical and decisional minorities enables us to see that not
everyone who votes for the losing side in an issue about rights should be regarded as a member
of the group whose rights have been adversely affected by the decision. See WALDRON, supra
note 14, at 13-14; Waldron, Rights and Majorities, supra note 19, at 64-66; Waldron,
Precommitment and Disagreement, supra note 19.

149
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

that it happens in cases when topical minorities are aligned with decisional
minorities. In Part VII, I shall consider the application of this to what are called
“discrete and insular minorities.”130 For now, though, we may note that this sort of
alignment is exactly what we should not expect under the core assumptions we are
considering. Assumption three was that most people, and therefore most members
of any given decisional majority, care about rights just as much as the members of
a given decisional minority. And our fourth assumption about disagreement was
that disagreement is not usually driven by selfish interests. Disagreement is
sufficiently explained by the complexity and difficulty of the issues themselves.
What Rawls called “the burdens of judgment”131 argue precisely against the sort
of alignment between opinion and interests that, we have just seen, responsible
talk of the tyranny of the majority ought to presuppose.
The conclusion is not, however, that tyranny of the majority is something we
need not worry about. Rather, the conclusion is that tyranny of the majority—if
that term is being used responsibly—is a characteristic of non-core cases, in
which people care little for minority or individual rights other than their own. I do
not want to deny that this happens. But I think it is important to emphasize its
incompatibility with my third assumption and not to try to talk simultaneously
about a society committed to rights in which tyranny of the majority is
nevertheless an endemic possibility.
The distinctions made in this Part can help us deal with two other arguments
about judicial review. First, Ronald Dworkin argues in Freedom’s Law that
democratic decisionmaking is inherently tyrannical if people’s rights are not
respected. This is not just because it may generate tyrannical outcomes, he
argues, but because respect for rights is a background condition for the legitimacy
of any system of political decisionmaking. Dworkin is not just making the familiar
point that democracy depends (constitutively) on certain rights, like the right to
vote or, indirectly, the right to free speech or freedom of association. His point is
more sophisticated than that. He maintains that processes like MD have no
legitimacy at all in a democratic context (or any other context) unless each voter
is assured that the others already regard him with equal concern and respect. A
bunch of terrorists deciding my fate by majority decision (even an MD process in
which I am given a vote) has no legitimacy at all, because this background
condition is not met. In general, Dworkin argues, a person can hardly be expected
to accept majority decisions as legitimate if she knows that other members of the
community do not take her interests seriously or if the established institutions of
the community evince contempt or indifference toward her or her kind.132

130. See infra text accompanying notes 137-140.

131
RAWLS, POLITICAL LIBERALISM, supra note 53, at 54-8.
132. DWORKIN, supra note 10, at 25.

150
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

Dworkin thinks this refutes the democratic objection to judicial review.133

Suppose a piece of legislation is enacted by an elected assembly and then
challenged by a citizen on the ground that it undermines right R, a right that is a
condition of democratic legitimacy. We imagine that others will disagree, some
because they think R is not a condition of democracy, others because they
understand R in a quite different way. And suppose the issue is assigned to a
court for final decision, and the court strikes down the statute, accepting the
citizen’s challenge. Is there a loss to democracy? The answer, Dworkin says,
depends entirely on whether the court makes the right decision. If it does—that
is, if the statute really was incompatible with the rights required as conditions for
legitimate application of MD—then democracy is surely improved by what the
court has done, because the community is now more democratically legitimate
than it would have been if the statute had been allowed to stand.134
There are many things wrong with this argument, some of which I have
pointed out elsewhere.135 For one thing, Dworkin seems to be suggesting that if a
political decision is about democracy, then there is no interesting question to be
raised about the institutional process by which the decision is made. This seems
wrong to me. If a decision about the majoritarian process (or about the conditions
of its legitimacy) were made using some procedure that, for example, precluded
the participation of women, equality-based objections to that procedure would not
be disqualified simply because the legitimacy of the majoritarian process was
actually the matter at issue. We care about process-values even when process is
what is at stake in our disagreements.
But the most telling objection is this: Let us grant Dworkin’s premise—that
democratic procedures are legitimate only among people who respect one
another’s rights. That may be read in two ways: (1) democratic procedures are
legitimate only among people who hold and act upon the correct view of one
another’s rights; or (2) democratic procedures are legitimate only among people
who take one another’s rights seriously and who in good faith try as hard as they
can to figure what these rights are. The first reading is far too strong; no
imaginable political system satisfies it. And I cannot see any objection to the

133. Dworkin is careful to say that it is not an argument for judicial review. Id. at 7 (“Democracy
does not insist on judges having the last word, but it does not insist that they must not have
it.”).
134. Id. at 32-33; see also id. (adding that, of course, “if we assume that the court’s decision was
wrong, then none of this is true. Certainly it impairs democracy when an authoritative court
makes the wrong decision about what the democratic conditions require—but no more than it
does when a majoritarian legislature makes a wrong constitutional decision that is allowed to
stand. The possibility of error is symmetrical.”).
135. For a full response, see WALDRON, supra note 14, Ch. 13 [SC2: cite to page numbers, not
chapter]

151
 43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

second reading of Dworkin’s premise. But if we read it this way, then Dworkin’s
premise is satisfied for the sort of society we are considering in this Essay. Even if
people disagree about rights, they may take one another’s rights seriously.
Decisional majorities may prevail. Sometimes they will be right about rights and
sometimes they will be wrong. But that is something they have in common with all
systems of decisionmaking and that alone cannot undermine their legitimacy, so
long as topical minorities have an assurance that most of their fellow citizens take
the issue of their rights seriously.
Second, we can also use the distinctions developed in this Part to help deal
with the allegation that unreviewable legislative decisionmaking about rights
involves the majority being the judge in its own case. Those who invoke the
maxim nemo iudex in causa sua in this context say that it requires that a final
decision about rights should not be left in the hands of the people. Rather, it
should be passed on to an independent and impartial institution such as a court.
It is hard to see the force of this argument. Almost any conceivable decision-
rule will eventually involve someone deciding in his own case. Unless we
envisage a literally endless chain of appeals, there will always be some person or
institution whose decision is final. And of that person or institution, we can always
say that because it has the last word, its members are ipso facto ruling on the
acceptability of their own view. Facile invocations of nemo iudex in sua causa are
no excuse for forgetting the elementary logic of legitimacy: people disagree and
there is need for a final decision and a final decision-procedure.
What this second argument for the necessity of judicial review might mean is
the members of the topical majority—i.e., the majority whose rights and interests
are is at stake—should not be the ones whose votes are decisive in determining
whether those rights and interests are to remain ascendant. And there are
legitimate grounds for concern when topical majorities align with decisional
majorities. (If this alignment is endemic, then I think we are dealing with a non-
core case, for reasons I will explain in Part VII.) But it is striking how rarely this
happens, including how rarely it happens in the kinds of cases that are normally
dealt with by judicial review in the United States. Think of the two examples I
mentioned earlier: abortion and affirmative action. In neither case is there the sort
of alignment that might be worrying. Many women support abortion rights but so
do many men; and many women oppose them. Many African-Americans support
affirmative action, but so do many members of the white majority; and many
African-Americans oppose affirmative action. This is what we should expect in a
society in which our third and fourth assumptions, set out in Part II, are satisfied.
People who take rights seriously must be expected to disagree about them; but it
is a sign of their taking rights seriously that these disagreements will be relatively
independent of the personal stakes that individuals have in the matter.

152
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

vii. No n- co re cas es

The arguments I have made so far are based on four quite demanding
assumptions. What becomes of these arguments when the assumptions fail, or for
societies in which the assumptions do not hold? I have in mind particularly my
first assumption that a society has democratic and legislative institutions in good
shape so far as political equality is concerned, and my third assumption that the
members of the society we are considering are by and large committed to the idea
of individual and minority rights. For many people, I think the case for judicial
review rests on the refusal to accept these assumptions. Judicial review is in part
a response to perceived failures of democratic institutions, or it is in part a
response to the fact that many people do not take rights sufficiently seriously (so
they need a court to do it for them). In sum, supporters of the practice will say we
need judicial review of legislation in the real world, not the ideal world defined by
my assumptions.
A number of things need to be said in response to this, before turning to a
couple of specific issues about non-core cases. First, the assumptions on which I
have been proceeding are not utopian or demanding. Assumption three, for
example—a general commitment to rights in the society—is not unreasonable,
given that the case for judicial review almost always assumes that somehow the
society for which judicial review is envisaged has a Bill of Rights that stands in
some real relation to the views of citizens. The first assumption was about
electoral and legislative arrangements being in reasonably good shape, bearing in
mind that even in the name of political equality we are not entitled to demand
perfection. Also in Part V, when I talked about the legitimacy of legislatures and
courts, I again stressed that my argument did not turn on there being a perfect
response to individual citizens’ demand for voice and participation. The case for
the legitimacy of legislative decisionmaking does not depend on any assumption
of the utopian perfection of legislative institutions, nor on their perfectly
embodying the principle of political equality in their elective and procedural
aspects. It turns on these institutions being explicitly oriented to this principle,
organized in a way that is designed to satisfy the principle, and making a
reasonable effort to do so. Also, I took care to cite the actual deliberations of an
actual legislature—the House of Commons on the Medical Termination of
Pregnancy Bill 1966—as an example of how legislatures might work, not some
concoction of the philosophical a priori.
Having said all that, we still must ask: What happens to the argument against
judicial review if the assumptions fail?
In cases in which the assumptions fail, the argument against judicial review
presented in this Essay does not go through. As I emphasized in Part II, my

153
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

argument is a conditional one.136 However, it does not follow that judicial review
of legislation is defensible whenever the assumptions fail. There may be other
good arguments against judicial review that are not conditioned on assumptions
like mine. Or it may be the case that judicial review offers no hope of ameliorating
a particular situation. It may not be appropriate to set up judicial review of
legislation if judicial decisionmaking in a society is no less corrupt or no less
prejudiced than its legislative decisionmaking. The arguments we entertained for
the core case were in large part comparative, and this logic applies to non-core
cases as well.
Suppose we are dealing with a case that is non-core by virtue of the failure of
my first assumption: in this case, legislatures are inadequately representative or
deliberative, the system of elections is compromised, and the procedures used in
the legislature no longer bear any credible relation to political legitimacy. Two
questions then arise: (1) Is it possible to improve the situation, so far as the
legislature is concerned? (2) Should a final power of decision for important issues
of rights be vested in the courts, assuming that the courts would handle those
issues better? The questions are independent, for we may reasonably think that
some issues of rights are too urgent to await the emergence of a more responsible
and representative legislature. But they are not utterly independent. Vesting the
final power of decision in courts may well make it more difficult to reform the
legislature or more difficult to develop the legislative ethos that the first
assumption, and perhaps also the third assumption, presuppose. I have heard
speculation to this effect about the United States: The idea is that U.S.
legislatures, particularly state legislatures, operate irresponsibly and in a way that
fails to take rights seriously because the knowledge that the courts are there as
backup makes it harder to develop a responsible culture among legislators. How
far this is true, I don’t know. It is certainly worth considering.
I want to end by discussing one well known way in which my first assumption
might be thought to fail. I have in mind Justice Stone’s suggestion in the famous
Carolene Products footnote four: “[P]rejudice against discrete and insular
minorities may be a special condition, which tends seriously to curtail the
operation of those political processes ordinarily to be relied upon to protect
minorities.”137 This it seems to me is an excellent way of characterizing the sort of
non-core case in which the argument for judicial review of legislative decisions

136. See supra text accompanying note 43. For an example of the failure of the argument, see
Waldron, supra note 47.
137. United States v. Carolene Products Co., 304 U.S. 144, 152 n.4 (1938); see also Keith E.
Whittington, An “Indispensable Feature”? Constitutionalism and Judicial Review, 6 N.Y.U. J.
LEGIS. & PUB. POL’Y 21, 31 (2002-2003 [SC2: is this the appropriate date?]) (stating that my
neglect of this idea in Law and Disagreement is “striking from the perspective of American
constitutional theory”).

154
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

has some plausibility. Minorities in this situation may need special care that only
non-elective institution can provide—special care to protect their rights and
special care (as John Hart Ely points out) to repair the political system and
facilitate their representation.138
We have to be cautious about this argument, however. It follows from what I
said in Part VI that not every minority deserves this special treatment: certainly
not every decisional minority, and not even every topical minority.139 There is no
reason to suppose even that every chronic minority deserves this special
treatment, certainly not chronic decisional minorities—Bolsheviks in the United
States, for example.
Too often the phrase “discrete and insular” is used thoughtlessly. Not every
distinct and identifiable minority is discrete and insular. There is nothing magical
about Justice Stone’s language. But if taken seriously, “discrete” and “insular”
are useful adjectives, for they convey not just the idea of a minority that exists
apart from political decisionmaking—in other words a topical minority—but also
a minority whose members are isolated from the rest of the community in the
sense that they do not share many interests with non-members that would enable
them to build a series of coalitions to promote their interests. The alignment of
decisional and topical minorities that we warned against in Part VI is a good
example of “insularity” in this sense. And it is a cause for concern.
What about the other criterion that Justice Stone mentioned—that the
minority is the victim of prejudice? Pervasive prejudice is certainly incompatible
with my third and fourth assumptions; it connotes indifference or hostility to the
rights of the group’s members, and it may lead members of the majority to differ
unreasonably from the minority members’ estimation of their own rights. But the
term “prejudice” may be too narrow and its connotations may fail to capture the
depth of entrenched and unconscious antipathy between one group and
another.140 The point is not to insist on any particular mode of antipathy, but to
distinguish between its various modes and the phenomenon of reasonable
disagreement about rights.141

138. ELY, supra note 126, 135-79.

139. TUSHNET, supra note 11, at 159 (“Every law overrides the views of the minority that loses. . . .
We have to distinguish between mere losers and minorities that lose because they cannot
protect themselves in politics.”).
140. See Charles R. Lawrence, III, The Id, the Ego, and Equal Protection: Reckoning with
Unconscious Racism, 39 STAN. L. REV. 317 (1987). I am grateful to Ian Haney-Lopez for
emphasizing this point.

141
It is important also to distinguish between prejudices and views held strongly on
religious or ethical grounds. We should not regard the views of pro-life advocates as
prejudices simply because we do not share the religious convictions that support them.

155
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

In such cases, the core argument against judicial review that I have outlined
cannot be sustained. But, again, this is not the same as saying that a case has
been made in favor of judicial review. Everything depends on whether judicial
majorities are infected with the same prejudice as legislative majorities. If they
are, then the case may be not only non-core but hopeless. A practice of judicial
review cannot do anything for the rights of the minority if there is no support at all
in the society for minority rights. The affirmative case that is often made for
judicial review in these circumstances assumes that there is some respect for the
relevant minority’s rights outside the minority’s own membership, but that it is
largely confined to political elites. The idea is that most ordinary members of the
majority do not share this sympathy. Now the elite members who do share it—I
shall call them elite sympathizers—may be in the legislature or they may be in
the judiciary. The argument for giving final authority to judges is that elite
sympathizers in the judiciary are better able than elite sympathizers in an elected
legislature to protect themselves when they accord rights to the members of an
unpopular minority. They are less vulnerable to public anger and they need not
worry about retaliation. They are therefore more likely to protect the minority.
Notice how this argument for judicial review depends on a particular
assumption about the distribution of support for the minority’s rights. The
sympathy is assumed to be strongest among political elites. If that is false—if the
sympathy is stronger among ordinary people—then there is no reason to accept
the argument of the previous paragraph. On the contrary, elective institutions may
be better at protecting minority rights because electoral arrangements will provide
a way of channeling popular support for minority rights into the legislature,
whereas there are no such channels into the judiciary. No doubt, the distribution
of support for minority rights varies from case to case. But I find it interesting that
most defenders of judicial review, when they assume that there will be some
support for minority rights in a society, are convinced that in all cases it will be
found among elites if it is found anywhere. They will defend this as an empirical
claim, but I must say it is entirely consonant with ancient prejudices about
democratic decisionmaking.
One other factor to take into account is whether an established practice of
judicial review will make it easier or harder in the long-term to remedy the
elective and legislative dysfunctions we are imagining here. In certain
circumstances, discrete and insular minorities may benefit from judicial
intervention to protect their rights. But institutionally, judicial solicitude may
make things worse, or at least fail to make them much better. As the United States
found in the 1950s and 1960s, for all the excitement of judicial attacks on

Almost all views about rights – including pro-choice views -- are deeply felt and rest in
the final analysis on firm and deep-seated convictions of value.

156
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the core of the case against judicial review

segregation in Brown and other cases, what was needed in the end was strong
legislative intervention (in the form of the Civil Rights Act), and it turned out that
the main difference was not courts versus legislatures per se, but federal
institutions versus state institutions, with the federal legislature finally playing the
decisive role.
Overall, we should not read the Carolene Products footnote or any similar
doctrine as a way of “leveraging” a more general practice of judicial review into
existence.142 The problem of discrete and insular minorities is not to be seen as a
sort of Trojan Horse for judicial review or as a basis for embarrassing the
arguments against it. The aim of considering such cases is not to defend judicial
review; rather it is to do whatever best secures the rights of the minorities
affected. We should aim directly at that, conscious of the fact that there is no
convincing general argument for judicial review of which this could be treated as
a sort of ideological vanguard.

co nclus io n

I have not sought to show that the practice of judicial review of legislation is
inappropriate in all circumstances. Instead I have tried to show why rights-based
judicial review is inappropriate for reasonably democratic societies whose main
problem is not that their legislative institutions are dysfunctional but that their
members disagree about rights.
Disagreement about rights is not unreasonable, and people can disagree about
rights while still taking rights seriously. In these circumstances, they need to
adopt procedures for resolving their disagreements that respect the voices and
opinions of the persons—in their millions—whose rights are at stake in these
disagreements and treat them as equals in the process. At the same time, they
must ensure that these procedures address, in a responsible and deliberative
fashion, the tough and complex issues that rights-disagreements raise. Ordinary
legislative procedures can do this, I have argued, and an additional layer of final
review by courts adds little to the process except a rather insulting form of
disenfranchisement and a legalistic obfuscation of the moral issues at stake in our
disagreements about rights.
Maybe there are circumstances—peculiar pathologies, dysfunctional
legislative institutions, corrupt political cultures, legacies of racism and other
forms of endemic prejudice—in which these costs of obfuscation and
disenfranchisement are worth bearing for the time being. But defenders of judicial
review ought to start making their claims for the practice frankly on that basis—
and make it with a degree of humility and shame in regard to the circumstances

142. See TUSHNET, supra note 11, at [AU: please provide pincite].

157
43A1A8E4-5F61-084271.DOC 12/15/2005 9:33 AM

the yale law journal 115:100 2006

that elicit it—rather than preaching it abroad as the epitome of respect for rights
and as a normal and normatively desirable element of modern constitutional
democracy.

158