Nature of Judicial Review The

The Nature of Judicial Review*
Harry H. WeIfingtont
In The Nature of the JudicialProcess,' Cardozo is concerned with the

sources of common law: how do judges decide cases; what counts as justifi-
cation for decisions that rest uneasily, if at all, on precedent and that are
not required by a written text? Although evidently struck by the power
common-law judges exercise in such situations, Cardozo was not moved to
question the legitimacy of that power. Instead, he considered it sufficient
to explain and classify the various methods courts use in reaching deci-
sions. This predominantly descriptive approach assumes, among other
things, that judges ought to make the common-law, that is, ought to fash-
ion principles and policies into standards of decision. Cardozo's assump-
tion remains the conventional understanding: while we may disagree
strongly with particular decisions, we rarely question the authority of
common-law courts, even in pivotal cases.
Judicial review is another matter. Almost every important case that dis-
pleases some sizeable group leads to questions about the legitimacy of the
famous doctrine proclaimed in Marbury v. Madison.2 To speak only of
the recent past, it was some sixteen years after publication of The Nature
of the Judicial Process that judicial review of New Deal legislation
3
brought the executive and the Supreme Court into direct confrontation.
More recently, the segregation, 4 reapportionment,' and abortion cases'
have provoked very serious attacks on the Court. In addition, the attempt
to find a justification for judicial review is again engaging the attention of
* This Article was originally presented, in somewhat different form, as the Thirty-Sixth Annual
Benjamin N. Cardozo Lecture, delivered on March 31, 1981, to the Association of the Bar of the City
of New York. See 36 REC. A.B. CITY N.Y. 360 (1981).
t Dean and Edward J. Phelps Professor of Law, Yale Law School.
1. B. CARDOZO, THE NATURE OF THE JUDICIAL PROCESS (1921).
2. 5 U.S. (1 Cranch) 137 (1803).
3. See R. JACKSON, THE STRUGGLE FOR JUDICIAL SUPREMACY (1941); Lerner, The Great Con-
stitutional War, 18 VA. Q. REV. 530 (1942); Leuchtenburg, The Origins of Franklin D. Roosevelt's
"Court-Packing" Plan, 1966 SUP. CT. REV. 347.
4. Brown v. Board of Educ., 347 U.S. 483 (1954).
5. Reynolds v. Sims, 377 U.S. 533 (1964); Baker v. Carr, 369 U.S. 186 (1962).
6. Doe v. Bolton, 410 U.S. 179 (1973); Roe v. Wade, 410 U.S. 113 (1973).
HeinOnline -- 91 Yale L.J. 486 1981-1982

Judicial Review
several of our prominent constitutional scholars,' even as it did an earlier

generation some twenty years ago.
This difference in attitudes toward constitutional and common-law ad-
judication is fostered by a popular and influential view of our legal and
political landscape. We tend to think of courts at common law as acting
because the legislature has not and as making law the legislature can un-
make. When statutes are involved, we see our courts either as effectuating
legislative will or, through an occasional misreading of legislative intent,
as producing an incorrect decision that can be remedied easily by a legis-
lative reform. But in picturing judicial review, we imagine Justices ap-
pointed for life permanently thwarting the will of the people by striking
down the work of their elected representatives.
This general picture, although partially accurate, seriously distorts real-
ity. Statutory interpretation may be contrary to legislative intent and diffi-
cult for legislators to alter. Constitutional decisions, in contrast, may be
short-lived and involve no legislative judgment.9 Moreover, the
majoritarian prerogatives allegedly infringed by judicial review may, for
reasons unrelated to judicial power, be weaker than is commonly sup-
posed. Our governmental system-quite apart from its judicial branch-is
itself designed to temper and check the power of majorities. Furthermore,
interest groups of one sort or another and, of course, the press, not only
help to form majorities, but also act to slow and deflect the efforts of mi-
nority coalititions that have achieved majority status.
I will argue that when our practices are viewed in this light, the burden
of justification commonly imposed on judicial review appears unduly
heavy. The power to act in derogation of the immediate or apparent
wishes of the majority is peculiar neither to constitutional adjudication
nor, more generally, to the courts. But judicial review does stand out from
other countermajoritarian practices by virtue of the apparent finality of
constitutional decisions. It is thus the issue of finality, more than other
questions posed by majoritarian theory, that proponents of judicial review
must address. Finality raises at least two questions: One asks how final
constitutional decisions really are. The second wonders how transitory are
7. See, e.g., J. CHOPER, JUDICIAL REVIEW AND THE NATIONAL POLITICAL PROCESS (1980); J.
ELY, DEMOCRACY AND DISTRUST (1980); M. PERRY, THE CONSTITUTION, THE COURTS, AND
HUMAN RIGHTS: AN INQUIRY INTO THE LEGITIMACY OF CONSTITUTIONAL POLICY MAKING BY THE
JUDICIARY (forthcoming).
8. See, e.g., A. BICKEL, THE LEAST DANGEROUS BRANCH (1962); C. BLACK, THE PEOPLE AND
THE COURT (1960); Wechsler, Toward Neutral Principlesof ConstitutionalLaw, 73 HARV. L. REV.
1 (1959). See also Deutsch, Neutrality, Legitimacy, and the Supreme Court: Some IntersectionsBe-
tween Law and Political Science, 20 STAN. L. REV. 169 (1968).
9. See generally C. BLACK, STRUCTURE AND RELATIONSHIP IN CONSTITUTIONAL LAW 77-78, 89-
90 (1969) (nonlegislative actions, such as those of law enforcement officers, also give rise to constitu-
tional decisions).

The Yale Law Journal Vol. 91: 486, 1982
other nonmajoritarian decisions of consequence. Each of these issues,

moreover, inevitably brings one to the questions regarding the sources of
law that Cardozo addressed in The Nature of the Judicial Process.
2
Majoritarianism, taken by itself, may be thought to require that each
citizen be granted one equal vote, and that simple majorities control the
course of government.10 Although adopted in more or less this form by the
reapportionment cases," this simple theory of political fairness1 2 is neither
established by nor easily reconciled with the Constitution. That document
forbids the direct popular electioni of the President; it does not forbid ger-
rymanders. 13 And by direct constitutional command, Senators chosen by
electorates of varying size wield equal voting power.14 Moreover, the Con-
stitution protects the important advantage of wealth in the political pro-
cess: the First Amendment forbids significant governmental restraints on
individuals who have the means and the desire to finance the widespread
expression of their political opinions.1"
When we move from the deep structure of the Constitution to well-
established legislative practice we find-and it is hardly a surprise-a fur-
ther dilution of majority rule. Thus, although recently diminished, senior-
ity prerogatives remain considerable, 6 and committee chairmen exercise
substantial power." Congressmen are dependent on their staffs, and con-
10. See J. RAWLS, A THEORY OF JUSTICE 221-28 (1971).

11. See Reynolds v. Sims, 377 U.S. 533 (1964).
12. Although simple in its requirements, this theory may be impossible to implement:
[I]f we expect majority rule to produce what a majority really wants, then democracy cannot
work flawlessly even in theory. Voting theorists. . . have known that the outcome of majority
voting, even in a direct democracy with no opportunity to form coalitions, may be merely the
artifact of the placement of issues on the public agenda.
Mashaw, ConstitutionalDeregulation:Notes Toward a Public, PublicLaw, 54 TULANE L. REV. 849,
858 (citing Levine & Plott, Agenda Influence and its Implications, 63 VA. L. REV. 561 (1977)). See
also B. ACKERMAN, SOCIAL JUSTICE IN THE LIBERAL STATE 289-93 (1980) (effect of agenda manipu-
lation on system of majority rule).
13. U.S. CONST. art. II, § 1. See Gaffney v. Cummings, 412 U.S. 735, 751-54 (1973).
14. U.S. CONST. amend. XVII.
15. See Buckley v. Valeo, 424 U.S. 1 (1976) (limitations on "independent" expenditures held
unconstitutional). The circuit court decision that Bucdey reversed in part had upheld independent
expenditure limitations on the theory that they implemented the "one-man, one-vote" principle estab-
lished in the reapportionment cases. See Bucldey v. Valeo, 519 F.2d 821, 841 (D.C. Cir. 1975); see
also Common Cause v. Schmitt, 50 U.S.L.W. 4161, 4168 (U.S. Jan. 19, 1982) (per curiam) (udg-
ment affirmed by a divided Court) (limitation on political expenditures of independent groups held
unconstitutional).
16. "What the congressional seniority system does as a system is to convert turf into property...
[a]nd the property automatically appreciates in value over time." D. MAYHEW, CONGRESS: THE ELEC-
TORAL CONNECTION 95-96 (1974); see also R. DAVIDSON & W. OLESZEK, CONGRESS AGAINST ITSELF
272 (1977); Polsby, Gallaher & Spencer, Growth of the Seniority System in the US. House of Repre-
sentatives, in CONGRESSIONAL BEHAVIOR 172-202 (N. Polsby ed. 1971).
17. See J. CHOPER, supra note 7, at 19-21, 38-42; R. DAVIDSON & W. OLESZEK, supra note 16,
at 49; B. HINCKLEY, THE SENIORITY SYSTEM IN CONGRESS 92-93 (1971).

Judicial Review
gressional staffs advance their own agenda.18 The press,1 and lobbyists
and lawyers representing powerful corporate and union clients, exert ex-
traordinary influence on the legislature and, of course, on other branches
of government-elected and appointed.20
The standard practices of government departments and agencies diverge
still further from simple majoritarian precepts. Extensive job security and
sheer numerical weight invest bureaucrats with considerable control over
policy. 21 Indeed, the bureaucracy is often entrenched firmly enough to ex-
ercise its power in the teeth of popular mandates. 22 Nor is it easy for the
people's representatives to revoke the broad delegations of power that
make such entrenchment possible. The needs that initially prompted the
establishment of an agency or bureau may still exist, or, indeed, have
grown increasingly acute. In such circumstances, the legislature's depen-
dence on the agency will force it to acquiesce in at least certain unpopular
agency initiatives. 23 Moreover, even when the legislature is willing to
abolish a particular agency or bureau, private groups that have relied on
its policies or existence may fiercely oppose the revocation of delegated
power. 4 It is the rare case in which neither a sizeable group of legislators
18. See M. MALBIN, UNELECTED REPRESENTATIVES: CONGRESSIONAL STAFF AND THE FUTURE
OF REPRESENTATIVE GOVERNMENTS (1980).
19. See, e.g., D. CATER, THE FOURTH BRANCH OF GOVERNMENT (1959) (discussing power of the
press); Robinson, Television and American Politics: 1956-1976, 48 PUB. INT. 3, 11-14 (1977) (dis-
cussing influence of television news coverage on political opinions).
20. As one commentator notes:
[Ilt is generally agreed that by transmitting pertinent information to key lawmakers, by skill-
fully and selectively applying pressure at critical points in the system, and by expending mas-
sive sums of money-not infrequently in an abusive, occasionally criminal, manner-they
[lobbyists] are able to exercise power well beyond the force of the numbers of people they
represent.
J. CHOPER, supra note 7, at 23. See also R. DAHL, DEMOCRACY IN THE UNITED STATES 454 (3d ed.
1976); M. MAYER, THE LAWYERS 51-52 (1966); D. TRUMAN, THE GOVERNMENTAL PROCESS 352-94
(2d ed. 1971).
21. See W. NISKANEN, BUREAUCRACY AND REPRESENTATIVE GOVERNMENT 138-54 (1971) (bu-
reaucracy acts contrary to popular will). The point is well made by Hodding Carter with respect to
the Department of State. See Carter,The Unequal BureaucraticContest, Wall St. J., Jan. 8, 1981, at
21, col. 3; see also A. DOWNS, INSIDE BUREAUCRACY 134-35, 152-53 (1967) (difficulty of controlling
bureaucracy); Cutler & Johnson, Regulation and the Political Process, 84 YALE L.J. 1395, 1408-09
(1975) (ability of politicians to influence agency policy decreases as agency ages and presumption of
independence and expertise grows).
22. See W. NISKANEN, supra note 21, at 138-54; Cutler & Johnson, supra note 21, at 1401 n.21
(Federal Reserve continues to implement monetary policies despite attacks); c. Nathan, The "Admin-
istrative Presidency," 44 PUB. INT. 40 (1976) (controlling bureaucracy is one of most difficult jobs
facing President).
23. With respect to many regulatory endeavors, "it may be impossible in the nature of the subject
matter to specify with particularity the course to be followed." Stewart, The Reformation of American
Administrative Law, 88 HARV. L. REV. 1667, 1695 (1975). In addition, Congress' own institutional
shortcomings compel it to accept some agency discretion as an inevitable cost of effecting its regulatory
goals. Id. at 1695-96.
24. Thus, for example, businesses typically support regulations that insulate them from competi-
tion in the marketplace. See R. NOLL, N. PECK, & J. MCGOWAN, ECONOMIC ASPECTS OF TELEVI-
SION REGULATION 205-07 (1973); Wilson, The Rise of the BureaucraticState, 41 PUB. INT. 77, 98-99

nor a powerful private group has a large stake in the survival of a govern-
ment bureau. Therefore, it is common for individuals who are neither
elected nor recently appointed by elected officials to direct or influence
significantly the course of government.
It is possible to view these countermajoritarian forces in the Constitu-
tion, legislatures, the bureaucracy, and private associations as flaws in our
democratic system, and to characterize government as a failed attempt to
achieve the simple majoritarian ideal. "5 But such a view seems mistaken.
While countermajoritarian forces impede the immediate implementation
of majority preferences and sometimes result in their reformulation, the
delay and reformulation often is a prudent constraint on governmental
action.
Such forced incrementalism provides several important benefits: First,
impediments to the instant gratification of majorities allow proposals for
legislative change to be considered carefully. This may be desirable even
though the creation of a particular majority itself has been time-consum-
ing. For it is surely the case that, in creating a majority for a proposal,
proponents may fail to examine fully the proposal's demerits.2 6 Second,
reliance on extended debate allows the political system to gauge the inten-
sity with which a position is advanced, rather than merely the number of
people who advance it.27 Third, while it is easy to belittle an earlier gen-
eration's exaggerated faith in administrative expertise, it is also easy to
forget that the influence of the bureaucracy in departments and agencies
does allow dedicated administrators with on-site experience to make con- 28
structive and often unpopular contributions to government policy.
Fourth, the deliberate allocation of power to groups not dominated by a
present coalition of fifty-one percent insures both that no one "faction"
will acquire too much power, and that affirmative governmental decisions
(1975). They also favor regulations that simply impose additional costs; having invested heavily to
comply with governmental requirements, businesses often fear the comparative cost-advantage deregu-
lation would afford future entrants into their markets. See N.Y. Times, Aug. 2, 1979, at A14, col. 1
(GM, after investing to meet fuel-efficiency standards, supports them).
25. See infra note 68. But see Shapiro, JudicialModesty, Political Reality, and Preferred Posi-
tion, 47 CORNELL L.Q. 175 (1962).
26. See Bickel, supra note 8, at 70 (by refusing to act through jurisdictional and other techniques,
the Court seeks to "elicit the correct answers to certain prudential questions" from the political pro-
cess); Bickel & Wellington, Legislative Purposeand the Judcial Process: The Lincoln Mills Case, 71
HARV. L. REV. 1, 27 (1957) (Congress, in responding to popular will, may unknowingly sacrifice
long-range values for immediate results; such values are protected through judicial review).
27. See Kendall & Carey, The "Intensity" Problem and Democratic Theory, 62 AM. POL. SC.
REV. 5 (1968). See also A. BICKEL, REFORM AND CONTINUITY 16, 17 (1971) (American political
system recognizes that opinions, preferences and interests vary in intensity); R. MUSGRAVE & P.
MUSGRAVE, PUBLIC FINANCE IN THEORY AND PRACTICE 113-15 (2d ed. 1976) (intensity of preference
registered through log-rolling and coalition formation).
28. A. BURNS, REFLECTIONS OF AN ECONOMIC POLICY-MAKER 418-22 (1978) (discussing ability
of Federal Reserve to advance unpopular policies).

Judicial Review
will rest on a broader and more stable base than simple majorities can
provide.29 Finally, by promoting stability, countermajoritarian practices
protect from abrupt defeat expectations invited by existing arrangements."
The majoritarian theory adopted by the reapportionment decisions does
not override these justifications of countermajoritarian practices. For the
reapportionment decisions have had no effect on countervailing doctrines
that either explicitly or implicitly established the constitutional importance
of incremental change. The vagueness and delegation doctrines, for exam-
ple, have been described as requiring legislatures to engage in careful de-
liberation before they substantially alter the legal landscape." When so
employed, these doctrines impose upon law-makers duties of care that are
neither mandated by, nor in many circumstances consistent with, a simple
majoritarian scheme.
Other doctrines (also unaffected by the reapportionment decisions) do
not directly foster, but instead presuppose incremental change, and thus
provide constitutional support for those countermajoritarian practices that
make incremental change possible. In particular, the efficacy of doctrines
that emphasize the importance of an individual's "reliance interest" de-
pends substantially on the political stability that our countermajoritarian
practices foster. Such doctrines have been developed, among other areas,
under the due process, takings, and contract clauses. For example,
whether an interest rises to the level of "property" for purposes of the
due process clause depends to a considerable extent on whether the rele-
vant authorities have led the possessor of that interest to rely on its secur-
ity and inviolability." Similarly, a takings or contract clause claim may be
seriously weakened by a showing that the possessor of the interest at issue
had always been aware of the existence of a superseding regulatory mech-
anism.3 Indeed, one commentator has noted that the contracts clause, al-
29. See A. BICKEL, supra note 27, at 16-19; R. DAHL, supra note 20, at 53-59 (American politics
not fully democratic; partitioning ecourages incremental changes by creating strongholds from which
minorities can block policy changes); Black, NationalLawmaking by Initiative?Let's Think Twice, 8
HUM. RTS. 28, 30-31 (1979) (legislative process contains structural and procedural safeguards that
protect minority interests and promote compromise). On incremental change generally, see D.
BRAYBROOKE & C. LINDBLOM, A STRATEGY OF DECISION (1963).
30. See G. CALABRESI, A COMMON LAW FOR THE AGE OF STATUTES 3 (1982).
31. See id. at 18.
32. See, e.g., Perry v. Sindermann, 408 U.S. 593 (1972) (whether state provided de facto support
for complainant's reliance is relevant to status of complainant's interest under due process clause);
Board of Regents v. Roth, 408 U.S. 564 (1972) (property interest not created by unilateral or merely
subjective expectation).
33. With respect to the contract clause, compare Allied Structural Steel Co. v. Spannaus, 438 U.S.
234, 249-50 (1978) (law impairing private pension agreement void because, among other things, it
imposed an unanticipated retroactive obligation in field that legislature had not previously regulated)
with Veix v. Sixth Ward Bldg. & Loan Ass'n, 310 U.S. 32, 38 (1940) (fact that complainant had
"purchased into an enterprise already regulated in the particular to which he now objects" relevant to
constitutionality of impairment).
With respect to the takings clause, compare Kaiser Aetna v. United States, 444 U.S. 164 (1979)

though originally of broader scope, has been narrowed to bar only the
disruption of reasonable expectations.34
The instability fostered by a government that instantly gratified majori-
ties would slow or halt the growth of reasonable expectations."5 By thus
destroying the foundations of reliance, such an overridingly majoritarian
scheme would eventually deprive the property and contract clauses of
what they are currently understood to protect.
Indirect support for the countermajoritarian value of continuity is also
provided by the broad constitutional prohibition of ex post facto laws.3"
That proscription is intended to allow reliance upon the particular rights
and duties established by the legislature.37 Such security and repose as the
proscription seeks to protect, however, could never develop in the absence
of mechanisms that often impede short-run, direct accountability.
These features of our social and legal organization reinforce the pro-
position that the bureaucracy, the congressional staff, the special pleaders,
and others cannot be condemned because of their countermajoritarian
character. To the contrary, it is precisely by exercising their limited free-
dom from the majority that these political forces advance values important
to our complex constitutional design.
3
Although not worrisome simply because they are countermajoritarian,
at least some of the political forces discussed above are worrisome indeed.
The perspective of minority power holders is often constrained by the
needs of their constituents. Yet those constituents are not themselves disin-
terested: where they seek to check the efforts of the majority, they do so
out of self interest. Accordingly, these countermajoritarian forces cannot
be relied upon merely to restrain majority willfulness. Instead, if given the
opportunity, such forces may hold hostage to minority demands even those
(government's attempt to compel free public use of corporation's assets went so far beyond ordinary
regulation as to interfere unconstitutionally with corporation's reasonable investment-backed expecta-
tions) with Penn Cent. Transp. Co. v. New York City, 438 U.S. 104, 124, 135-36 (1978) (relevant to
landmark preservation law's validity that it did not interfere with "distinct investment-backed expec-
tions") and City of Pittsburgh v. Alco Parking Corp., 417 U.S. 369 (1974) (incremental increase in
preexisting gross-receipts tax not taking even where tax, in combination with government-subsidized
competition, renders business unprofitable).
34. See J. ELY, supra note 7, at 92.
35. See G. CALABRESI, supra note 30, at 3 ("If legal-political institutions are too responsive to
change, however, temporary and unstable majorities are apt quickly to impose their will. New laws
are passed only to be followed by quick reversal at the next election, leading to uncertainty and to the
defeat of legitimate expectations.")
36. U.S. CONST. art. I, § 10.
37. See Weaver v. Graham, 101 S. Ct. 960, 964 (1981) (Framers sought through prohibition "to
assure that legislative acts give fair warning of their effect and permit individuals to rely on their
meaning until explicitly changed").

Judicial Review
many majoritarian proposals that are carefully fashioned, incremental so-

lutions to important problems."
Where countermajoritarian institutions do not overreach, but merely
moderate majority claims, personal interest or a narrowness of vision still
determines and confines their ability to contribute to the process of gov-
ernment. Because they often deflect the momentary passions of the major-
ity, countermajoritarian political forces may well provide protection for
longer range concerns in politics. Their interest in, or perspective on, par-
ticular results, however, necessarily renders such protection episodic and
often erratic. Accordingly, to ensure the considered and careful protection
of those principles that constitute our political inheritance, we require a
distinctive political force.
The special attributes of courts-the fact that judges are disinterested
generalists 39-enables them to serve this distinctive function better than
other political institutions. Judges-and especially Article III judges-are
deliberately removed from the pressures to which many other governmen-
tal actors are deliberately exposed.4 Their relative insulation from the
direct claims of special interest constituencies protects judges from the par-
tisan views of other political actors. At the same time, the generality of
their jurisdiction enables judges to discern, better than others with narrow
jurisdiction and, therefore, limited vision, the enduring principles and
longer range concerns that tend to be forgotten where either the interests
of factions collide4 or the perspectives of bureaucrats prevail.
Indeed, it is the special role of courts, when confronted with a bureau-
cratic distortion or, more importantly, a sharp clash of interests, to ex-
amine the views the community expressed in calmer moments, and to in-
fer from those expressions that recur principles of general application.42
Thus, while other countermajoritarian institutions out of parochial inter-
ests slow the rate of change, the courts alone deliberately search the past
for elements worth preserving. By applying those elements to the often
heated controversies before them-thereby reminding the polity of the val-
38. This problem is discussed in Calabresi's splendid book, see supra note 30, at 6, 48-49, 124-29.
It is worth noting that Calabresi's solution is carefully tailored to preserve the balance between con-
tinuity and change that has distinguished American common law. Id. at 118-19.
39. See Fiss, The Supreme Court, 1978 Term-Foreword: The Forms of Justice, 93 HARV. L.
REV. 1, 12-13 (1979). By disinterested, of course, I do not mean uninterested. I do mean one who is
not influenced by selfish motives or personal advantage. And by generalist, I do not necessarily mean a
Renaissance Person. I certainly do not mean an amateur. I do mean one who is engaged with and
knowledgeable about the law, conceived broadly.
40. See id. at 1-12; Wellington, Common Law Rules and ConstitutionalDouble Standards:Some
Notes on Adjudication, 83 YALE L.J. 221, 248-49 (1973).
41. See A. BICKEL, THE SUPREME COURT AND THE IDEA OF PROGRESS 86-87 (1970); Wellington,
supra note 40, at 246-48.
42. This point is elaborated in Wellington, supra note 40, at 246-48.

ues to which it has long adhered 4 3-the courts may be seen as the key
political institution charged with taking account of our public traditions.
The costs of uniting present and past should be acknowledged: our
predecessors too were imperfect. Nevertheless, a governmental structure
that fails to unite a nation's present with its past necessarily fails to pre-
serve values to which its citizens may attach considerable weight. It fails
to make a contemporary effort to understand what we have been or have
wished as a people to become, and thus it fails to give effect to what might
be called the moral ideals of the community. Those ideals cannot be un-
derstood by the bureaucracy, the special pleaders, and the congressional
staffers. Theirs is a tunnel vision, and the tunnel vision of one is not offset
by that of the others. Nor would these ideals be given adequate voice in a
simple majoritarian government where the passionate and self-interested
concerns of the moment were too easily accorded sovereignty. 4
Many may have trouble with the view that courts should seek to dis-
cover and use the moral ideals of the community as a source of legal prin-
ciples.4 5 But it is less controversial, and for present purposes sufficient, to
note that courts do in fact seek to preserve principles that are threatened
by majority preferences, and that they do so even in cases where the aban-
donment of those principles would not raise constitutional questions.
The most important method courts use in this task is statutory interpre-
tation. It is standard and appropriate for courts to employ general legal
principles to construe the open texture of statutes.46 This is not necessarily
countermajoritarian, although sometimes statutory purpose and general
legal principles pull in different directions.47 Sometimes interpretation is
clearly countermajoritarian as the following three cases show. In none am
I concerned with whether the particular decision is correct. The cases are
used to demonstrate an approach, a judicial function, that contributes to,
and is inherent in, the judicial process.48
Roto-Lith v. Bartlett," a contract case, provides an example of this ju-
dicial function. The common law, concerned with the principle of volunta-
rism, had long required assent as a necessary condition of binding con-
43. Id. at 245.

44. See infra note 133.
45. For a brief defense of this concept, see infra note 133; for its elaboration, see Wellington,
supra note 40, at 243-54, 272-311.
46. See Republic Aviation Corp. v. NLRB, 324 U.S. 793 (1945).
47. See NLRB v. Babcock & Wilcox Co., 351 U.S. 105 (1956).
48. This judicial approach has been both praised, see H. Hart & A. Sacks, The Legal Process
1240 (tent. ed. 1958) (unpublished manuscript on file with Yale Law Library), and criticized, see
Pound, "Common-Law and Legislation," 21 HARV. L. REV. 383 (1908). Although occasionally
abused through judicial overreaching, c G. CALABRESI, supra note 30, at 24 (criticism of "strict
construction of statutes in derogation of the common law" provoked by judicial abuse), the approach
has over the generality of cases provided a unique and important check on legislative action.
49. 297 F.2d 497 (1st Cir. 1962).

Judicial Review
tracts-absent a precise acceptance by the offeree of each aspect of the

offer, a contract would not be formed."0 The Uniform Commercial Code
broke with this position. In particular, Section 2-207 of the Code provided
that a binding agreement could be found even if a response to the offer
stated additional or different terms.' The Code's innovation had the effect
of requiring offerees wishing to alter the terms of an offer to make such
alterations express conditions of acceptance.12 If an offeree did not do this,
and if the transaction was between "merchants," the Code undertook to
sever those disputed terms that "materially altered" the contract, and to
find an acceptance of the remainder of the offer.
The Roto-Lith court was unwilling to countenance such a divergence
from the common law:
The statute is not too happily drafted. Perhaps it would be wiser in

all cases for an offeree to say in so many words, "I will not accept
your offer until you assent to the following: . . . ." But businessmen
cannot be expected to act by rubric. It would be unrealistic to sup-
pose that when an offeree replies setting out conditions that would
be burdensome only to the offeror he intended to make an uncondi-
tional acceptance of the original offer, leaving it simply to the of-
feror's good nature whether he would assume the additional restric-
tions. To give the statute a practical construction we must hold that
a response which states a condition materially altering the obligation
solely to the disadvantage of the offeror is an "acceptance . . .ex-
pressly . . . conditional on assent to the additional . . . terms.""3
By misconstruing the Code in this manner, 4 the court in Roto-Lith de-

flected the majoritarian attempt to compromise (in this type of situation)
the underlying and central common-law principle that emphasizes the
need for voluntary acceptance of contractual burdens.5
50. See Iselin v. United States, 271 U.S. 136, 139 (1926); Prescott v. Jones, 69 N.H. 305, 305-6,
41 A. 352, 352-53 (1898); More v. New York Bowery Fire Ins. Co., 130 N.Y. 537, 547, 29 N.E. 757,
759 (1892). For an interesting historical perspective on voluntarism and competing concepts in con-
tracts, see Gordley, Equality in Exchange, 69 CALIF. L. REV. 1587 (1981).
51. See U.C.C. § 2-207 (1972).
52. Id.
53. 297 F.2d 497, 500 (1st. Cir. 1962).
54. See, e.g., Comment, Commercial Law-Offer and Acceptance-Under Uniform Commercial
Code Purchaserof Goods Is Bound by Disclaimerof WarrantiesContainedin Seller's Expression of
Acceptance, 76 HARV. L. REV. 1481, 1482-83 (1963) (criticizing Roto-Lith); Comment, Nonconform-
ing Acceptances Under Section 2-207 of the Uniform Commercial Code: An End to the Battle of the
Forms, 30 U. CHI. L. REV. 540, 553-54 (1963) (same).
55. The Supreme Court recently applied a very similar principle to protect states from the inad-
vertent assumption of burdens stemming from "contracts" or agreements executed by the federal gov-
ernment pursuant to its spending power. See Pennhurst State School & Hosp. v. Halderman, 451
U.S. 1 (1981). The Court narrowly construed a federal "grant-in-aid" statute to limit the obligation it
imposed on states as a condition of the receipt of federal funds. The Court did not prospectively forbid
the imposition of substantial obligations on states, but merely required, in accordance with the com-
495

Textile Workers Union v. DarlingtonManufacturing Co." is a differ-

ent sort of case, but also one in which the judiciary thwarted majoritarian
goals in the service of what it took to be an important general principle.
In that case the Supreme Court held, among other things, "that so far as
the Labor Relations Act is concerned, an employer has the absolute right
to terminate his entire business for any reason he pleases . . . ."I The
proposition that had been advanced by the Textile Workers was "that an
employer may not go completely out of business without running afoul of
the . . . Act if such action is prompted by a desire to avoid unioniza-
tion." ' To this the Court replied that the union's position "would re-
present such a startling innovation that it should not be entertained with-
out the clearest manifestation of legislative intent . . .,,9
Of course the Act dearly does make it an unfair labor practice for an
employer to discharge its employees because they join a union." And, of
course, the union's position was that for an employer to go out of business
because its employees joined a union is tantamount to a discharge. Surely
this is correct. Nor was the question of remedy troublesome: while rein-
statement with back pay is standard, in this case the order below was that
the employer pay its employees "until they obtain substantially equivalent
work . -.6.
Accordingly, it seems that the purpose of the statute supported the
union's claim and, one would suppose, the statute's purpose reflected
majoritarian will in the same way that any non-anachronistic enactment
does. But there is an issue of economic freedom, at least at a symbolic
level: one-even if the one is a major corporation that generally is referred
to by the pronoun "it"-should be free to be or not to be. It is this princi-
ple that the Court employed to deflect the indicated majoritarian result.
Observe, moreover, that this principle has no constitutional standing: the
mon-law principle advanced in Roto-Lith, that those obligations be stated clearly and specifically by
Congress. Id. at 22-27. The Court also instructed Congress to make unmistakably plain its intent to
rely on Section 5 of the Fourteenth Amendment. Id. at 15-16. This instruction makes more difficult
the enactment of legislation that coerces state behavior; it also insures that states receive notice of the
federal requirements with which they must comply.
Neither this discussion of Pennhurstnor my use of any other case is intended to imply agreement
with substantive results. While endorsing the judicial technique that moderates majorities without
permanently foreclosing legislative choices, I would readily concede that this technique, like every
other, can be abused.
56. 380 U.S. 263 (1965).
57. Id. at 268.
58. Id. at 269-70.
59. Id. at 270.
60. "It is no longer disputed that workers cannot be dismissed from employment because of their
union affiliations." Phelps Dodge Corp. v. NLRB, 313 U.S. 177, 183 (1941). See J. GETMAN, LABOR
RELATIONS 113-14 (1978).
61. 380 U.S. 263, 267 (1965).
496

Judicial Review
state of economic due process today makes it certain that "the clearest
manifestation of [contrary] legislative intent" would succeed.
Vincent v. Pabst 6 2 provides a more complicated
example of counter-
majoritarian judicial revision and reveals a different aspect of the judicial
role. That case involved a Wisconsin statute that denied recovery to a tort
plaintiff whose own negligence was equal to or greater than the negli-
gence of the defendant."3 Its retention of contributory negligence put the
statute at odds with the common law, which, by the time Vincent was
decided, was moving rapidly toward full comparative negligence.6" Thus,
the circumstances underlying Vincent were both different from and simi-
lar to those from which Roto-Lith and Darlington arose: in each case
legislation, for better or worse, compromised principles of importance in
the relevant jurisprudence. In Vincent, however, the common law had
overtaken the legislation. The principle involved was relatively new in
terms of its legal status. Unlike the courts in Roto-Lith and Darlington,
the Vincent court was not acting as a guardian of the past.
The Vincent court, moreover, did not directly vitiate and rewrite the
Wisconsin statute to conform to the principle of the common law. Perhaps
because it was aware that the legislature had considered and rejected com-
parative negligence, members of the court chose the less intrusive, though
still countermajoritarian, tactic of threatening to rewrite if the legislature
did not act itself. 5 In order to make that threat credible, they had first to
establish that the court possessed the power to apply the common law."6
They accomplished this through a deliberate and plain misconstruction. In
particular, several Justices held that the statute covered only those cases in
which the plaintiffs negligence was less than the defendant's; as to all
other cases, these Justices found the statute to be silent. Tailored to per-
mit a consistent application of the common law, this misconstruction con-
stituted not direct, but "prospective" countermajoritarian revision.
The insight to be taken from this brief excursion into non-constitutional
adjudication is that even when no constitutional question is raised, the
courts, like other imperfectly representative institutions, may sometimes
blunt the efforts of majorities. Unlike those other institutions, however,
the courts check majorities in a selective and deliberate fashion. In partic-
ular, they seek to preserve principles that, from their generalist perspec-
tive, are, for one reason or another, significant. 7
62. 47 Wis. 2d 120, 177 N.W.2d 513 (1970). My knowledge of the case, and my analysis of it,
stem from G. CALABRESI, supra note 30, at 36-43, 210 nn.16-29.
63. WIS. STAT. ANN. § 895.045 (West 1966).
64. See G. CALABRESI, supra note 30, at 210 nn.18-19.
65. Id. at 36.
66. Id. at 36-37.
67. See Bickel & Wellington, supra note 26, at 27.

4
The foregoing discussion provides reasons for challenging the standard
critique of judicial review. That critique supposes that from our practice
in the normal course, adherence to the principle of simple majority control
can be inferred. According to the critique, divergences from this principle
are not deliberate; instead, no matter how frequently encountered, they
are dismissed as inadvertent failures to achieve the majoritarian ideal. 8 As
I have suggested, however, few of our practices are overridingly
majoritarian. To the contrary, wherever our system creates a danger of
majority willfulness, some tempering device is interposed. Accordingly, the
critics of judicial review should not ask us to assume that any failure to
support the preferences of majorities requires special or extraordinary jus-
tification. Indeed, it may be that the critics should bear the burden of
persuasion. After all, it is they who question a one hundred and fifty-year
old precedent that fits comfortably within our complex governmental
structure and that makes a distinctive contribution to that structure.
Of course, the critics may respond that other countermajoritarian insti-
tutions do not exercise power after the enactment of legislation. Thus, one
commentator has recently suggested that "[t]he existing antimajoritarian
influences in . . . legislatures, capable though they may be in blocking
legislation, are not well situated to get legislation passed in the face of
majority opposition."'" This commentator does recognize that "[tihere may
• . . exist situations in which a majority cannot pass a law repealing old
legislation because of minority resistance." 70 And I am sure he knows that
countermajoritarian institutions do, apart from judicial review, exercise
power after the fact. Statutory interpretation by agencies, departments,
and courts, for example, often departs from legislative intent. But more
importantly, it is not so much the timing, as it is the nature, of a declara-
tion of unconstitutionality that is significant. Assume that the Court, in
order to make the short-run effects of judicial review more like "[t]he ex-
isting antimajoritarian influences in legislatures," issued binding constitu-
tional opinions only before the legislature acted. Would it matter very
68. See A. BICKEL, supra note 8, at 18 ("[I]mpurities and imperfections, if such they be, in one
part of the system are no argument for total departure from the desired norm in another part."); J.
CHOPER, supra note 7, at 57-59 (that representative institutions are imperfectly majoritarian does not
affect legitimacy of judicial review); R. DWORKIN, TAKING RIGHTS SERIOUSLY 141 (1977) (that legis-
latures are imperfectly majoritarian "does not so much undermine the argument from democracy as
call for more democracy"). See also J. ELY, supra note 7, at 67. But see Hazard, The Supreme Court
as a Legislature, 64 CORNELL L. REV. 1, 10-11 (1978) (arguing that democratic principles tolerate
allocation of some power to unrepresentative institutions); Rostow, The DemocraticCharacterof Ju-
dicial Review, 66 HARV. L. REV. 193, 197-200 (1952) (noting similarity between judicial review and
other well-established countermajoritarian practices).
69. See J. ELY, supra note 7, at 67 (citing Choper, The Supreme Court and the Political
Branches: Democratic Theory and Practice, 122 U. PA. L. REV. 810, 830-32 (1974)).
70. Id.

Judicial Review
much, for these purposes?" I think not because I believe that the real
anxiety over judicial review is not its countermajoritarian nature as such;
it is rather the seeming finality of a constitutional pronouncement by the
Supreme Court.
5
The task of defending judicial review is a different one if the anxiety
over the practice derives from problems of finality rather than from the
inescapable fact that judicial review is countermajoritarian. In order to
distinguish between these tasks, it may be helpful, before addressing the
issue of finality, to examine briefly a justification of judicial review that is
concerned primarily with its countermajoritarian nature.
Today the most prominent such justification draws its inspiration from
the second and third paragraphs of footnote four of a 1938 case, United
States v. Carolene Products Co. 72 But for the footnote, the case would not
be remembered. It stands with others of its time in sustaining economic
legislation that less than a decade earlier would have been invalidated
under the due process clause. Speaking for the Court, Justice Stone made
it plain that economic legislation was to enjoy a powerful presumption of
constitutionality, indeed, that it would survive challenge under that once
formidable clause if the legislation had a "rational basis."
The footnote, after suggesting that this presumption might have less
force when legislation is questioned under constitutional provisions that
are more linguistically precise than due process, continues as follows:
It is unnecessary to consider now whether legislation which restricts

those political processes which can ordinarily be expected to bring
about repeal of undesirable legislation, is to be subjected to more
exacting judicial scrutiny under the general prohibitions of the Four-
teenth Amendment than are most other types of legislation ....
Nor need we enquire whether similar considerations enter into the
review of statutes directed at particular religious . . . or national
. . . or racial minorities . . . ; whether prejudice against discrete
and insular minorities may be a special condition, which tends seri-
ously to curtail the operation of those political processes ordinarily to
be relied upon to protect minorities, and which73
may call for a corre-
spondingly more searching judicial inquiry.
Footnote four can be seen as "a participation-oriented, representation-
71. For reasons largely unrelated to the countermajoritarian difficulty, Article III has been con-
strued to forbid premature judicial judgments. See Poe v. Ullman, 367 U.S. 497 (1961). For a differ-
ent view of Article III, see A. BICKEL, supra note 8, at 115-16.
72. 304 U.S. 144, 152 n.4 (1938).
73. Id. at 152, 152 n.4, 153.

reinforcing approach to judicial review,"" and, as such, it is wonderfully

attractive. It aims our attention to failures of process in the legislative or
executive branches rather than to failures of substance. And process, after
all, is a comfortable and familiar domain of lawyers and judges.
Professor John Hart Ely, in a recent book, puts footnote four in terms
of the malfunctioning of the more majoritarian branches:
Malfunction occurs when the process is undeserving of trust, when

(1) the ins are choking off the channels of political change to ensure
that they still stay in and the outs will stay out, or (2) though no one
is actually denied a voice or a vote, representatives beholden to an
effective majority are systematically disadvantaging some minority
out of simple hostility or a prejudiced refusal to recognize commonal-
ities of interest, and thereby denying that minority the protection af-
forded other groups by a representative system."
While malfunction in the process may be a good reason for the exercise
of judicial review, the notion that its absence requires the denial of review
is, at least on first impression, both striking and contrary to our intuitions.
Are there not substantive malfunctions that require serious judicial re-
view? Is it not the case, for example, that government sometimes intrudes
on individual autonomy and privacy without sufficient justification? Is not
the Court's capacity here as great, and its role as important, as when the
perceived malfunction is one of process?
Professor Ely's "answer" is emphatically negative:
Our government cannot fairly be said to be 'malfunctioning' simply

because it sometimes generates outcomes with which we disagree,
however strongly. . .. In a representative democracy value determi-
nations are to be made by our elected representatives,and if in fact
most of us disapprove we can vote them out of office.7"
Ely's attempt to withdraw value determinations from the courts results

from his commitment to the simple majoritarian premises that I have sug-
gested do not accurately reflect the nature of our governmental structure.
In a representative democracy, value determinations are made, and are
meant to be made, by many actors who cannot possibly be described as
"our elected representatives."
Moreover, this unrealistic attachment to simple majoritarian prem-
ises-either as an is or an ought-may indeed impose an insuperable bur-
74. J. ELY, supra note 7, at 87.

75. Id. at 103 (emphasis in original) (footnote omitted).
76. Id. (emphasis added).

Judicial Review
den on proponents of judicial review. Surely Ely does not overcome the
problem that he thinks they present. For there are at least two inevitable
and fatal difficulties with his attempt to bar consideration of values in
constitutional adjudication. First, the judicial diagnosis of a process mal-
function may itself entail a value determination. For example, given the
method for selecting United States senators and the gerrymander, it is far
from clear how, in many situations, a court can know whether a legisla-
tive apportionment is improper, unless, of course, it develops a theory of
political fairness. But can anyone develop such a theory without making
at least one value determination?" Second, the judicial cure for a process
malfunction also may require judges to make value determinations. Con-
sider free speech. "Courts must," as Ely tells us, "police inhibitions on
expression and other political activity because we cannot trust elected offi-
cials to do so: ins have a way of wanting to make sure the outs stay out." 8
Agreed. But a court must place a value on reputation if it is to fashion a
First Amendment rule protecting expression considered defamatory under
state law. New York Times v. Sullivan79 may be the correct cure for a
process malfunction. But after reading the case, it is apparent that Justice
Brennan found it necessary to make more than one value determination.
Judicial value determinations are inescapable. As I have argued, they
are also unexceptional. What is exceptional in judicial review is that value
determinations in constitutional cases have the appearance of finality. And
this is perceived even though the Court has neither purse nor sword; 0
even though the Congress has power over the Court's appellate jurisdic-
77. For the value determinations that Ely makes, see id. at 123.
78. Id. at 106.
79. 376 U.S. 254 (1964).
80. As Alexander Hamilton put it:
Whoever attentively considers the different departments of power must perceive that, in a
government in which they are separated from each other, the judiciary, from the nature of its
functions, will always be the least dangerous to the political rights of the Constitution; because
it will be least in a capacity to annoy or injure them. The executive not only dispenses the
honors but holds the sword of the community. The legislature not only commands the purse,
but prescribes the rules by which the duties and rights of every citizen are to be regulated. The
judiciary, on the contrary, has no influence over either the sword or the purse; no direction
either of the strength or of the wealth of the society; and can take no active resolution
whatever. It may truly be said to have neither FORCE nor WILL, but merely judgment; and
must ultimately depend upon the aid of the executive arm even for the efficacy of its
judgements.
THE FEDERALIST No. 78, at 465 (A. Hamilton) (C. Rossiter ed. 1961).

tion;8" indeed, even though there is, within the Constitution itself, the pos-
sibility of amendment. 2
If it can be shown, however, that often there is less finality in a consti-
tutional decision than meets the eye, and that (unlike bureaucrats) the
value determinations of judges in constitutional cases are constrained by
norms applicable generally in adjudication, then perhaps we can accept
more readily the legitimacy of judicial review.
6
The finality of judicial review, even as generally perceived, may seem
less troublesome if compared with the accepted judicial function of statu-
tory interpretation. In the Rotolith, Darlington, and Vincent cases, we
saw how courts sometimes depart sharply from legislative intent; some-
times they do more. In some situations statutory interpretation forecloses
legislative change. Thus, it may have more finality than we imagine (more
indeed than many constitutional interpretations), and therefore, it can be
more closely related to the standard picture of judicial review than is often
supposed. 3
Consider two examples: administrative agencies-particularly when
they are young-sometimes interpret statutes with a zeal that is a variety
of the tunnel vision I mentioned earlier. Instead of harmonizing the law it
administers with the general law, an agency may ignore the surrounding
jurisprudence, and its constitutional configurations. On review, the courts
may then reinterpret the statute to avoid confronting the constitutional
questions presented by the agency's single-minded devotion to its mis-
sion. 4 This describes the National Labor Relations Board in its spring-
81. The scope of the power is not clear. Ex parte McCardle, 74 U.S. (7 Wall.) 506 (1869), is the
leading case. It should be read with Ex parte Yerger, 75 U.S. (8 Wall.) 85 (1869). See C. BLACK,
DECISION ACCORDING TO LAW (1981); Van Alstyne, A Critical Guide to Ex Parte McCardle, 15
ARIZ. L. REV. 229 (1973). See also Sanger, The Supreme Court, 1980 Term-Foreword: Constitu-
tional Limitations on Congress' Authority to Regulate the Jurisdiction of the Federal Courts, 95
HARV. L. REV. 17 (1982) (discussing limits of congressional authority to contract jurisdiction of fed-
eral courts). Compare Wechsler, The Appellate Jurisdictionof the Supreme Court: Reflections on the
Law and the Logistics of Direct Review, 34 WASH. & LEE L. REV. 1043 (1977) (to insure against
nullification of constitutional rights, Supreme Court must scrutinize state court decisions of law and
fact) with Hart, The Power of Congress to Limit the Jurisdiction of FederalCourts: An Exerdse in
Dialectic, 66 HARV. L. REv. 1362 (1953) (suggesting that state courts may be ultimate guarantors of
constitutional rights).
82. Only four Supreme Court decisons have been set aside by constitutional amendment: Oregon
v. Mitchell, 400 U.S. 112 (1970); Pollock v. Farmers' Loan & Trust Co., 157 U.S. 429 (1895); Dred
Scott v. Sanford, 60 U.S. (19 How.) 393 (1857); Chisholm v. Georgia, 2 U.S. (2 Dall.) 419 (1793).
Justice Black, criticizing the Court's reinterpretation of a statute, has said: "[W]here the only alter-
native to action by this Court is the laborious process of constitutional amendment and where the
ultimate responsibility rests with this Court, I believe reconsideration [of a constitutional decision] is
always proper." Boys Mkts. v. Retail Clerks Union, 398 U.S. 235, 259 (1970) (Black, J., dissenting).
83. As will become clearer later, my claim is also that the reasons for this are related to the
reasons for finality-when there is finality-in judicial review.
84. For an example of judicial disagreement over the propriety of this type of agency behavior in

Judicial Review
time. It had a propensity to read the Wagner Act as if there were no First
Amendment to the United States Constitution. 5 When unions organized,
employers were at risk if they spoke, as they often did, in favor of the
open shop."6
The Supreme Court, through statutory interpretation, denied the Board
this power. To read the statute as the Board had would raise serious con-
stitutional issues although it might be that the reading was constitutional.
The issue could be avoided by interpretation that was sensitive to the
value our law attaches to free expression. 7
Because the Court did not render a constitutional decision in these
cases, Congress retained the formal power to amend the statute to make it
conform to the agency's interpretation. But in fact this would not have
been easy. It is hard to rewrite labor statutes, and often it is difficult for
Congress to ignore values that may be of constitutional dimension and
that are called to its attention by the Supreme Court.8
In this statutory situation there is more finality than meets the eye, but
little doubt about the legitimacy of the Court's role. Indeed, even if the
Court did not have the power to declare legislation unconstitutional, it
would have the authority to review agency action and the obligation to
accommodate, through statutory interpretation, particular legislation to
the principles that underpin our law.
The second example is again from labor law, but does not involve an
administrative agency. Under the Railway Labor Act,"9 the union selected
by a majority of employees in a bargaining unit is the exclusive represen-
tative of all employees in the unit. The statute is absolutely silent, how-
ever, on the question of the union's obligations to the employees it repre-
sents. In 1944 the practice of the brotherhoods, as some of the railway
unions are called, was to discriminate and to do so with vigor."' Some
employees who suffered sued. Their case made its way to the Supreme
Court, which held that the "fair interpretation of the statutory language is
a non-constitutional setting, compare the majority and dissenting opinions in D.A. Schulte, Inc. v.
Gangi, 328 U.S. 108 (1946).
85. See H.R. REP. NO. 245, 80th Cong., 1st Sess. 33 (1947) ("Although the Labor Board says it
does not limit free speech, its decisions show that it uses against people what the Constitution says
they can say freely.")
86. See NLRB v. Golub Corp., 388 F.2d 921, 926 (2d Cir. 1967) ("Under the Wagner Act.
the Board condemned almost any antiunion expression by an employer.")
87. See NLRB v. Virginia Elec. & Power Co., 314 U.S. 469 (1941) (construing National Labor
Relations Act narrowly to avoid clash between requirements of Act and employer's First Amendment
rights); cf.Thomas v. Collins, 323 U.S. 516, 537 (1945) (application of state labor statute that im-
posed prior restraint on organizer held unconstitutional).
88. Some of the subsequent history of employer speech and the congressional response to Board
and Court protection is traced in NLRB v. Golub Corp., 388 F.2d 921 (2d Cir. 1967).
89. 45 U.S.C. §§ 151-88 (1976).
90. See H. WELLINGTON, LABOR AND THE LEGAL PROCESS 145-55 (1968).

that the organization chosen to represent . . . is to represent all . . . and

it is to act for and not against those whom it represents."'" Chief Justice
Stone, in his opinion for a unanimous Court, reasoned as follows:
It is a principle of general application that the exercise of a granted

power to act in behalf of others involves the assumption toward them
of a duty to exercise the power in their interest and behalf, and that
such a grant of power will not be deemed to dispense with all duty
toward those for whom it is exercised unless so expressed."
The common-law doctrine that the Chief Justice grafted onto the stat-
ute-namely, that an agent must represent his principal fairly-is not a
doctrine that a legislature could, even under tremendous political pressure
from powerful groups, easily reject by amending the statute over which it
theoretically has authority. 3 There is more finality in the interpretation of
statutes, when that interpretation reflects the principles of our jurispru-
dence, than we sometimes imagine.'4
7
In addressing the issue of finality in judicial review itself, it is essential
to distinguish among various types of constitutional decisions. Some con-
stitutional decisions do not even appear to be final. Others deal primarily
with means rather than ends, and thus leave more than a little legislative
discretion intact. Finally, certain types of decisions, although facially final,
are properly subject to thorough revision, the principle of stare decisis
notwithstanding.
The class of plainly non-final decisions might best be understood in
terms of the Court saying to another governmental entity: "You may be
able to achieve the substantive result you desire but you must proceed
toward your objective in a different fashion." There is a family of such
91. Steele v. Louisville & N. R.R., 323 U.S. 192, 202 (1944).
92. Id.
93. Nor can this finality be explained simply on the theory that Steele was, at bottom, a constitu-
tional decision. Justice Murphy, concurring, did argue that Steele should have been decided on consti-
tutional grounds. 323 U.S. at 192, 208-09. At the time, however, this suggestion was innovative. The
state action doctrine in 1944 was a long way from what it is today. For its application to unions, see
Machinists v. Street, 367 U.S. 740 (1961), decided seventeen years after Steele. Moreover, Steele
preceded Brown v. Board of Educ., 347 U.S. 483 (1954), by ten years; in 1944, the common-law
notion of fair representation may have operated more powerfully on behalf of minorities than the
equal protection clause.
94. See supra p. 502.

Judicial Review
doctrines, procedural or structural in 97nature, such as "delegation,""5 "void

for vagueness,"' 6 and "overbreadth."
These doctrines have been much analyzed, and some have been de-
scribed as techniques for introducing flexibility into constitutional adjudi-
cation.'" The focus here is not on their flexibility or on the notion, as such,
that they enable the Court to initiate a dialogue with another branch of
government." Nor am I now concerned with the related idea, that the
more majoritarian branches must assume responsibility by speaking
clearly and precisely if they want to achieve a goal that either lies in the
"foothills of the Constitution," or that stigmatizes through the use of the
criminal sanction."' Of course, these ideas are closely intertwined with my
conception of the importance of countermajoritarian forces-especially the
judiciary-to American democracy; but my interest at this point is in the
degree of finality that attends a judicial decision.
In this respect the infrequently employed (at the federal level) delega-
tion doctrine should be seen as requiring that the legislature spell out in
more detail the standards a regulatory agency should apply in effectuating
the substantive goal of the statute; the void for vagueness doctrine as re-
quiring more specificity in articulating the goal to be achieved by the leg-
islation under review; and the overbreadth doctrine as requiring that the
substantive goal be stated as narrowly as possible. Far from foreclosing
legislative choices, therefore, these doctrines shape the process by which
legislative goals may be achieved.
8
Justice Stevens, dissenting in Fullilove v. Klutznick, 10 applied another
variety of structural or procedural, and therefore non-final, review. The
case involved the constitutionality of the minority business enterprise, or
"set aside," requirement of the Public Works Employment Act of 1977.
That provision requires, absent an administrative waiver, that at least ten
percent of federal funds allocated to local public works projects be used to
95. See, e.g., Hampton v. Mow Sun Wong, 426 U.S. 88 (1976); Kent v. Dulles, 357 U.S. 116
(1958); see also Gewirta, The Courts, Congress,and Executive Policy-Making: Notes on Three Doc-
trines, 40 LAW & CONTEMP. PROBS. 46 (1976).
96. See, e.g., Cramp v. Board of Pub. Instruction, 368 U.S. 278 (1961); Thornhill v. Alabama,
310 U.S. 88 (1940).
97. See, e.g., United States v. Robel, 389 U.S. 258 (1967); Zwickler v. Koota, 389 U.S. 241
(1967).
98. See A. BICKEL, supra note 8, at 111-98.
99. See generally Bickel & Wellington, supra note 26 (discussing circumstances in which courts
should "remand" statutes to Congress for further consideration).
100. See Wellington & Albert, Statutory Interpretationand the Political Process: A Comment on
Sinclair v. Atkinson, 72 YALE L.J. 1547, 1560 n.42 (1963).
101. 448 U.S. 448, 548-49 (1980).

procure services or supplies from businesses owned by United States citi-

zens "who are Negroes, Spanish-speaking, Orientals, Indians, Eskimos,
2
and Aleuts."''
In a six to three split, the Court upheld the statute from a predictable
equal protection challenge. Justice Stevens' dissent rested on due process
grounds, and procedural due process at that:
The very fact that Congress for the first time in the Nation's history
has created a broad legislative classifiction for entitlement to benefits
based solely on racial characteristics identifies a dramatic difference
between this Act and the thousands of statutes that preceded it. This
dramatic point of departure is not even mentioned in the statement
of purpose of the Act or in the reports of either the House or the
Senate Committee that processed the legislation, and was not the
subject of any testimony
10 3
or inquiry in any legislative hearing on the
bill that was enacted.
The Justice went on to describe this as a "malfunction of the legislative

process" because of the "perfunctory consideration of an unprecedented
policy decision of profound constitutional importance to the Nation." He
said:
Whenever Congress creates a classification that would be subject to

strict scrutiny under the Equal Protection Clause of the Fourteenth
Amendment if it had been fashioned by a state legislature, it seems
to me that judicial review should include a consideration of the pro-
cedural character of the decisionmaking process. A holding that the
classification was not adequately preceded by a consideration of less
drastic alternatives or adequately explained by a statement of legisla-
tive purpose would be far less intrusive than a final determination
that the substance of the decision is not "narrowly tailored to the
achievement of that goal." 104
Justice Stevens' approach invites attention to separation of powers

questions, as he notes.'0° They are questions that are also present in one
of the statutory interpretation examples (the Railway Labor Act case) dis-
cussed earlier, and are in any event triggered by the doctrines of vague-
ness, delegation, and overbreadth. For surely whenever a court tells a leg-
islature that it must speak clearly if it is to accomplish the goal it seems to
have in mind-and that happens not infrequently'° 6-the court is in-
102. 42 U.S.C. § 6705(0(2) (1976 & Supp. I 1978).

103. 448 U.S. 448, 548-49 (1980).
104. Id. at 550-51 (footnote omitted).
105. Id. at 552.
106. See Wellington & Albert, supra note 100, at 1559-66.

Judicial Review
structing the legislature on its internal procedures. Thus, Justice Stevens'

approach-his insistence on structuring the law-making process-may be
no more than an extension of an established judicial practice.
9
The Justice's approach invites us to examine other procedural due pro-
cess or related constitutional interventions by the Court that may have less
finality than we generally associate with judicial review. Consider the at-
tempted judicial regulation of police behavior through such continuingly
controversial decisions as those that established the Miranda require-
ments0 7 and imposed the exclusionary rule. 0 8 I say "attempted" because
part of the controversy turns on whether the requirements and the rule do
make any difference in police behavior. Of course, in determining whether
the Court's decisions are justified it may not matter. The point is that
neither Miranda nor the exclusionary rule questions any substantive goal
of any governmental entity. Both structure the means of achieving an end;
both, perhaps, make the end's achievement more difficult. So do other
things we tolerate, indeed, embrace. For example, police unions may in-
terfere with efficient police work and the efficient administration of crimi-
nal justice as much or perhaps more than do these constitutional decisions.
Moreover, there is as much certainty of the continuing existence of police
unions as there is of the survival of Miranda and the exclusionary rule,
and the voter may have as little direct control over one as he does over the
other.
10
This discussion of structural or procedural judicial review highlights
the important role of the Supreme Court as umpire-at-the-margin of the
appropriate processes of other governmental entities. It reveals that judi-
cial review, when so employed, does not preclude substantive legislative
goals and accordingly that, so far as substantive finality is concerned, such
constitutional adjudication closely resembles common-law and statutory
interpretation: all three may make it harder for legislatures to reach par-
ticular goals; usually none prevents the attainment of those goals.
The examples employed of structural or procedural judicial review
were, of course, meant to be suggestive; plainly, they are not exhaustive. I
do, however, want to continue the discussion in one additional context,
that of equal protection. The place to begin is with an observation made
by Justice Jackson in Railway Express Agency v. New York.109 The case
107. Miranda v. Arizona, 384 U.S. 436 (1966).

108. Linkletter v. Walker, 381 U.S. 618 (1965); Mapp v. Ohio, 367 U.S. 643 (1961).
109. 336 U.S. 106 (1949).

involved a New York City traffic regulation that barred advertising on

delivery vehicles, unless the vehicles were "engaged in the usual business
of regular work of the owner and not used merely or mainly for advertis-
ing." Railway Express had violated the regulation and was fined. It
claimed that its due process and, more emphatically, its equal protection
rights had been infringed. The Court was not sympathetic. But in his
concurring opinion Justice Jackson said:
The burden should rest heavily upon one who would persuade us to
use the due process clause to strike down a substantive law or ordi-
nance. . . . Invalidation of a statute or an ordinance on due process
grounds leaves ungoverned and ungovernable conduct which many
people find objectionable.
Invocation of the equal protection clause, on the other hand, does
not disable any governmental body from dealing with the subject at
hand. It merely means that the prohibition or regulation must have a
broader impact.110
Justice Jackson has a point, but the point is of limited utility if the
issue is-as it is for us-the relative finality of judicial review. For equal
protection cases vary considerably in the breadth and permanence of the
restraints they impose on legislatures. Some equal protection cases are
structural in the same way procedural due process cases may be;"' some
are essentially substantive due process cases;" 2 some are republican form
of government cases,' or at least they seem to be when the verbal mask is
stripped away. Others, such as the race cases, which involve affirmative
governmental discrimination, are final in any realistic sense because they
imagine a nation that must be profoundly different from one in which the
Constitution would permit the discrimination. Moreover, they entail a re-
medial program that, while it may not achieve the ideal of racial equality,
does substantially rearrange the country's priorities." 4
Yet ei6en the desegregation cases were initially provisional. The finality
we see today reflects today's acceptance of the value derived and articu-
lated as law by the Supreme Court in 1954, namely (with certain qualifi-
cations), that government may not affirmatively discriminate on the basis
of race. Without today's acceptance there would be no finality; but at the
time of decision, such acceptance was in doubt.
In the most substantive of constitutional cases there is always doubt
110. Id. at 112.

111. See, e.g., Boddie v. Connecticut, 401 U.S. 371 (1971); Griffin v. Illinois, 351 U.S. 12 (1956).
112. See Shapiro v. Thompson, 394 U.S. 618 (1969).
113. See, e.g., Harper v. Virginia Bd. of Elections, 383 U.S. 663 (1966); Reynolds v. Sims, 377
U.S. 533 (1964).
114. These cases begin, of course, with Brown v. Board of Educ., 347 U.S. 483 (1954).

Judicial Review
about finality at the time of decision because there is always the possibility
of judicial mistake and the inevitability of social change. When we talk of
mistake and social change-both terms that need elucidation-we are face
to face with questions about the sources of law.
11.
Two types of judicial mistake concern us here. Each is related to a type
of justification that a court might offer for a decision. And it is this rela-
tionship that connects the topic of judicial mistake with that of the sources
of law. Courts-at least appellate courts-generally believe themselves
under an obligation to justify what they hold. There are many standard
modes of justification: an appeal to a legislative command, to a judicial
precedent, to legislative history, or to analogous developments in other ju-
risdictions or in related areas of law. Moreover, the consequences that
may attend a legal rule can sometimes serve as the justification for a judi-
cial decision; so sometimes can an appeal to the moral ideals of the
community.
The latter two types of justification are the most problematic in appli-
cation. And they typically come into play when the more standard modes
of justification themselves become problematic. It is fair to conclude, there-
fore, that they most often give rise to judicial mistake.
The first type of justification is consequentialist; it looks to the future.
The rule on which a holding rests will, according to the rationale of the
decision, change the behavior of individuals or institutions. The articu-
lated rule serves a policy that in turn is designed to effectuate a societal
goal; the justification sounds in terms of benefits and costs.
The second type of justification looks to the past. Its persuasiveness is
not in what will be; it does not move us toward a goal. Instead, the rule
has persuasiveness because it vindicates a principle embedded in the moral
ideals of the community; the justification sounds in terms of rights and
obligations.
Of course, the vindication of a principle may have significant effects.
Indeed, it may entail-as in desegregation-remedial efforts that change
the nature of society. But this does not mean, in this class of cases, that
effect should be equated with justification any more than it should be
when a court carries out a legislative command by enforcing a statute.
There too the decision has an effect, but the justification for the decision
does not therefore become its effect; it remains the authority of the
legislature.
Earlier I suggested that final responsibility for the moral ideals of the
community cannot adequately be lodged in either countermajoritarian in-
stitutions with tunnel vision or majoritarian institutions in which the pas-

sionate and self-interested concerns of the moment may prevail too easily.
It does not follow, however, that courts-because they are better suited to
derive and articulate the community's moral ideals-will not make mis-
takes in doing so. And, of course, courts are also apt to make mistakes
when using forward looking policies to justify their rules; indeed, in the
policy area courts have no comparative institutional advantage.
It should be added that the two types of justifications I have adum-
brated are hard to discover in pure form in the judicial world. Courts tend
to mix them together. And, although they may be easy enough for a phi-
losopher to describe, a lawyer cannot easily separate them. Still, either a
forward looking policy or a backward looking principle may be dominant
in the explanation of a legal rule. And it behooves us, in trying to under-
stand legal phenomena, to try to draw the distinction. While Cardozo in
The Nature of the JudicialProcess did not explicitly make that effort, his
lectures are-as he might have put it-instinct with the idea.
Both policies and principles are important sources of law, including
constitutional law. In a formal sense-so far as judicial review is con-
cerned-they, along with history and precedent, give content to the open
texture of the Constitution. An example of each will serve to clarify the
distinction between them.
12
The major justification for the rule announced in New York Times v.
Sullivan"' is a policy. The case held that, absent actual malice,"' the
First Amendment (via the Fourteenth) bars a public official from recover-
ing damages for false and defamatory statements relating to his official
conduct. The goal to be fostered through this rule is-at its most ab-
stract-the sound working of American democracy and, at a somewhat
reduced level of abstraction, "a commitment . . . that debate on public
issues should be uninhibited, robust, and wide-open ... .""I The Court's
justification in part took the form that without such a rule, "critics of
official conduct may be deterred from voicing their criticism, even though
it is believed to be true and even though it is in fact true because of doubt
whether it can be proved in court or fear of having to do so.'"" The
common law of defamation, the Court reasoned, "dampens the vigor and
limits the variety of public debate.""' The rule, in short, is justified by
predictable consequences that serve a desirable goal.
115. 376 U.S. 254 (1964).

116. A defendant is guilty of actual malice if his statement was made "with knowledge that it was
false or with reckless disregard of whether it was false or not." Id. at 280.
117. Id. at 270.
118. Id. at 279.
119. Id.

Judicial Review
Contrast this with the dissent of Justice Brandeis in the famous 1928
case of Olmstead v. United States.120 At issue in that case was whether
wiretapping by federal officers constituted an unreasonable search and
seizure under the Fourth Amendment. According to Brandeis:
The makers of our Constitution undertook to secure conditions

favorable to the pursuit of happiness. . . . They sought to protect
Americans in their beliefs, their thoughts, their emotions and their
sensations. They conferred, as against the government, the right to
be let alone-the most comprehensive of rights and the right most
valued by civilized men. To protect that right, every unjustifiable
intrusion by the government upon the privacy of the individual,
whatever the means employed, must be deemed a violation of the 4th
Amendment. 121
Of course, if Brandeis is "doing" history, he is attributing a great deal to

the "makers of our Constitution," given the fact that he is concerned with
a contemporary technology in a changed world. But he is not "doing"
history in any conventional sense. His is a quest for our community's
22
moral ideals.
One might argue that the rule urged by Brandeis in Olmstead is justi-
fied by him in much the same way the Sullivan rule is justified: in conse-
quentialist or policy terms. That would be mistaken. It would show either
an insensitivity to linguistic nuance or a determination to conflate concepts
that it is useful, if difficult, to keep separate. In Sullivan, the justification
for the rule protecting expression looks to the timid behavior of the gov-
ernment's critics, who are faced with the common law of defamation. In
Olmstead, Brandeis is not concerned directly or principally with the be-
havior of government officials who authorize wire tapping; his is a con-
cern with the individual's right to be let alone. The rule that he would
have had the Court adopt did not depend for its validity upon the subse-
quent conduct of such officials: less wiretapping was desirable, but the
justification for the proposed rule was individual privacy. That right was
what Brandeis wished the Court to vindicate. In contrast, the rule that the
Sullivan Court adopted depends upon the conduct of potential critics of
government when released from fear of litigation: an increase in "the
vigor and variety of public debate" is not only desirable, it is the justifica-
tion for the rule. The right of the individual to speak-to exercise his
autonomy-is a welcome side effect.
120. 277 U.S. 438, 471-85 (1928).

121. Id. at 478.
122. C Fiss, supra note 39, at 11-12 (role of judge to give content to constitutional values).

13
This distinction between a forward looking policy justification and a
backward looking justification from principles embedded in the moral ide-
als of the community in turn suggests a distinction between two types of
judicial mistake. If it were to become clear that the rule in New York
Times v. Sullivan did not have the consequences the Court predicted-if it
became clear that the behavior of potential critics of government did not
change and that "the vigor and variety of public debate" had not been
altered-then New York Times v. Sullivan would be either a constitu-
tional mistake or, at least, a decision in search of a new justification.
Speaking in a different context and about common-law cases, Cardozo
said: "Those that cannot prove their worth and strength by the test of
experience, are sacrificed mercilessly and thrown into the void." 'M But
what do we mean by "the test of experience"? A plausible answer is social
science investigation, empirical research based upon a statistically sophisti-
cated methodology. In addition to New York Times v. Sullivan, consider
some other subjects-hypothetical and real-for this "test of experience":
a constitutional rule excluding the introduction of illegally obtained evi-
dence, justified by the rule's deterrent effect on unlawful police behav-
ior; 12 a constitutional rule requiring at least twelve on a jury, justified by
the proposition that the protection afforded the defendant is decreased by
a decrease in jury size; a determination that the death penalty is unconsti-
tutional in felony-murder cases, because it fails to deter felony murder.
Without in any way implying criticism of the social sciences, one can
agree that in at least some of these situations-or others one could think
of-scientific tests of experience would not lead to clear conclusions, if
indeed scientific tests were possible.2 2 Nor is any other "test of experi-
ence," such as informed intuition, apt to be better.
This conclusion-that there may be no clear conclusions-raises ques-
tions about the use of consequentialist or policy justifications in constitu-
tional law. From this perspective it. seems acceptable enough if the evi-
dence on which the policy is based is clear. When it is not, when judges
must rely on intuition because social scientists or other data collectors can-
not tell them very much about the consequences of legal rules, should not
policy choices be left to the legislatures? Should choices not be subject to
123. B. CARDOZO, supra note 1, at 22.

124. See, eg., Stone v. Powell, 428 U.S. 465 (1976); Linkletter v. Walker, 381 U.S. 618 (1965).
125. See, e.g., Klevorick & Rothschild, A Model of the Jury Decision Process,8 J. LEGAL STUD.
141 (1979); Levin, Education, Life Chances,and the Courts: The Role of Social Science Evidence, 39
LAW & CONTEMP. PROBS. 217 (1975); Oaks, Studying the ExclusionaryRule in Search and Seizure,
37 U. CHI. L. REV. 665 (1970).

Judicial Review
state "laboratories of experimentation" and the trade-offs of national

politics? "'
If this problem is viewed through the lens of judicial mistake, a normal
presumption in favor of stare decisis should make a judicial declaration of
unconstitutionality difficult to reverse because of the difficulty in disprov-
ing its empirical assumptions. And unless the societal goals that the policy
serves have changed, that is the only proper reason for overruling the de-
cision. Public or political reaction, for example, is not evidence that the
exclusionary rule is wrong if the exclusionary rule deters illegal searches
and seizures and if deterrence is the reason for the rule. This suggests that
forward looking justifications for constitutional rules may be troublesome:
either they may have more finality than is healthy for a constitutional
solution based on problematic assumptions or they may tend not to endure
for improper reasons.
Two caveats: First, the exclusionary rule, as judicial policy, is substan-
tially less troublesome than defamation and the First Amendment because,
as noted earlier, it redirects governmental activity without changing the
government's goal. In this sense it is far less final than Sullivan, which
substantially restricts the government's protection of reputation. This may
suggest that, when outcomes are problematic, courts should be entitled to
use consequentialist justifications only for constitutional rules of a struc-
tural or procedural nature. Second, while this discussion should be taken
as requiring a justification, not a rationalization, for constitutional rules, it
should not be taken as a criticism of the holding in New York Times v.
Sullivan, or the exclusionary rule cases, or any other constitutional case.
It questions only the approach;127 it is neutral as to results. The exclusion-
ary rule, for example, could perhaps be justified in Brandeisian terms as
an aspect of "the privacy of the individual," without regard to its effect on
128
police practices.
14
The second type of judicial mistake is associated with rules that are
justified by looking to the past, rules that are persuasive because they vin-
dicate principles embedded in the moral ideals of the community. The
Brandeis dissent in Olmstead is the example of this second type of justifi-
126. See Wellington, supra note 40, at 267.

127. "Further, if social science findings increasingly are used to create what appear to be technical
issues out of essentially moral dilemmas, this presents a potential social danger." Levin, supra note
125, at 240.
128. Compare Mapp v. Ohio, 367 U.S. 643 (1961) (emphasizing "Fourth Amendment's right of
privacy" as warranting exclusionary rule) with Linkletter v. Walker, 381 U.S. 618 (1965) (emphasiz-
ing deterrent effect as justification for exclusionary rule).

cation. There Brandeis spoke of the "privacy of the individual." As a con-

stitutional principle-as well as a concept in torts-privacy has had an
interesting judicial history. It appears in Griswold,129 the Connecticut con-
traceptive case, and, of course, in the abortion cases.130 Some years ago in
these pages I concluded, after a detailed analysis, that the 1973 abortion
decisions were partially right but overly broad in terms of legal principles
derived from conventional morality, M and in The Nature of the Judicial
Process, Cardozo talks of "customary morality" 2 as a source of law. But
it seems more in keeping with the conception of the judicial function ad-
vanced here to speak about the moral ideals of the community.'3 This is
so not because it is easier to get our moral ideals straight, but. because that
terminology shifts-perhaps ever so slightly-the emphasis from obliga-
tion to aspiration. While this shift does not change my estimate of the
abortion cases, it may with some issues lead one to slightly different con-
clusions. However, conventional or customary morality and the moral.
ideals of the community raise similar problems where, as here, the issue is
mistake. This is so because both approaches are-perhaps equally
-problematic in result.
It is certainly not difficult to imagine that the moral ideals of the com-
munity are often less than scrupulously regarded in the give and take of
the legislative process with its necessary compromises, trade-offs, and es-
sential goal orientation. Nor should it be surprising that mistakes occur in
the comparatively better situated judicial process. Even the most conscien-
tious judge can get the community's moral ideals wrong.
But what is the criterion here for judicial mistake? What is the mean-
ing of Cardozo's "test of experience" in this context? It does not seem to
be empirical investigation. Perhaps it is the community's reaction to a ju-
dicial decision.
It is, of course, obvious that the articulation by the Court of the moral
ideals of the community is not neutral. The articulation itself has an ef-
fect-it is bound to have an effect-on those ideals. What that effect will
be-indeed, the direction it will take-is another question, one that is not
129. Griswold v. Connecticut, 381 U.S. 479 (1965).

130. Doe v. Bolton, 410 U.S. 179 (1973); Roe v. Wade, 410 U.S. 113 (1973).
131. Wellington, supra note 40, at 243-51, 285-311.
132. See B. CARDOZO, supra note 1, at 104.
133. The notion that a court might use the moral ideals of a community as a bar against the
actions of majorities has been much criticized. In particular, it has been claimed that the actions of
majorities are the best evidence of communities' ideals, and thus that the latter cannot coherently be
said to restrain the former. See J. ELY, supra note 7, at 69 ("It makes no sense to employ the value
judgments of the majority as a means of protecting minorities from the value judgments of majori-
ties.") This criticism, however, overlooks the familiar fact that communities, like individuals, may well
violate principles to which they usually adhere. Such inconsistency is most likely in the legislative
process, when self-interest is least neutralized by considerations of conscience or moral obligation.

Judicial Review
susceptible to a general answer. Yet we do know that when the Justices

get the community's moral ideals wrong in constitutional cases, they hear
about it. Of course it is also the case that when they get them right they
hear about it."14 In both situations there may be turmoil, resistance, and
threats from other governmental entities, from private groups, institutions
and individuals. There is always discussion and analysis; some of it may
even be informed and dispassionate. All of it helps to clarify.
Contemporary historical examples might begin with the Court and the
New Deal," 5 and continue with Brown v. Board of Education.136 That
case gave rise to a heightened level of political activity that was sustained
for years and that includes, on the one hand, such unforgettable events as
the Southern Congressional Manifesto 37 and the opposition to integration
at Little Rock,"' and, on the other, the heroic activities that constituted
the civil rights movement.' Moving on in time, the political unrest asso-
134. Consider, as related to the theme I am developing, Andrew Jackson's veto message of July
10, 1832 on a bill to recharter the Bank of the United States:
It is maintained by the advocates of the bank that its constitutionality in all its features ought
to be considered as settled by precedent and by the decision of the Supreme Court. To this
conclusion I cannot assent. Mere precedent is a dangerous source of authority, and should not
be regarded as deciding questions of constitutional power except where the acquiescence of the
people and the States can be considered as well settled.
Jackon's message is reproduced in 2 MESSAGES AND PAPERS OF THE PRESIDENTS 576 (Richardson ed.
1896). But cf Cooper v. Aaron, 358 U.S. 1 (1958) (Court is supreme arbiter of Constitution). Had
Cooper been decided in 1832, Jackson could not have fashioned his argument in the manner presented
above: that all officers of government are bound by the judgments of the Supreme Court would not
have been a matter open to debate. But Jackson's argument could have been reshaped. As Bickel
noted, decisions of the Supreme Court, unlike the injunctive order at issue in Cooper, see infra note
138, are not self-executing. See A. BICKEL, THE MORALITY OF CONSENT 111 (1975). The legal free-
dom thus accorded parties not bound by concluded litigation may well, like any freedom, be abused.
But there is no reason to suppose that private resistance to constitutional decisions is always or usually
invidiously motivated; nor is there reason to assume the infallibility of the courts.
135. See supra note 3.
136. 347 U.S. 483 (1954).
137. The "Southern Manifesto" was a document signed by 101 Congressmen from Southern
states. The document maintained that Brown v. Board of Education was an abuse of judicial power,
and expressed its signers' intention to reverse the decision. Pursuant to the Manifesto's purpose, legis-
latures in the South sought to evade integration decrees by, among other things, closing schools, pro-
viding tuition grants for private education, and establishing "freedom of choice" attendance plans. See
H. HOROWITZ & K. KARST, LAW, LAWYERS, AND SOCIAL CHANGE 256 (1969).
138. Governor Faubus of Arkansas had relied on the threats of violence made by groups opposed
to integration to justify a failure to comply with remedial injunctions ordering desegregation. In
Cooper v. Aaron, 358 U.S. 1 (1958), the Supreme Court cut short the governor's attempts to evade the
rule of Brown v. Board of Education, 347 U.S. 483 (1954), holding that desegregation decrees must
be enforced, local hostility notwithstanding. See also R. COLE, CHILDREN OF CRISIS 111, 373 (1967);
A. GRIMSHAW, RACIAL VIOLENCE IN THE UNITED STATES 448, 520 (1969), R. KLUGER, SIMPLE
JUSTICE 223, 747, 754 (1975).
139. See, e.g., D. BATES, THE LONG SHADOW OF LITTLE ROCK 69-76 (1962) (discussing treat-
ment of "Little Rock Nine"); D. GARROW, PROTEST AT SELMA 133-61 (1978); C. KING, MY LIFE
WITH MARTIN LUTHER KING JR. 188-207 (1969) (discussing impact of sit-ins and freedom riders);
M. KING JR., STRIDE TOWARD FREEDOM: THE MONTGOMERY STORY (1958) (discussing successful
boycott of Montgomery bus system).

ciated with the abortion cases 140 includes sophisticated election tactics by
highly visible "right-to-life" and "freedom-of-choice" groups, campaign
rhetoric at every level, and efforts at constitutional amendment' 41 and leg-
142
islative nullification.
When the Justices are right about the moral ideals of the community,
their decisions become settled and accepted. The turmoil, the resistance,
and the threats from other governmental entities, from private groups, in-
stitutions and individuals diminish with time. Thus, few today can be
heard to endorse government supported racial segregation or other state
practices that discriminate against blacks.
When the Court is wrong, criticism and analysis help to reveal the mis-
take-so do the turmoil, the threats, the approval and the resistance, from
all the sources that make up our community. Remember, it is the moral
ideals of the community and not of the wise philosopher that concern the
Court. And it is a wise court that pays attention to the community-not
3
out of fear, but out of obligation. 1
When the Court recognizes that it has made a mistake, it should, in the
appropriate case, rectify the situation. Needless to say, there is no rule in
constitutional law, any more than there is a rule at common law, to tell a
court when to disregard or modify a precedent. The subject is difficult and
no generalization is readily apparent. It is instructive, however, to con-
sider some aspects of the Court's handling of the abortion issue since its
140. For a discussion of the political reaction to Roe v. Wade and Doe v. Bolton, see Rubin, The
Abortion Cases: A Study in Law and Social Change, 5 N.C. CENT. L.J. 215, 247-53 (1974). See also
DeWitt, Abortion Foes March in Capitolon Anniversary of Legislation, N.Y. Times, Jan. 23, 1979,
at C1, col. 1; N.Y. Times, April 24, 1981, at A16, col 1.
141. See, e.g., S.J. Res. 110, 97th Cong., 1st Sess. (1981) (proposed constitutional amendment
permitting Congress and states to legislate with respect to abortion); H.R.J. Res. 372, 97th Cong., 1st
Sess. (1981) (same); S.J. Res. 19, 97th Cong., 1st Sess. (1981) (proposed constitutional amendement
guaranteeing "right-to-life" to unborn); H.R.J. Res. 32, 97th Cong., 1st Sess (1981) (same).
142. See, e.g. S. 158, 97th Cong., 1st Sess. (1981) (declaring human life to begin at conception);
H.R. 900, 97th Cong., 1st Sess. (1981) (same). See also S. 9583, 97th Cong., 1st Sess. (1981) (limit-
ing jurisdiction of federal courts to hear abortion cases); H.R. 73, 97th Cong., 1st Sess. (1981) (same).
143. Is this not the flip side of the following segment of President Lincoln's First Inaugural Ad-
dress, delivered on March 4, 1861?
I do not forget the position assumed by some that constitutional questions are to be decided
by the Supreme Court, nor do I deny that such decisions must be binding in any case upon the
parties to a suit as to the object of that suit, while they are also entitled to very high respect
and consideration in all parallel cases by all other departments of the government. And while it
is obviously possible that such decision may be erroneous in any given case, still the evil effect
following it, being limited to that particular case, with the chance that it may be overruled and
never become a precedent for other cases, can better be borne than could the evils of a different
practice. At the same time, the candid citizen must confess that if the policy of the Government
upon vital questions affecting the whole people is to be irrevocably fixed by decisions of the
Supreme Court, the instant they are made in ordinary litigation between parties in personal
actions the people will have ceased to be their own rulers, having to that extent practically
resigned their Government into the hand of that eminent tribunal.
2 THE COLLECTED WORKS OF ABRAHAM LINCOLN 494 (Basler ed. 1953).

Judicial Review
1973 decision in Roe v. Wade,"" for the Court seems to have some, per-
haps as yet only partially conscious, sense that its decision to permit abor-
tions freely was overly broad, and, in that respect, a mistake. The Court
itself has, indirectly and perhaps unknowingly, retreated in some subse-
quent cases by distinguishing its 1973 holding.
In Maher v. Roe,' 4- decided in 1977, the Court sustained a Connecticut
welfare regulation denying Medicaid recipients payments for medical ser-
vices related to nontherapeutic abortions. And in the summer of 1980, in
Harrisv. McRae,146 the Court upheld the Hyde Amendment's prohibition
on the use of federal Medicaid funds for abortion, "except where the life
of the mother would be endangered if the fetus were carried to term; or
except for such medical procedures necessary for the victims of rape or
incest .... ,47
Since Roe v. Wade had held unconstitutional a statute making it a
crime to "procure an abortion," except to save the life of the mother, 14 8 it
might seem sufficient for the Court to point out, as it did, the "basic dif-
ference between direct state interference with [a woman's freedom of
choice which, under Roe v. Wade, is absolute during the first trimester of
pregnancy] and state encouragement of an alternative activity consonant
with legislative policy,"' 49 namely, carrying the fetus to term. The trouble
with this is that the holding in Roe v. Wade rests on the premise that the
morality of an abortion is none of the government's business, while the
Connecticut welfare regulation and the Hyde Amendment represent gov-
ernmental decisions against a woman's freedom of choice based exclusively
on the proposition that government assistance for all but a limited class of
abortions is immoral. There can be no other explanation for the Connecti-
cut regulation and the Hyde Amendment. Those governmental decisions
were not taken to save money: medical services incident to childbirth are
generally more expensive than abortion. Nor is there any reason to imag-
ine that they were taken because of a direct governmental conunitment to
population growth. 150
The Connecticut welfare regulation and the Hyde Amendment reflect
144. 410 U.S. 113 (1973).

145. 432 U.S. 464 (1977).
146. 448 U.S. 297 (1980).
147. Id. at 301-02.
148. 410 U.S. 117-18.
149. 448 U.S. at 315 (citing Maher v. Roe, 432 U.S. 464, 475-76 (1977)).
150. Cf Perry, Why the Supreme Court Was Plainly Wrong in the Hyde Amendment Case: A
Brief Comment on Harris v. McRae, 32 STAN. L. REV. 1113, 1122-26 (1980) (Hyde Amendment
unconstitutional because "motivated" by view that abortion is immoral). On the proper role of legisla-
tive motivation in constitutional adjudication, see Brest, Palmer v. Thompson: An Approach to the
Problems of Unconstitutional Legislative Motive, 1971 SUP. CT. REV. 95; Ely, Legislative and Ad-
ministrative Motivation in ConstitutionalLaw, 79 YALE LJ. 1205 (1970). See also J. ELY, supra
note 7, at 136-48.

an understanding of the moral ideals of the community different from the

understanding that informs Roe v. Wade. By sustaining the constitutional-
ity of these two provisions, the Court has undermined the foundation for
the rules it announced in Roe v. Wade. It has implied that its earlier
understanding of the moral ideals of the community was mistaken. Roe v.
Wade, accordingly, should be seen as a shaky precedent, and the Court
should see itself as under an obligation to reexamine the breadth of that
1973 decision. 5 '
Indeed, there is reason to believe that this process has begun. In the
Matheson case,"' decided in March of 1981, the Court sustained the con-
stitutionality of a Utah statute at least to the extent that it required a
physician, "if possible," before performing an abortion, to notify the par-
ents of a minor who could not show special circumstances1 3 and was liv-
ing at home. The case involved a fifteen-year old in the first trimester of
pregnancy. She consulted a social worker and a physician. The physician
advised her that an abortion was in her best medical interest. She agreed
but did not wish to notify her parents, a decision with which her social
worker agreed. On these facts it seems that the Utah law burdened the
pregnant young woman's right to choose, with the advice of her physician,
whether to terminate her pregnancy. That is a right that Roe v. Wade
seemed to grant.
Once again, two caveats: First, whatever the Court believed it was do-
ing in the Matheson case, it may not have realized that the Connecticut
and Hyde Amendment cases constituted a retreat from Roe v. Wade. But
if the Court did, it can have no pride in its method. The Court should not
have begun to undo its mistake by sustaining a limitation of choice im-
posed on poor women, without explicitly modifying the 1973 decision.
Second, and more generally, some may say that the views expressed
here foster disrespect for law, because the doctrine of mistake in judicial
review implies that it is legitimate, as a political matter, to resist the deci-
sions of the Supreme Court. While my views, if articulated by the Court,
would be apt to increase political activity both against and for the Court,
151. This does not mean that there is nothing to the distinction between direct state interference
with a woman's freedom of choice and state encouragement of the way in which she exercises that
freedom. It does mean that Roe v. Wade is too broad, given the Court's subsequent conclusion that a
legislature has the constitutional authority to effectuate the proposition that government assistance, for
all but a limited class of abortions, is immoral.
152. H.L. v. Matheson, 450 U.S. 398 (1981).
153. Although the Court assumed that the statute did not apply to "emancipated" or "indepen-
dent" minors, id. at 406-07, this narrowing construction may well fail to constrain the statute's actual
reach. To avoid the parental notification requirement, a minor must first convince her doctor that she
is "emancipated" or "independent." It is not unreasonable to assume that doctors, lacking any clear
standard against which to judge their patients' maturity, and subject to criminal and civil sanctions if
their judgment regarding "emancipation" is later reversed, would err on the side of parental
notification.

Judicial keview
there is as much reason to think that this would be good as to suspect that
it would be bad. Moreover, and more importantly, increased political ac-
tivity over Supreme Court decisions does not constitute disrespect for law.
Our law must be based on consent. It is everyone's obligation to insist that
all branches of government, including the courts, remain true to this cen-
tral understanding."" Peaceful resistance is part of the minorities' arsenal
of weapons against the majority. 55' It is a standard political weapon of
16
minorities and the majority against the Court.
15
It is time to examine where we are and what judicial review looks like
from this position. It comes to this: concern- with judicial review is exag-
gerated if-as is generally the case-the concern focuses on the counter-
majoritarian nature of the practice. Concern is appropriate, however, if
prompted by the apparent finality of constitutional decisions. It is not
helpful, in trying to alleviate this concern, to seek to insulate the Court
from the realm of values by limiting judicial review to diagnosing and
curing the process malfunctions of other governmental entities. That ap-
proach fails to remove the Court from the realm of values and unwisely
excludes it from areas of substantive malfunctions.
The key to mitigating concern with judicial review is found when one
analyzes the concept of finality and relates it to the judicial process. Many
constitutional decisions deal with means and not ends. Accordingly, they
are often less final than might be supposed. Moreover, because value de-
terminations of either the policy or the principle variety are problematic,
judges are apt to make mistakes. But, at least where principles are in-
volved, mistakes can be discovered and mistaken decisions amended by
normal judicial processes. '
154. See H. ARENDT, Civil Disobedience, in CRISES OF THE REPUBLIC 51, 51-57 (1972). See also
A. BICKEL, supra note 8, at 244-72.
155. My views on this are developed to some extent in Wellington, On Freedom of Expression,
88 YALE L.J. 1105, 1136-42 (1979).
156. See supra pp. 1, 515-16. Justice Jackson was surely correct when he said, in Brown v. Allen,
"We are not final because we are infallible . . . ," 344 U.S. 443, 540 (1953) (concurring in the
results), and just as surely incorrect when he continued, "but we are infallible only because we are
final." Id. In fact, the Justices are neither final nor infallible.
157. A remaining question concerns the relationship between the techniques of avoidance, or
"passive virtues," see Bickel, supra note 8, at 111-98, and the arguments concerning the judicial
function presented here. That relationship has been touched upon in the discussion of th delegation,
void-for-vagueness and overbreadth doctrines and in some of the analysis of statutory interpretation.
Moreover, where severe social flux imposes particular difficulty on the judicial task I have described,
the use of techniques of avoidance may seem attractive. In such circumstances the danger of jldicial
mistake may be substantial. But there are difficulties with this approach if carried as far as Bickel
proposes. Quite apart from the familiar criticisms, see Gunther, The Subtle Vices of the "Passive
Virtues'--A Comment on Principle and Expediency in Judicial Review, 64 COLUM. L. REV. 1
(1964), one might ask whether passive avoidance is as non-intrusive as commonly supposed. Deliber-
ate passivity, after all, seeks to shift responsibility by placing burdens on other institutions. Some of

16
Mistake might seem a relatively static concept that should be contrasted
with growth and change or-if one is optimistic-progress in constitu-
tional law. But the distinction is not sharp because mistaken judicial deci-
sions affect the moral ideals of the community in ways that defy general-
ization. And the ensuing political unrest in turn affects the law. At any
rate, the question of growth and change, as a discrete question, has been
much remarked."' 8 If there is anything new to say about it, it would best
be left for another time. Let me only observe, and with this observation
close: growth, change, and progress are inconsistent with finality in any
strong sense; nor is there reason to suppose that they present problems in
understanding the nature of the judicial process that can be ignored in
attempting to understand the nature of judicial review.
the time, and in some situations, those institutions may be ill-suited to perform the additional task.
Bickel, of course, was moved to his position, at least in part, by his anxiety over the countermajori-
tarian nature of judicial review. If one is less worried about this, perhaps one can afford to worry
more about the misallocation of institutional responsibility that may be entailed in the otherwise ques-
tionable avoidance of a constitutional decision.
158. In The Nature of the Judicial Process, supra note 1, Cardozo's central concern is with
growth, change, and progress.

The Yale Law Journal
Volume 91, Number 3, January 1982
David Wippman
Editor-in-Chief
Randall Kennedy Ellen R. Porges
Note & Topics Editor Managing Editor
Richard Bartlett Tim Atkeson
George C. Harris Russell Hayman
Robert Silver Jonathan R. Macey
George Wyeth Theodore A. Paradise
Note Editors Alan 0. Sykes
Article & Book Review Editors
Kathleen Cully Rick Levy Alfred 0. Rose
Robert D. Gilbert David R. Lewis Dietrich L. Snell
Charles Glick John Lewis Hector Torres
Mark A. Goldberg Rory K. Little Lynn A. Stout
Candice Hoke Jonathan E. Richman Monique Witt
Senior Editors
Jose Berrocal Steven Gunby John Motulsky

Gregg Bloche David Hansell Gene Oshman
David Brodsky Doron Henkin Richard Revesz
Steven Calabresi Michele Hirshman Robert Richman
Ludy Cardwell Samuel Issacharoff William Rooney
Robert Cooper, Jr. Michael Jacobs Victoria Rostow
Nicholas De Feis Valerie Lambiase Bennett Rushkoff
Davison Douglas Janet Langford Donald Sparks
Martha Duncan Noel V. Lateef Hubertus van der Vaart
George Ellard Eric Lewis Wendy Warring
Cynthia Estlund Carl Loewenson, Jr.
Editors
Business Manager:Pamela Standish EditorialAssistant: Liz Doyle
Student Contr'butorsto This Issue

Charles Glick, Free Speech, the Private Employee, and State
Constitutions
John Lewis, The Open-Ended Investigation:A Method for Regulation of
New Medical Services
Ken Taymor, The Military and State Secrets Privilege: Protection for the
National Security or Immunity for the Executive
Hector Torres, Conflicts in Class Actions and Protection of Absent Class
Members
David Wippman, Worker Ownership and Section 8(a)(2) of the National
Labor Relations Act

Nature of Judicial Review The

Uploaded by

Copyright:

Available Formats

Nature of Judicial Review The

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Nature of Judicial Review The

Uploaded by

Copyright:

Available Formats

The Nature of Judicial Review*

In The Nature of the JudicialProcess,' Cardozo is concerned with the

HeinOnline -- 91 Yale L.J. 486 1981-1982

several of our prominent constitutional scholars,' even as it did an earlier

HeinOnline -- 91 Yale L.J. 487 1981-1982

other nonmajoritarian decisions of consequence. Each of these issues,

10. See J. RAWLS, A THEORY OF JUSTICE 221-28 (1971).

HeinOnline -- 91 Yale L.J. 488 1981-1982

HeinOnline -- 91 Yale L.J. 489 1981-1982

HeinOnline -- 91 Yale L.J. 490 1981-1982

HeinOnline -- 91 Yale L.J. 491 1981-1982

HeinOnline -- 91 Yale L.J. 492 1981-1982

many majoritarian proposals that are carefully fashioned, incremental so-

HeinOnline -- 91 Yale L.J. 493 1981-1982

43. Id. at 245.

HeinOnline -- 91 Yale L.J. 494 1981-1982

tracts-absent a precise acceptance by the offeree of each aspect of the

The statute is not too happily drafted. Perhaps it would be wiser in

By misconstruing the Code in this manner, 4 the court in Roto-Lith de-

HeinOnline -- 91 Yale L.J. 495 1981-1982

Textile Workers Union v. DarlingtonManufacturing Co." is a differ-

HeinOnline -- 91 Yale L.J. 496 1981-1982

HeinOnline -- 91 Yale L.J. 497 1981-1982

HeinOnline -- 91 Yale L.J. 498 1981-1982

It is unnecessary to consider now whether legislation which restricts

Footnote four can be seen as "a participation-oriented, representation-

HeinOnline -- 91 Yale L.J. 499 1981-1982

reinforcing approach to judicial review,"" and, as such, it is wonderfully

Malfunction occurs when the process is undeserving of trust, when

Our government cannot fairly be said to be 'malfunctioning' simply

Ely's attempt to withdraw value determinations from the courts results

74. J. ELY, supra note 7, at 87.

HeinOnline -- 91 Yale L.J. 500 1981-1982

HeinOnline -- 91 Yale L.J. 501 1981-1982

HeinOnline -- 91 Yale L.J. 502 1981-1982

HeinOnline -- 91 Yale L.J. 503 1981-1982

that the organization chosen to represent . . . is to represent all . . . and

It is a principle of general application that the exercise of a granted

HeinOnline -- 91 Yale L.J. 504 1981-1982

doctrines, procedural or structural in 97nature, such as "delegation,""5 "void

HeinOnline -- 91 Yale L.J. 505 1981-1982

procure services or supplies from businesses owned by United States citi-

The Justice went on to describe this as a "malfunction of the legislative

Whenever Congress creates a classification that would be subject to

Justice Stevens' approach invites attention to separation of powers

102. 42 U.S.C. § 6705(0(2) (1976 & Supp. I 1978).

HeinOnline -- 91 Yale L.J. 506 1981-1982

structing the legislature on its internal procedures. Thus, Justice Stevens'

107. Miranda v. Arizona, 384 U.S. 436 (1966).

HeinOnline -- 91 Yale L.J. 507 1981-1982

involved a New York City traffic regulation that barred advertising on

110. Id. at 112.

HeinOnline -- 91 Yale L.J. 508 1981-1982

HeinOnline -- 91 Yale L.J. 509 1981-1982

115. 376 U.S. 254 (1964).

HeinOnline -- 91 Yale L.J. 510 1981-1982

The makers of our Constitution undertook to secure conditions

Of course, if Brandeis is "doing" history, he is attributing a great deal to