Rule of Law

1
Cardozo Law School

Jacob Burns Institute for Advanced Legal Studies
March 2001
Working Paper Series No. 36
THE RULE OF LAW, AND THE

LEGITIMACY OF
CONSTITUTIONAL DEMOCRACY
By Michel Rosenfeld
This paper can be downloaded without charge from the

Social Science Research Network Electronic Paper Collection:
http://papers.ssrn.com/paper.taf?abstract_id=262350
a: RULE OF LAW.01 March 27, 2001

2
THE RULE OF LAW, AND THE

LEGITIMACY OF
CONSTITUTIONAL DEMOCRACY
By Michel Rosenfeld *
I. Introduction
The rule of law is a cornerstone of contemporary constitutional democracy1 as
underscored by its paramount role in cementing the recent transitions from authoritarian or
totalitarian regimes2 to constitutional democracy in Eastern Europe and elsewhere3. In the
broadest terms, the rule of law requires that the citizenry be subjected only to publicly
promulgated laws, that the legislative function be somewhat kept separate from the
adjudicative function, and that no one within the polity be above the law. Moreover, in the
absence of the rule of law, contemporary constitutional democracy would seem altogether
impossible because it would lack one of the three essential characteristics of modern
*
Justice Sydney L. Robins Professor of Human Rights, Benjamin N. Cardozo School
of Law.
See Michel Rosenfeld, “Modern Constitutionalism and the Interplay Between Identity
1
and Diversity” in CONSTITUTIONALISM, IDENTITY, DIFFERENCE AND LEGITIMACY:

THEORETICAL PERSPECTIVES (M. Rosenfeld, ed., 1994) 3.
2
For a discussion of the difference between authoritarian and totalitarian regimes, see
Juan J. Linz and Alfred Stepan, PROBLEMS OF DEMOCRATIC TRANSITION AND
CONSOLIDATION ch. 3 (1996).
See, e.g., Mark F. Brzezinki and Leszek Garlicki, “Judicial Review in Post-Communist
3
Poland: The Emergence of a Rechtsstaat?”, 31 Stanford J. of Intl. L. 35 (1995); “Special

Report: Retroactivity Law Overturned in Hungary” 1 East European Constitutional Review,
n. 1,7-8 1992; and Krisztina Morvai, “Retroactive Justice Based on International Law: A
Recent Decision of the Hungarian Constitutional Court”, 3-4 East European Constitutional
Review 32-34 (1993-4).
3
constitutionalism, these being: limiting the powers of government, adherence to the rule of law,
and protection of fundamental rights4. Thus, the rule of law must figure in constitutional
democracy as an indispensable ingredient of constitutionalism. Beyond that, however, it is
not clear what precise characteristics the rule of law must possess to help sustain
constitutional democracy or what specific role it must assume to ensure a working
constitutional democracy, or how it might ultimately contribute to the latter’s legitimacy.
Although it may be widely believed that rule of law and constitutional democracy go
hand in hand closer scrutiny raises several vexing questions. For one thing, constitutionalism
and democracy may not always be in harmony5, with the consequence that the rule of law may
clash or come in tension with democracy. For another, the proper role and scope of the rule
of law within constitutionalism is itself ambiguous inasmuch as the rule of law may ultimately
spill over to the other two essential features of constitutionalism rather than figuring exclusively
as one of the three. Indeed, a written constitution may have the force of law6, and thus its
provisions limiting the powers of government and those devoted to the protection of
fundamental rights may become part and parcel of the rule of law regime instituted by the
relevant constitutional regime. Moreover, the rule of law may encompass the entire field
4
See Michel Rosenfeld, supra note 1, at 3.
5
This is made manifest by the famous “counter-majoritarian” difficulty discussed below.
See infra, at —. While this difficulty is most often considered in the context of judicial
interpretation of constitutional rights, it can also he raised in connection with at least some
such rights themselves.
See e.g., Marbury v. Madison, 5 U.S. 137 (1803) (treating the constitution as legally
6
enforceable and as the highest law of the land).

4
coming within the sweep of constitutionalism7 or it may only play a limited role in the
maintenance of a prescribed constitutional order8.
The above difficulties are compounded because there is no consensus on what “the
rule of law” stands for, even if it is fairly clear what it stands against. An important part of the
problem stems from the fact that “the rule of law” is an “essentially contestable concept”, that
is a concept with both descriptive and prescriptive content over which there is a lack of
widespread agreement9. Just as in the case of concepts such as “liberty” or “equality”, “the
rule of law”’s descriptive meaning is dependent on the prescriptive meaning one ascribes to
it, and typically in the context of complex contemporary polities there are likely to be vigorous
disagreements concerning the relevant prescriptive standards at stake10. Consistent with
7
This would occur if the constitution defines the nature and scope of governmental
powers and human rights and if these become implemented through laws or judicially
interpreted and elaborated as legal norms.
8
This would be the case if the constitution established a division of powers, but left it
to the political process to implement it, c.f. Garcia v. San Antonio Metropolitan Transit
Authority, 469 U.S. 528 (1985) (indicating that disputes concerning the boundaries between
federal and state powers in the United States should be left to the national legislature as state
interests are incorporated in the very composition of that legislature), or if it singled out a set
of fundamental rights for special protection, leaving their actual implementation to the
discretion of the political branches of government as opposed to the judiciary.
See Richard H. Fallon, Jr., “The Rule of Law“ as a Concept in Constitutional

9
Discourse”, 97 Colum. L. Rev. 1, 7 (1997).

10
While it is important for analytic purposes to keep the prescriptive and descriptive
aspects of essentially contestable concepts apart, no conception of such concepts can be
cogent unless it encompasses both prescriptive and descriptive elements. For example, it
is impossible to determine in the abstract whether mechanical application of rigid legal rules
amounts to implementation of “the rule of law”. If the relevant prescriptive objective is to insure
predictability, then the answer could well be in the affirmative. However, if the main
prescriptive aim were to insure fairness, then mechanical adherence to rigid rules would most
5
this, the rule of law has come to mean different things to various legal traditions as evinced by
the contrasts between Anglo-American “rule of law”, German Rechtsstaat and French État de
droit11. Moreover, even within a single tradition, it is not clear whether the rule of law ought to
be largely, if not exclusively, procedural or substantive, or whether it should be primarily
concerned with predictability rather than with fairness. Finally, at least in the context of
constitutional democracy the rule of law appears to rest on a paradox. In terms of the
institutional framework necessary for constitutional democracy, and of implementation of the
will of the majority through law, the rule of law seems definitely on the side of the state, and
often against the citizen12. In contrast, in connection with protection of fundamental
constitutional rights, the rule of law seems on the side of the citizen against the state to the
extent that constitutional law can be invoked by citizens against laws and policies of the state.
In order to determine whether, and how, the rule of law might contribute to establishing
the legitimacy of constitutional democracy in a contemporary pluralistic society, it is necessary
to tackle the above questions as well as certain key preliminary issues relating to the very
concepts of “rule of law,” “legitimacy” and “pluralistic society”. I shall briefly deal with these
preliminary issues in the next section. Then, I shall concentrate on the essential jurisprudential
characteristics of the respective conceptions of the rule of law in three different legal traditions.
likely fail to live up to the requirements of the rule of law.

11
See the discussion infra, at ---for an account of the differences among these.
12
This seems to be the case whenever a citizen is coerced to abide by laws which he
or she deems unjust or oppressive notwithstanding that such laws were approved by duly
elected legislative majorities. For further exploration of this issue, see infra, at ---.
6
In Section III, I will discuss the German conception of the Rechtsstaat; in section IV, the French
notion of the État de droit; and, in Section V the Anglo-American common law based
elaboration of the idea of “the rule of law”. In Section VI, I will explore the conditions which
the rule of law would have to fulfill on order to legitimate constitutional democracy in a
pluralistic society. I will focus particularly on the contrast between procedural and substantive
safeguards and on the divide between law and politics, with special reference to the tendency
by proponents of Critical Legal Studies (CLS) to collapse law into politics. In Section VII, I will
examine how the rule of law, particularly in common law jurisdictions, might reconcile the need
for predictability with that for fairness. Finally, after concluding that it is impossible for the rule
of law to satisfy the requisite conditions to the legitimation of constitutional democracy, I will
focus in the Section VIII on what role the rule of law might nonetheless play in the quest for such
legitimation.
II. Background Presuppositions:

Pluralism, Legitimacy as Consent,
and the Minimum Conditions for the
Rule of Law.
The belief that constitutional democracy under the rule of law is desirable does not
occur in a vacuum. As already mentioned, constitutional democracy looms as virtually
indispensable when contrasted to the totalitarian or authoritarian regimes which it replaced
in the second half of the twentieth century. Indeed, these regimes to various extents ignored,
violated or abused the rule of law (under almost all plausible conceptions of it), and those
subjected to them were clearly the worse for it. It does not necessarily follow from this ,
7
however, that constitutional democracy under the rule of law is always indispensable or the
best alternative in all circumstances. Indeed, one can think of cases in which constitutional
democracy could be superfluous, or even undesirable. For example, in a close knit
homogeneous society that is deeply religious and ruled by revered leaders who are widely
believed to have direct access to divine commands, a theocracy would plainly seem more
appropriate than a constitutional democracy. Furthermore, in such a society instructions and
directions imparted by the religious leaders would be paramount, leaving little, if any room for
the rule of law.
In contrast, in heterogeneous societies with various competing conceptions of the
good, constitutional democracy and adherence to the rule of law may well be indispensable
for purposes of achieving political cohesion with a minimum of oppression13. Such
heterogenous societies, moreover, can be characterized as being pluralistic - in-fact14. A
society is pluralistic-in-fact if it is divided along ethnic, religious, linguistic, cultural or
ideological lines, or, in other words, if it is comprised of different groups we do not share the
same values or conceptions of the good. Moreover. even a homogeneous society with an
individualistic conception of the good, in which every person is viewed as entitled to pursue
13
This is provided that the societies in question possess -- or are within striking
distance of -- the means necessary to sustain a working constitutional democracy. See
PROBLEMS OF DEMOCRATIC TRANSITION, supra note ---, at 7-11 (identifying five arenas
of “consolidated democracy”, to wit: 1) civil society; 2) a relatively autonomous political
society; 3) implementation of the rule of law; 4) a viable state bureaucracy; and 5) an
“institutionalized economic society”.
14
For a more extended discussion of “pluralism-in-fact” and of the contrast between it
and “pluralism-as-norm,” see Michel Rosenfeld, JUST INTERPRETATIONS: LAW BETWEEN
ETHICS AND POLITICS 200 - 03 (1998).
8
his or her own individual good, would qualify as being pluralistic-in-fact15. In short, a society
is pluralistic-in-fact either if it is pluralistic at the group level or the individual level. And,
consistent with this, all Western democracies, and for that matter most contemporary nation-
states qualify as being pluralistic-in-fact.
Because everyone does not share the same values or interests in societies that are
pluralistic-in-fact, the legitimacy of fundamental political institutions must ultimately depend on
some kind of consent among all those who are subjected to such institutions. There is a long
standing tradition that conceives institutional legitimacy and political justice in terms of
consent. That tradition is the one established by social contract theory as articulated in the
philosophies of Hobbes, Locke, Rousseau, Kant16, and, more recently, Rawls17. In the
broadest terms, under social contract theory, the legitimacy of government depends on the
consent of the governed. However, there is no agreement among social contract theorists,
as to what constitutes adequate consent. For some, like Locke it seems to be the actual
15
More precisely, such a society would not be pluralistic at the group level, but would
qualify as pluralistic at the individual level to the extent that pursuit of individual self-interest
were deemed normatively acceptable or desirable. Accordingly, such a society would be
pluralistic-in-fact as contrasted with a homogeneous society in which individuals are
normatively required to sacrifice private interests to the common good.
16
For a comparison of the similarities and differences among the respective social
contract theories espoused by these philosophers, see Michel Rosenfeld, “Contract and
Justice: The Relation Between Classical Contract Law and Social Contract Theory,” 70 Iowa
L. Rev. 769 (1985).
17
See John Rawls, A THEORY OF JUSTICE (1971).
9
consent of the governed18, while for others, like Rawls, a hypothetical consent on the basic
institutions of society suffices19. Furthermore, consent is also the basis for legitimacy in the
context of theories that do not strictly speaking fall within the social contract paradigm but
which nonetheless bear great affinity with it, such as Habermas’s consensus based discourse
theory of the justification of law20.
Leaving aside differences among various consent-based theories, legitimation based
on consent appears to be the optimal -- if not the exclusive -- means of normative justification
for both constitutional democracy and the rule of law in the context of societies that lack a
consensus on the good and are hence pluralistic-in-fact. Indeed, notwithstanding its
unmistakable attractiveness when contrasted with authoritarian or totalitarian regimes,
constitutional democracy can itself be oppressive as it generally imposes at least two kinds
of coercion. To the extent that it is democratic, constitutional democracy implements the will
of political majorities and coerces political minorities to contribute to the realization of majority
objectives with which such minorities may strongly
18
See John Locke, THE SECOND TREATISE OF GOVERNMENT §§100-102 (J.
Gough, ed. 1976).
19
See John Rawls, A THEORY OF JUSTICE, supra note ---, at 11-13.
See Jürgen Habermas, BETWEEN FACTS AND NORMS: CONTRIBUTIONS TO

20
A DISCOURSE THEORY OF LAW AND DEMOCRACY 118-31 (William Rehg, trans., 1996).
For an analysis of Habermas’s theory as the heir to social contract theory, see Michel
Rosenfeld, “Can Rights Democracy, and Justice Be Reconciled through Discourse Theory?
Reflections on Habermas’s Proceduralist Paradigm of Law” in HABERMAS ON LAW AND
DEMOCRACY: CRITICAL EXCHANGES 82 (Michel Rosenfeld and Andrew Arato eds.,
1998).
10
disagree 21. On the other hand, to the extent that constitutional democracy affords
protection to certain fundamental rights, and that certain vindications of such rights frustrate
the will of majorities in relation to certain objectives which they consider paramount, the
enforcement of constitutional rights would seem to lead to a significant amount of coercion22.
For its part, the rule of law itself is coercive inasmuch as citizens are subjected to laws
with which they disagree or which they find oppressive. Also, the above mentioned types of
coercion associated with constitutional democracy are likely to be imposed largely through
implementation of the rule of law. Specifically, infra-constitutional laws justify state backed
coercion of citizens, whereas constitutional laws allow citizens to challenge state backed infra-
constitutional laws, and may upset the pursuit of the good of the promoters of successfully
challenged infra-constitutional laws.
Consent to constitutional democracy and the rule of law may not eliminate coercion but
it would legitimate it. Presumably, the consent in question would be, in all relevant respects,
akin to consent in contract23. In the context of a valid legal contract, uncoerced agreement
21
Although not all policies designed to placate the majority need to be coercive from
the standpoint of political minorities, many of them undoubtedly are. To cite but two strong
examples, a citizen may certainly feel coerced if he or she must serve in the military to fight
a war which he or she strongly opposes; or if he or she must contribute a hefty sum to pay a
tax earmarked to finance policies which he or she deems in square contradiction with the
dictates of his or her conception of the good.
22
For example, striking down a widely supported antipornography law as violative of
freedom of expression rights, may well frustrate decency and moral objectives considered by
their proponents as vital to the preservation of their way of life.
23
For a discussion of the similarities and differences between legal contract and social
contract, see Michel Rosenfeld, “Contract and Justice,” supra note ---.
11
between the parties at the time of making the contract legitimates subsequent enforcement
of such contract even against a party who has come to regret his or her original agreement.
Thus, although someone who has changed his or her mind after having entered into a valid
contract may feel that enforcement of such contract is coercive, such coercion would still be
legitimate, and would derive in all relevant respects from valid consent rather than being
merely oppressive. Similarly, if constitutional democracy and the rule of law can be genuinely
legitimated on the basis of some plausible notion of consent, the mere fact that they may also
be experienced as coercive would not necessarily impair their legitimacy.
What is crucial in contract is that the consent be ex-ante, before the legally binding
transaction is set in motion. This is because if would-be contractors know what to expect in
case the contract goes forward, and if they nonetheless choose to go ahead with it, then
subsequent enforcement of such contract against them would be both fair and consistent with
respect for their freedom and autonomy. And assuming that consent would analogously
validate constitutional democracy, and that the rule of law would play an important rule in
making the consequences of ex ante commitment predictable, then the citizenry’s consent to
the rule of law would be a key factor in the legitimation not only of rule of law regime at stake,
but also in that of the particular constitutional democracy associated with it.
Before determining what the rule of law is, or what its potential may be, it is important
to note briefly what it is not, or in other words, to elucidate what are its minimum

12
requirements24. The “rule of law” is often contrasted to the “rule of men” 25. While the “rule of
men” (or, we might say today, “the rule of individual persons”)generally connotes unrestrained
and potentially arbitrary personal rule by an unconstrained and perhaps unpredictable ruler,
it will be understood here more broadly. For present purposes, even rule through law will be
deemed to amount to the “rule of men”, if the law can be changed unilaterally and arbitrarily,
if it is largely ignored, or if the ruler and his or her associates consistently remain above the
law. At a minimum, therefore, the rule of law requires fairly generalized rule through law; a
substantial amount of legal predictability -- through generally applicable, published and largely
prospective laws; a significant separation between the legislative and the adjudicative
function; and, widespread adherence to the principle that no one is above the law. Consistent
with this, any legal regime which meets these minimal requirements will be considered to
satisfy the prescriptions of the “rule of law in the narrow sense”.
The rule of law In the narrow sense may be preferable to the rule of men26, but is highly
insufficient for purposes of satisfying the minimum requirements of a legitimate constitutional
See Richard H. Fallon, Jr. “The Rule of Law” as a Concept in Constitutional

24
Discourse”, supra note ----- at 1 (“The Rule of Law is a much celebrated, historical ideal, the
precise meaning of which may be less clear today than ever before.”).
See Marbury v. Madison, 5 U.S. 137, 163 (1803) (contrasting a “government of laws”
25
to a “government of men”).
26
It is certainly conceivable that the members of a polity would be better off under the
unfettered rule of a benevolent monarch deeply committed to ruling in a just and
compassionate manner. Nevertheless, as a general rule it would seem preferable to be
subjected to a publicly implemented and largely predictable legal regime than to be at the
mercy of the whim of a single ruler or of a collective leadership bound by no law.
13
democracy. Indeed, the rule of law in the narrow sense need not be just or even democratic
as it is entirely compatible with legal regimes predicated on slavery, apartheid or countless
other oppressive and dehumanizing practices and policies grounded in law, shaped by law,
and carried out through law27. Accordingly, for the rule of law to measure up to the
requirements of a legitimate constitutional democracy, it must be more than the rule of law in
the narrow sense.
To become legitimate the rule of law would seem to have to be democratically
accountable, procedurally fair and even perhaps substantively grounded. Again, satisfying
these requirements may be necessary without being sufficient to produce legitimacy.
Democratic laws may be oppressive to minorities, procedural fairness may be consistent with
a significant measure of substantive inequity, and the substantive values vindicated by any
particular instantiation of the rule of law may be rejected by a sizeable portion of the polity,
particularly in pluralist settings marked by clashing conceptions of the good.
Also as already noted, a further difficulty stems from the split within the rule of law in
a constitutional democracy. Moreover, even if we assume that as shields against ordinary
laws constitutional rights command greater support among the citizenry than most ordinary
laws, there would still be constitutional rights opposed by some of the citizens -- or, at the very
The Dred Scott case furnishes a particularly apt illustration of this point as the
27
Supreme Court managed to enshrine a legally grounded property right of a slave owner in his
slave as a constitutional right in the course of resolving a conflict between state law and
federal law under the American Constitution. Dred Scott v. Sanford, 60 U.S. 393 (1857)
(federal law providing for emancipation of slave brought by his master to federal territory held
unconstitutional as deprivation of master’s state created property right in his slave without the
“due process of law” guaranteed by the Fifth Amendment to the Constitution).
14
least, some binding interpretations or applications of even the most popular constitutional
rights would still be opposed by some. For example, for those Americans who favor state aid
to parochial school, the Supreme Court’s interpretation of the Establishment Clause as
forbidding such aid28 might well seem as oppressive as the state prohibition against
contraceptives did to those who challenged it as violative of their constitutional privacy rights29.
Consistent with this, in a constitutional democracy all laws are prone to being considered
advantageous or acceptable by some and oppressive or coercive by others.
The fact that if genuinely subjected to the consent of the citizenry all laws -- whether
constitutional or ordinary -- would be approved by some but rejected by others seems to erect
a formidable barrier to the legitimation of the rule of law in a constitutional democracy. It may
be objected, however, that the lack of full consent is less daunting than it seems for it appears
ultimately reducible to the familiar conflict between majoritarian laws and antimajoritarian
constitutional constraints much debated in American constitutional circles30. While this debate
See Everson v. Board of Education, 330 U.S. 1 (1947) (holding that state cannot
28
finance parochial schools). More recent decisions, however, have upheld certain indirect or
incidental state aid to parochial schools. See, e.g., Agostini v. Felton, 521 U.S. 203 (1997)
(public school teachers may impart remedial education to disadvantaged parochial school
students).
29
See Griswold v. Connecticut, 381 U.S. 479 (1965).
30
This conflict around what has been labeled in the United States as the “counter-
majoritarian problem” has generated an immense literature. Two leading works are
Alexander Bickel, THE LEAST DANGEROUS BRANCH: THE SUPREME COURT AT THE
BAR OF POLITICS (1962) and John Hart Ely, DEMOCRACY AND DISTRUST: A THEORY
OF JUDICIAL REVIEW (1980).
15
has been quite vehement at times31, it seems to boil down to a conflict over how broadly or
narrowly antimajoritarian constitutional constraints ought to be construed without questioning
their inherent legitimacy. Moreover, so long as some constitutional constraints are legitimate,
it would seem that so is majoritarian law making. In other words, if there is a consensus
concerning certain fundamental constitutional constraints and a shared commitment to
democracy, then mere legislative setbacks or even the subjection to certain unfair yet
constitutional laws should not pose a serious challenge the legitimacy of the prevailing rule of
law regime32.
As long as the proper measure of legitimacy is the consent of each and every citizen33
then the above objection must ultimately fail. Even if there were a consensus on some
constitutional constraints, it seems highly implausible that in any pluralist constitutional
democracy there would be unanimity on a sufficient core of constitutional fundamentals to
directly or indirectly legitimate the rule of law all the way down. There is certainly no such
unanimity in the United States as evinced by deep splits over key constitutional issues ranging
See, e.g., Robert Bork, THE TEMPTING OF AMERICA: THE POLITICAL

31
SEDUCTION OF LAW (1990) for a scathing attack on the U.S. Supreme Court for failure to
remain faithful to the “original understanding” of the text of the Constitution.
32
Cf. John Rawls, POLITICAL LIBERALISM 11, 237-40 (1993) (framing political justice
in terms of the “basic structure” of society and “constitutional essentials”).
33
I restrict the test of legitimacy to citizens rather than persons to avoid certain
theoretical issues relating to immigrants and would be immigrants. Of course, if the test of
legitimacy fails for citizens within the polity, it would seem a fortiori to fail for those altogether
excluded from representation.
16
from federalism34 to abortion35, affirmative action36 equality for women37 or homosexuals 38
and the relationship between state and religion39. And there seems to be even less
agreement in other constitutional democracies divided along ethnic,
See e.g., United States v. Lopez 514 U.S. 549 (1995) (5-4 decision restricting scope
34
of federal commerce power).

35
See Lawrence Tribe, ABORTION: THE CLASH OF ABSOLUTES (2nd ed. 1992).
See Michel Rosenfeld, AFFIRMATIVE ACTION AND JUSTICE: A PHILOSOPHICAL

36
AND CONSTITUTIONAL INQUIRY(1991).
See, e.g., Martha Minow, “ Justice Engendered”, 101 Har. L. Rev. 10 (1987) (arguing
37
that Supreme Court adjudication of sex discrimination cases posits men’s experience as the
“norm” against which women are measured).
See Bowers v. Hardwick, 478 U.S. 186 (1986) (5-4 decision upholding
38
criminalization of consensual sex among homosexual adults).
See, e.g., Agostini v. Felton 521 U.S. 203 (1997) (5-4 decision repudiating previous
39
limitations on state assistance to parochial schools), and Employment Division, Dept. of

Human Resources v. Smith, 409 U.S. 872 (1990) (divided court narrows scope of protection
of religious liberty). It may be objected that notwithstanding these splits, Americans generally
agree that there ought to be divided government and liberty and equality for all. Assuming that
this is true, such consensus would remain at such an abstract level as to preclude any
meaningful legitimation of the rule of law. For example, so long as adherence to “equality for
all” is not deemed inconsistent with criminalizing homosexual sex, it seems reasonable for
homosexuals to challenge the legitimacy of a rule of law regime that deprives them of
fundamental equality and privacy rights accorded to their heterosexual fellow citizens.
17
linguistic or religious differences such as Canada,40 or Spain41.
It follows from these considerations that actual unanimous consent for any meaningful
constitutional constraints, let alone for any rule of law regime, seems highly implausible. Such
actual consent, however, is unnecessary. It is arguably sufficient for purposes of assessing
the legitimacy of a rule of law regime to determine whether acceptance of the latter would be
reasonably consistent with the diverse agendas of all concerned. Accordingly, I propose using
a test fashioned after Habermas’s criterion for the legitimacy of law. According to Habermas
the legitimacy of law can be established dialogically through communicative action among
persons who recognize each other as equals and who agree to accept as legitimate only
those laws which they would all consent both to enact as autonomous legislators and to follow
as law abiding citizens.42 This test allows for reconstruction on the basis of a counterfactual
in order to establish the legitimacy of law, and is used by Habermas to elaborate and defend
40
Quebec has thus far refused to accept the legitimacy of the 1982 Canadian
Constitution, see Jacques Pariseau, “The Case for a Sovereign Quebec”, Foreign Policy,
Summer 1995, at 69, and has come close to moving towards secession. Thus, in 1995 a
provincial refendum on Quebec sovereignty failed to endorse secession by a very slim margin.
See Rogers Worthington “50.5% - 49.5% unity wins in Quebec”, Chicago Tribune Oct. 31,
1995, at 1.
41
See, e.g. Jon Elster, “Forces and Mechanisms in the Constitution-Making Process,”
45 Duke L. J. 364, 393 (1995) (observing that Spain’s 1978 constitution was made in the
context of Basque terrorism), and Charles E. Ehrlich, “Ethno-cultural Minorities and
Federalism: Is Spain Instructive?”, 24 S. Ill. U.L.J. 291, 305 (2000) (Basque nationalists chose
in substance to stay out of process of constitution-making).
See Jürgen Habermas, “Paradigms of Law” in HABERMAS ON LAW AND

42
DEMOCRACY: CRITICAL EXCHANGES 19-21 (Michel Rosenfeld and Andrew Arato eds.,
1998).

18
his “proceduralist paradigm of law”43. As I have discussed elsewhere, however, I do not share
Habermas’s belief that law can be legitimated on purely procedural grounds44. I therefore
propose to adapt Habermas’s test to account for this key difference. More specifically, I
intend to rely on two modifications which somewhat weaken the conditions of legitimacy
envisaged by Habermas: 1) I will consider the counterfactual requirement of self-legislation
coupled with willing submission to law to be satisfied if it can be used to legitimate a rule of
law regime taken as a whole without separately legitimating individual laws within that regime;
and 2) I will construe the requirement of consent more loosely so as to include within it a
criterion of reasonableness based on lack of coercion coupled with the meeting of certain
conditions which make it reasonable to endorse a particular rule of law regime consistent with
one’s substantive aims.
Although these proposed modifications make the requirements of legitimacy weaker
than those endorsed by Habermas, it is by no means obvious that any contemporary rule of
law regime would actually be able to measure up even to the lesser of these criteria.
Because of this, I will not seek to justify adoption of a weaker test before determining what
kind of rule regime could satisfy it. Moreover, before tackling this latter task, it would seem
useful to get a better handle on different traditions relating to the rule of law. Accordingly, I first
turn briefly to these traditions, then attempt to determine whether any rule of law regime is
43
Id., at 13.
See Michel Rosenfeld, “Can Rights, Democracy, and Justice Be Reconciled through
44
Discourse Theory? Reflections on Habermas’s Proceduralist Paradigm of Law” in

HABERMAS ON LAW AND DEMOCRACY, supra, note ...., at 82.
19
likely to meet the test I have set above, and finally assess whether a legitimate rule of law
regime is possible or merely conceivable as a useful counterfactual.
III The German Rechtsstaat as State

Rule though Law
Any attempt to put more flesh and bones on the concept of the rule of law should be
mindful that diverse conceptions of the rule of law have taken root in different traditions. A
brief comparison among these traditions will allow for a better grasp of certain key nuances
concerning the rule of law and thus make it easier to appreciate its scope and limitations with
a view to testing its legitimacy in the context of a pluralist constitutional democracy.
Accordingly, I shall focus on the salient differences among the three major traditions that have
given shape to the rule of law, namely the German, French, and Anglo- American one.
The German Rechtsstaat, the French État de droit and the corresponding British and
American conceptions all endorse the rule of law in the narrow sense, but otherwise diverge
significantly from one another45. Moreover, some of these four traditions are more ancient
than others, and all of them have evolved over the years; though some of them have done so
more than others. Since my primary focus is conceptual and directed to issues of legitimacy
in pluralist settings, I shall only refer to the history and evolution of these traditions with an eye
to gaining further insights into the potential, limitations, and multiple dimensions of the rule of
45
The rule of law in the narrow sense has a much more ancient pedigree than the
traditions being considered as it dates at least as far back as Aristotle. See his POLITICS
bk. III, 15-16.
20
law.
The Rechtsstaat is often treated as the German equivalent to the concept of the rule
of law in the Anglo-American tradition46. Both concepts certainly share certain important
elements in common. Chief among these, is the relationship between the state and the
institutionalization of a legal regime or, in other words, the state’s duty to wield its power
through laws and in accordance with fundamental principles of legality--including consistent
implementation of publicly disseminated generally applicable rules giving citizens notice
regarding what conduct is subject to legal sanctions coupled with fair procedural safeguards.
Beyond that, however, the two concepts differ significantly, particularly in terms of their
understanding of the relationship between the state and the law. Whereas as we shall see,
the American conception of the rule of law is rooted in a somewhat antagonistic relationship
between the state and the rule of law--which gives prominence to the above noted paradox
47
between the law as dependent on, and independent from, the state -- its German
counterpart is squarely predicated on a veritable symbiosis between the law and the state.
In the broadest terms, in the Rechtsstaat law becomes inextricably tied to the state as the only
legitimate channel through which the state can wield its power. Accordingly, “state rule through
law” would be a much better approximation in English for “Rechtsstaat” than “rule of law”.
If any state rule through law would do then the Rechtsstaat would boil down in essence
46
See, e.g., Edward Eberle, “Human Dignity, Privacy and Personality in German and
American Constitutional Law”, 1997 Utah L. Rev. 963, 967.
47
See Supra, at --.
21
to little more than the rule of law in the narrow sense. Actually, though the concept has
significantly evolved since its implantation in the nineteenth century, the Rechtsstaat has
always stood for much more than the rule of law in the narrow sense. The Rechsstaat, which
had its intellectual origins in Kant’s theory 48, stood in the first half of the nineteenth century for
rational state rule encompassing universal protection of formal rights for every individual within
the ambit of a unified legal order crafted by legislation and administered through a separate
and independent process of adjudication. In contrast to American notions of separation of
powers, so long as legislation was kept separate from adjudication, the nineteenth century
Rechsstaat was equally compatible with a government (as opposed to a staat) that was
monarchic as with one that was democratic49.
As it evolved from its Kantian roots toward more positivistic configurations in
Bismarck’s late nineteenth century Germany, the Rechsstaat became increasingly tied to
issues of form rather than substance 5 0 . What binds together both the Kantian and the
positivistic conceptions of the Rechtsstaat, however, is the rejection of older notions which
anchored the state’s legitimacy in the pursuit and implementation of transcendental religious
48
See Hans Reiss, KANT’S POLITICAL WRITINGS 11 (1991).
See Donald Kommers, THE CONSTITUTIONAL JURISPRUDENCE OF THE

49
FEDERAL REPUBLIC OF GERMANY 36 (2d. ed. 1997).
See Rainer Grote, “Rule of Law, Rechtsstaat and État de Droit” in

50
CONSTITUTIONALISM, UNIVERSALISM AND DEMOCRACY: A COMPARATIVE

ANALYSIS 269, 278-81 (Christian Starck, ed. 1999).
22
or ethical values51. Accordingly, the Rechtsstaat opened the door to a state rule-through law
that could function properly without having to rely on a value system derived from any particular
religion or transcendental conception of ethics. In other words, the Rechtsstaat made
possible the systematic deployment of a legal regime poised to accommodate a plurality of
conceptions of the good.52
The severance of the Rechtsstaat from the external constraints of transcendental
religion or ethics further specifies what state rule through law is not without revealing what it
ought to be, or whether it could ever altogether escape from the grip of contested religious or
ethical values which might stubbornly linger within it. To better appreciate the potential for
positive contribution of the nineteenth century Rechtsstaat, it is thus necessary to turn briefly
to a comparison between its Kantian and its positivist dimension, and to examine how these
might be reconciled.
For Kant, a legal regime is legitimate if it is grounded in the right. Granting that citizens
have different interests and competing ideas about the pursuit of happiness, Kant recognizes
that nothing like an actual consent of the entire citizenry could ever validate any concrete piece
of legislation53. Consistent with this, legitimacy cannot be established at the level of interests
51
Id., at 279.
52
Significantly, this precisely is what the anti-pluralist Carl Schmitt reproached to the
late nineteenth century conception of the Rechtsstaat. See Carl Schmitt, “The Liberal Rule
of Law” in Arthur J. Jacobson and Bernhard Schlink, eds., WEIMAR: A JURISPRUDENCE
OF CRISIS 294, 297-99 (2000). See the discussion of Schmitt infra.
See Immanuel Kant, “On the Common Saying: ‘This May Be True in Theory, But It
53
Does Not Apply in Practice’’‘ in KANT’S POLITICAL WRITINGS, supra, at 61, 80.
23
or of the good, but only at that of the just and the right -- that is, by categorically treating every
citizen as free and equal and as an end in him/herself. In other words, a law can only be
legitimate if it is reasonable for every citizen to accept it as being right and just. Pursuant to
this criterion, the legislator is obligated, in Kant’s own words,
to frame his laws in such a way that they could have been
produced by the united will of a whole nation, and to regard each
subject in so far as he can claim citizenship, as if he had
consented within the general will. This is the test of rightfulness
of every public law54.
Kant’s test thus sets a counterfactual against which the rightness of law is to be
measured. Regardless of how citizens actually feel or whether they would have actually voted
for a law, the key question is whether it is proper for citizens -- conceived as free and equal
and as treating one another as ends in themselves -- to have enacted the law in question as
legislators and to have willingly accepted to be bound by it as citizens. If the answer is in the
affirmative, then the law in question is legitimate.
As Kant’s counterfactual test does away with consideration of interests or of the good,
it raises the question of whether the just and the right can be found beyond the realm of
interests or whether it remains altogether beside it. In other words, is satisfaction of the
counterfactual dependent on there being a realm of justice compatible with all conflicting
interests and all competing conceptions of justice -- a very strong requirement. Or is it
possible to satisfy the counterfactual regardless of what interests or conceptions of the good
may be involved -- a seemingly weak requirement.
54
Id., at 79.
24
A close look at Kant’s argument reveals that his conception of the just and the right lies
both beyond and besides the realm of interests. More precisely, the right ascends beyond the
good in connection with the fundamental rights and freedoms of the individual, but remains
besides it when it comes to the criterion of self-legislation. And as a consequence of this,
Kant’s counterfactual criterion of self-legislation ultimately seems extremely weak. Under the
test of self-legislation, as Kant explains, if a law is
such that a whole people could not possibly agree to it (for

example if it is stated that a certain class of subjects must be
privileged as a hereditary ruling class), it is unjust; but if it is at
least possible that a people could agree to it, it is our duty to
consider the law as just, even if the people is at present in such
a position or attitude of mind that it would probably refuse its
consent if it were consulted.55
Kant’s test for self-legislation thus boils down to a requirement of formal equality before
the law. So long as laws are equally applied to all, they must be deemed
legitimate regardless of their content. When placed in its Enlightenment context, this test is
by no means trivial as it delegitimates all vestiges of status-based legislation typical of the
Ancien Régime. In today’s world where feudal hierarchy has been widely banished from
constitutional democracy, however, Kant’s test of self -legislation rings rather hollow. This can
be illustrated, moreover, through Kant’s own example in support of his test. Kant’s argues that
if a proportional tax is imposed on the entire citizenry in order to finance an unpopular war, this
would meet the test of self-legislation for it would be possible for all accept the tax if they
55
Id. (emphasis in the original).

25
supported the war56. If the tax were only imposed on part of the citizenry, on the other hand,
those singled out for that burden would have good reason not to accept it voluntarily even if
they enthusiastically supported the war. While all this may be true, it seems largely besides
the point. Without any meaningful consent on the war -- whether directly or indirectly through
endorsement of the decisionmaking process responsible for the war policy -- it is difficult to
conceive the tax as a plausible product of self-legislation.57
Fundamental rights and freedoms seem on more solid ground than self-legislation in
so far as they can be legitimated beyond the realm of interests. Such legitimation, moreover
may depend on these rights being beyond interests in the sense of remaining consistent with
all conceivable differences in interests. Or else, they could be beyond interests by stacking
up against them and thus imposing boundaries on the legitimate pursuit of interests.
The key right for Kant is that to individual autonomy58, which he envisages as requiring
freedom, equality and the right to own property59. Also, in organized society, preservation of
the individual’s autonomy depends on others treating that individual as an end in him/herself
rather than merely as means.
If Kantian autonomy requires treating every individual only as an end then the criterion
56
Id.
57
‘Arguably, confronted with an unpopular war, citizens could still agree that it ought to
be financed to avoid disaster, and that the resulting burden ought to fall equally on all rather
than being disparate. Nevertheless, the latter agreement would fail the weak consent test set
out above.
58
Id., at 77-78.
59
Id., at 74ff.
26
of legitimacy for law is clear, but legitimate law is impossible. If, on the other hand, Kantian
autonomy may be satisfied by treating every individual as more than a mere means -- for
example, through implementation of some version of the rights to liberty, equality and property
which would not foreclose all treatment as means -- then law is certainly possible but could
never be more than partially legitimated.60
To illustrate, consider the case of employment law. Every complex contemporary
society committed to the rule of law needs to adopt legal standards to govern the employment
relationship. Both employers and employees, however, typically relate to one another as
means toward the achievement of particular economic objectives. Accordingly, no law
pertaining to employment could altogether forbid (or avoid legitimating) treating another as
means with the consequence that it would be impossible to come up with a legitimate
employment law so long as the right is construed as requiring treating people exclusively as
ends. Conversely, even if there were a consensus concerning the requirements for the
individual to maintain an acceptable measure of autonomy while nonetheless being treated
by others as means, the scope of treatment as means would effectively have to be determined
60
It may seem that those difficulties might be avoided by charging legislators with the
duty to treat citizens as ends without extending that duty to citizens in their interactions with
fellow citizens. Thus, laws could cast persons as ends regardless of the positions espoused
by legal actors. This possibility must be rejected, however, for at least two crucial reasons.
First, consistent with Kantian autonomy, for any legislation to be normatively justified, it must
in some sense qualify as self-legislation, thus invalidating any normative distinction between
the legislator and those subjected to legitimate legal norms. And, second, the legislator
cannot, in the last analysis, remain above contested (or contestable) interests. Accordingly,
given that in a complex society it is virtually impossible to remain self-sufficient in the quest
to satisfy one’s interests, all legal norms requiring citizens to act to refrain from certain actions
cannot avoid enlisting some as means in the pursuit of interests held by others.
27
in terms of clashing interests. Consistent with this, there would seem to be no compelling
reasons for any loser in such struggle over interests to accept unfavorable legal outcomes as
legitimate. In short, if autonomy is the source of law’s legitimacy, how can that which lies
beyond autonomy in any way contribute to such legitimacy. Returning to the employment law
example, if respect for liberty, equality and the right to own property is deemed sufficient to
sustain the necessary measure of autonomy, then any employment law consistent with that
would be presumptively legitimate. But if more than one such law would be consistent with the
above mentioned rights, and one of the latter would better serve the interests of employers
and another those of employees, then at least in part these laws would depend on choices
among interests and to that extent could not be deemed, strictly speaking, legitimate 61.
I have thus far assumed that the fundamental rights to liberty, equality and property
essential to the Kantian conception remain beyond the realm of interests - - either because
compatible with all interests or because they can bracket out all interests. But of course this
need not be so. By emphasizing negative liberty, formal equality and private property62, Kant
endorses particular rights (or particular versions of certain rights) which can be attacked as
prone to favoring certain interests over others. Thus, while Kant’s rights may well suit the
interests of libertarians, they seem ill equipped to promote those of egalitarians. And if this
is true, then these rights do not really rise above interests.
61
It would of course be different if the test of legitimacy were limited to requiring
compliance with the above mentioned rights without any reference to the requirement to treat
persons as ends or to interests. In the latter case, interest biases might well have no impact
on legitimacy, but such legitimacy would not be truly Kantian in nature.
62
See KANT’S POLITICAL WRITINGS, supra at 74ff.
28
A Rechsstaat rooted in the Kantian conception would, in addition to complying with the
rule of law in the narrow sense, require the state to adhere to equality before the law, to grant
citizens (negative) liberty, formal equality and (bourgeois) property rights. Consistent with the
above discussion, such Rechtsstaat can only be plausibly viewed as rising beyond interests
from the perspective of the Enlightment looking back at the Ancien Régime. Once the Ancien
Régime completely vanishes from the horizon, however, the Kantian Rechtsstaat becomes
inevitably mired in the realm of conflicting interests, and therefore cannot provide sufficient
backing to the legitimacy of law. In other words, the Kantian Rechtsstaat rises above all the
interests it encompasses in its confrontation with what it stands against, but has no way of
doing so in its quest of what it should stand for. Similarly, the Kantian Rechtsstaat remains
compatible with the ideal of treating its citizens exclusively as ends so long as it is pitted
against pre-Enlightment legal regimes relying on transcendental values. However, once these
transcendental values have completely lost their hold, the ideal of treating citizens exclusively
as ends can only serve as a counterfactual reminder that establishing law’s legitimacy above
interests has become impossible. Accordingly, either post Enlightment legal legitimacy is
impossible ir it is only legitimate to the extent that the exclusion of certain interests can be
justified. Thus, post Enlightment legal legitimacy’s very possibility depends on finding an
acceptable justification for endorsing certain interests and a conception (or several
conceptions) of the good while rejecting others.
In light of these theoretical observations, it is not surprising that the historically
grounded early nineteenth century Rechtsstaat has given way to the much different

29
substantively grounded post World War two Rechtsstaat associated with the current legal -
constitutional regime in Germany63. Before focusing on the latter, however, it is necessary
to briefly consider the late nineteenth century positivistic Rechtssaat and the crisis that led to
the demise of the Weimar Republic in order to place the German experience in proper
theoretical perspective.
From a historical standpoint, the shift from a Kantian to a positivistic Rechtsstaat was
traceable to the failure of the liberal revolution of 1848 in Germany64. As a consequence of
this failure, there followed a de-emphasis of fundamental rights as constitutional principles
coupled with the emergence of a conception of the Rechtsstaat as primarily formal in nature.65
According to this conception, the Rechtssaat was not concerned with the content or purpose
of the law of the state but only with the methods employed by the state to foster its
realization66. This, moreover, tended to reduce state rule-through law to a principle of
legality67.
Placed in the context of the rise of the bourgeoisie, the Kantian Rechtsstaat
incorporates two separate elements, one formal and the other substantive. From a formal
standpoint, the Kantian Rechtsstaat incorporates the categorical imperative and requires law
63
See Rainer Grote, supra note ---, at 285ff.
64
Id.
65
Id
66
Id.
67
Id., at 281.
30
as norm to promote individual autonomy by treating all persons as ends rather than means.
From a substantive standpoint, on the other hand, the Kantian Rechtsstaat promotes equal
(negative) rights to liberty and property which clearly lend support to bourgeois values as
against those that had been entrenched in the Ancien Régime, but also -- perhaps much less
obviously, except in hindsight -- against those of the propertyless or of those who in spite of
their newly minted rights still depend on others for their very survival.
Unlike in neighboring France, there was no bourgeois revolution in Germany during
Kant’s lifetime, thus casting the substantive vision emanating from the Kantian Rechtsstaat
as a pure ideal with counterfactual implications. Germany did have its bourgeois revolution
in 1848, but it was defeated, and this together with the subsequent military successes of the
Prussian monarchy, left the bourgeoisie rather powerless. This prompted it to seek limits on
the powers of the Reich, hence settling on the positivistic Rechstaat68.
From the standpoint of the positivistic Rechtssaat, the Reich was conceived as a legal
person rather than as a league of princes69. Moreover,
The positivist theory of the law of the state viewed state

institutions above all from the viewpoint of limits. The law making
power of the Reichstag was limited by the Bundesrat, but at the
same time the power of the monarchic administration to interfere
with the freedom and property of citizens was limited by the
requirement of statutory authorization70.
68
See Arthur J. Jacobson and Bernhard Schlink, “Introduction: Constitutional Crisis” in
WEIMAR: A JURISPRUDENCE OF CRISIS, supra note —, at 5-6.
69
Id., at 6.
70
Id.
31
In other words, the positivistic Rechsstaat affords protections through the institutional
framework of state rule through law, and through imposition of the requirement that the state
act through promulgation and implementation of laws rather than through mere deployment of
the will of the monarch. Although in practice it may be poised to advancebourgeois economic
interests as distinguished from their political interests71, in theory it lacks both the formal and
substantive tendencies of its Kantian counterpart.
Taken out of context, the positivistic Rechtsstaat might be regarded as marking a
regression to the rule of law in the narrow sense. When viewed in proper historical
perspective as set against the Kantian Rechtsstaat as ideal rather than reality, however, the
focus on legality acquires a somewhat different meaning. Indeed, particularly given the rise
of the administrative state, the positivistic Rechsstaat’s requirement of legality potentially
endows state rule through law with a much needed measure of rationality and predictability,
by facilitating stabilization of the expectations of legal actors confronted with an ever more
complex social and institutional setting 72. To be sure, law’s predictability in a complex social
universe is no guarantee of its legitimacy, but it does make for an important step in that
direction. Thus, at least in some important fields of contemporary law, the content of legal
norms is less important than their stability. For example, in the realm of private contractual
71
See Id.
72
For Niklas Luhmann stabilizing expectations in an ever more complex world is the
paramount function of contemporary law. See his :”The Unity of the Legal System” in
AUTOPOIETIC LAW: A NEW APPROACH TO LAW AND SOCIETY (Gunther Tuebner, ed.
1987), and his “Operational Closure and Structural Coupling: The Differentiation of the Legal
System”, 13 Cardozo L. Rev. 1419 (1992).
32
relationships among commercial firms of roughly comparable bargaining power, knowing
there are stable relevant applicable legal norms is in most cases far more important than what
the features of such norms may happen to be. Indeed, if the parties to a contractual
arrangement are unhappy about any of the applicable legal rules, they can contract around it.
Accordingly, if in the context of a sales contract, the law provides that the seller bears the risk
of loss of goods prior to delivery to the buyer, the parties know what to expect, and the seller
can either get insurance to counter the risk of such loss, or attempt to persuade the buyer (e.g.
by offering the latter otherwise more favorable terms) to contractually assume the risk of loss
prior to delivery73.
In at least certain fields of law then, the predictability associated with formal legality is
efficient if not inherently equitable. Moreover, by stabilizing expectations, such formal legality
makes for greater control over one’s interests in a complex social setting rife with a vast array
of administrative regulation. Formal legality would certainly not live up to the rigorous
demands of Kantian autonomy, but it may well contribute to the legitimacy of contemporary
legal regimes to the extent that it equally endows all legal actors with rational means to
coordinate their pursuit of self-interest. Before pursuing this any further, however, it is
necessary briefly to examine the evolution of the concept of the Rechtsstaat during the
Weimar Republic and the crisis that led to its demise.
Constitutional protection of fundamental rights, which had been coveted by the 1848
73
Even if the law imposing the burden on the seller were rigid and it were impossible
strictly speaking to contract around it, so long as expectations remained stable, the parties
could still adjust their exchange, by taking the seller’s burden into account when determining
the sales price, etc.
33
revolutionaries, finally came to Germany after the country’s defeat in World War I. As noted
by Carl Schmitt, however, the new constitutional rights received in defeat did not garner the
enthusiasm which they would have generated in 184874. Moreover, the new Weimar
Constitution was to be mired in crisis and short lived. The new liberal rule of law may have
suited the bourgeois, but -- at least in Schmitt’s view -- could not integrate the working class
within the unity of the state 75. Schmitt’s own proposed solution was to abandon liberalism, to
foresware pluralism, and to turn to the people as a homogenous whole to take over the
political destiny of the German nation-state 76.
For Schmitt, politics is anchored in the distinction between friend and foe, and
bourgeois liberalism can only serve to detract from what is essential 77. From the standpoint
of legitimating constitutional democracy for a pluralist polity, however, the principal lesson to
be drawn from the Weimar experience is that supplementing the positivistic Rechtsstaat with
liberal bourgeois constitutional rights may leave out a significant number of members of the
polity, thus falling far short the minimum level of consent required for legitimacy.
In the end, the Kantian ideal of autonomy falls short both from a formal and from a
substantive viewpoint. As formal, it remains too abstract, and as substantive, consistent with
the Weimar experience, insufficiently universal. Positivistic rule through law, on the other hand,
74
See Carl Schmitt, “The Liberal Rule of Law”, supra note —, at 295.
75
Id., at 297.
76
Id., at 298-300.
77
See Volker Neumann, “Carl Schmitt: Introduction” in Arthur J. Jacobson and Bernard
Schlink, eds., WEIMAR: A JURISPRUDENCE OF CRISIS, supra note —, at 280, 283.
34
makes for increased predictability, which is particularly important as social and legal
relationships become more complex, but provides no assurances that laws will be fair.
Against this background, and emerging on the heels of Germany’s total defeat at the end of
World War II, the new Rechtsstaat seeks to reconcile the need for predictability with the quest
for autonomy, through commitment to substantive values, and in particular through
entrenchment of respect for human dignity as the paramount constitutional value.
Set against the horrors of the Nazi era, the contemporary German Rechtsstaat shaped
by its new constitution, the 1949 Basic Law, subordinates positive legality to entrenched
substantive principles and values.78 And chief among the latter is human dignity which is
enshrined as an unamenable constitutional value in Article I of the Basic Law, and which has
been interpreted as the paramount value in the German constitutional order in numerous
decisions of the German Constitutional Court.79 More generally, today’s Rechtsstaat has
become inextricably tied to constitutional democracy framed by fundamental substantive
values, and its legality has become subjected to a set of substantive norms embodied in
constitutional justice.80
Although today’s Rechtsstaat in some sense incorporates elements of both its Kantian
and positivistic counterparts, it is in key respects different than its predecessors and thus
raises novel questions regarding law’s legitimacy. Like its Kantian counterpart, today’s
78
See Rainer Grote, Supra at 285.
79
Id. At 286 and n. 71.
80
See Donald Kommers, supra, at 36.
35
Rechtsstaat enshrines fundamental rights above the realm of ordinary laws, albeit that these
rights are substantive rather than formal and that they differ significantly in content from their
Kantian predecessors. On the other hand, like its positivistic predecessor, today’s
Rechtsstaat institutionalizes legality, but it is a legality that is not merely dependent on
consistency and predicatablity, but also contingent on constitutional conforminity and on the
realization of constitutionally recognized substantive goals. This, in turn, tends to
constitutionalize all politics and to convert the Rechtsstaat into a Verfassungsstaat (i.e. a
state rule through the constitution) as some German scholars have argued.81 Finally, even
beyond constitutionalization as such, today’s Rechtsstaat judicializes realms such as the
promotion of welfare, which were clearly relegated to politics in both of its nineteenth century
predecessors. Thus, the Basic Law commands the German states -- the Länder – to
promote the sozialer Rechtsstaat or sozialstaat (i.e. the social welfare state through law) as
well as democracy and republicanism.82
Whereas the constitutionalization of politics and the state’s obligation to promote the
social welfare of all seem unobjectionable if not indispensable from the standpoint of erasing
all vestiges of the legacy of Nazism, they raise serious legitimacy issues for established and
unthreatened constitutional democracy. To be sure, commitment to human dignity taken in the
See, e.g., Ulrich Karpen, “Rule of Law” in THE CONSTITUTION OF THE FEDERAL
81
REPUBLIC OF GERMANY 173 (Ulrich Karpen, ed. 1988) (defining the Verfassungsstaat as
a state which “means to organize politics and evaluate goals by applying, executing the
constitution”).
82
See German Basic Law, Art. 28, para 1.
36
abstract should command universal approval, but this should not necessarily extend to diverse
and often conflicting more concrete instantiations of the concept.83 For example, does human
dignity require forbidding or permitting abortion? Or assisted suicide? Does it call for
pervasive social welfare legislation or does it render it demeaning?
The more pervasive constitutionalization based on substantive values becomes, the
less it is likely to command widespread support. Overconstitutionalization forces some in the
polity to become subordinate to the values and conceptions of the good of others and thus
threatens to delegitimize the Verfassungsstaat.84 In the last analysis, over-
constitutionalization gives rise to a very similar problem to that produced by strict Kantian
autonomy. In the latter case, legitimate law is bound to alienate one from one’s own interests
as the right must remain above all interests; in the former, one always risks alienation from
one’s own interests to the extent that the constitution enshrines conflicting interests. And in
a pluralist polity, this means that a sizeable portion of the citizenry will variously remain
significantly alienated from the dictates emanating from the prevailing substantively grounded
legal-constitutional regime.
For a similar argument concerning human rights, see Michel Rosenfeld, “Can Human
83
Rights Bridge the Gap between Universalism and Cultural Relativism?” 30 Columbia Human
Rights L. Rev. 249 (1999).
84
For a critique of overconstitutionalization in Germany by a German constitutional
scholar, see Bernhard Schlink, “German Constitutional Culture in Transition” in
CONSTITUTIONALISM, IDENTITY, DIFFERENCE AND LEGITIMACY, supra note , at
197.
37
IV. The French État de droit as Means to

Vindicate Fundamental Rights Through Law
The French État de droit is much more recent than the German Rechtsstaat and was
originally derived from the latter.85 However, even though the French term “ ”État de droit” is
a literal translation of the German “Rechtsstaat”, the French adapted and transformed the
concept they found in nineteenth century German legal thought so thoroughly that the French
expression came to acquire a completely different meaning than that connoted by the
positivistic Rechtsstaat86. Indeed, in its current meaning as understood in French legal theory
and as institutionalized in the contemporary French constitutional order, “État de droit” does
not mean “state rule through law”, but rather “constitutional state as legal guarantor of
fundamental rights” (against infringements stemming from law made by parliament).87
The French État de droit has come to supplement and, in an important sense, limit the
“État Légal”, which may be roughly translated as “democratic state rule through law.” The Etat
Légal is therefore what comes closest in France to the positivistic Rechtsstaat, with the key
difference that the French concept is inextricably linked to parliamentary sovereignty and
85
See Jacques Chevallier, L’ ÉTAT DE DROIT 11 (3d ed. 1999).
86
Id. At 11, 22-31.
See id. Just as its German or Anglo-American counterparts, the French conception
87
of État de droit is dynamic and highly contested. Given the present aim of identifying essential
characteristics for purposes of conceptual analysis and comparison, however, no attempt will
be made to account for most nuances or matters of dispute. For a thorough account of the
origins of the French conception, see Marie Joëlle Redor, DE L’ ÉTAT LÉGAL À L’ÉTAT DE
DROIT: L ÉVOLUTION DES CONCEPTIONS DE LA DOCTRINE PUBLICISTE FRANÇAISE
(1992).
38
parliamentary democracy. Accordingly, whereas the positivistic Rechsstaat’‘s primary function
is to set institutional limitations on the uses of governmental powers by the monarch against
the people, the État de droit embodies the democratic will of the French nation as transformed
into law by the parliament.88
The concept of État de droit was first articulated in the aftermath of World War I in the
works of Carré de Malberg,89 and it became institutionalized within the French constitutional
system after World War II90. As already indicated “État de droit” unlike ”rule of law” or
“Rechtsstaat” does not refer to law as a whole but rather to fundamental rights as having the
force of law. In other words, the État de droit is the state backed legal regime shaped by
fundamental liberal rights which places constraints on the État légal.91 Moreover, as Carré
de Malberg emphasized, the État de droit could not be fully realized until the adoption of
constitutional review of parliamentary laws 92 - - a development that took place in France in
1971 when the Constitutional Council for the first time invalidated a law of parliament for
infringement of a fundamental right enshrined in the 1789 Declaration of the Rights of Man.93
88
See Jacques Chevallier, L’ETAT DE DROIT, supra note , at 18.
89
CONTRIBUTIONS À LA THÉORIE GÉNÉRALE DE ´ ÉTAT (vol. I, 1920) pp. 488 ff.
90
See Rainer Grote, supra note , at 292-94.
91
See Carré de Malberg, supra note , at 490.
92
Id. at 492.
See CC No. 71-44, July 16, 1971 (parliamentary law held unconstitutional as
93
violation of fundamental right to freedom of association).

39
France’s recourse to the État de droit as a check on the laws issuing from
parliamentary democracy is a stunning development in a culture long persuaded that
parliamentary sovereignty would best express the nation’s will and at the same time
adequately protect its citizens’ fundamental rights. Moreover, not only was France’s
commitment to parliamentary sovereignty solidly entrenched, but it also had deep theoretical
roots in the political philosophy of Jean-Jacques Rousseau.
According to Rousseau, the conflict between clashing individual interests, on the one
hand, and the common good of the polity, on the other, could be solved through pursuit of
democratic self-government. In Rousseau’s conception, however, democracy is not mere
majority rule with the inevitable consequence that political minorities are compelled to obey
laws imposed against their will. Instead, democracy requires implementation of the general
will through the efforts of the entire citizenry working to overcome the disparate demands
arising form the realm of clashing private interests in order to embrace as their own what is
good for society as a whole.94 Pursuant to Rousseau’s analysis, moreover, by partaking in
the legal expression of the general will every citizen engages in self-legislation. Conversely,
as a person with private interests every member of the polity must voluntarily restrain his or
her particular interests in order to pave the way for the laws embodying the general will, thus
freely consenting to become bound by such laws.
The key to Rousseau’s democracy oriented towards the general will is self-restraint.
Such self-restraint, however, is not the consequence of some fear of adversity but rather that
See Jean-Jacques Rousseau, THE SOCIAL CONTRACT 14 - 18 (Charles Frankel,

94
ed. 1947).
40
of a free assumption of responsibility. For Rousseau, each individual is both a citizen and a
private person or bourgeois. As a bourgeois, each individual pursues private interests which
often clash with those of others. As a citizen, on the other hand, each citizen is part of the
sovereign charged with ruling society pursuant to the dictates of the general will. Accordingly,
the two perspectives of the governors and the governed coexist within each member of the
polity. Because freedom cannot be realized without self-government, every individual must
resolve the conflict between the two perspectives within him or her so as to be able to
participate in government while continuing to attend to as many of his or her private interests
as possible (without compromising any duties as governor) in his or her capacity as a
member of the governed.
If there were room for fulfillment of both the general will and part of the objectives
issuing from private wills, then the need for self-restraint and the requisite sacrifices regarding
self-interest might well be outweighed by the benefits of increased participation in democratic
law making and enhancement in the scope of self-determination. If, however, adherence to
the general will requires complete suppression of private interests, then Rousseauian
legitimation of law would remain unpersuasive as it is difficult to see why someone would give
up nearly everything he or she holds dear to take an active role in producing laws poised to
frustrate his or her most cherished objectives.95
Rousseau himself sheds little light on how recourse to the general will might impact the
95
If to attain the general will the interests of the private person must completely give
way to the collective duties of the citizen, then the ideal of self-government as envisaged by
Rousseau might well be a precursor of totalitarianism. See Roger Masters, THE POLITICAL
PHILOSOPHY OF ROUSSEAU 315 (1968).
41
balance between private and collective interests. Indeed, as used by Rousseau, the notion
of the general will remains somewhat mysterious and not altogether consistent. The general
will must be distinguished from individual or majority will, and Rousseau conceives it as the
sum of the differences between all the individual wills, or as the “agreement of all interests”
which “is produced by opposition to that of each.”96
Notwithstanding these difficulties, two important points emerge from consideration of
Rousseau’s general will. First, unlike Kant, Rousseau does not rely on the distinction between
interests and the right when dealing with the issue of law’s legitimacy. For Rousseau, law is
legitimate if it is an expression of the general will which itself emerges as a consequence of
a dynamic process which must account for all relevant interests. Second, self-government has
intrinsic value so that -- at least to a certain extent-- a trade off which would enhance self-
government at the expense of the pursuit of self-interest should not adversely affect the
legitimacy of law.
There is a strong affinity between Rousseau’s theory and the French conception of the
État légal. Indeed, there is an obvious congruity between Rousseau’s vision of all the citizenry
joining together to govern and the French Revolution’s conviction that the democratic fate of
the nation would be best served by a single nationwide legislature entrusted with the
transformation of the will of the people into law. This was set against an Ancient Régime
marked by absolute rule in favor of those privileged within the prevailing feudal hierarchy. In
contrast, the representatives of the people as a whole acting after the repeal of feudal
96
Jean-Jacques Rousseau, THE SOCIAL CONTRACT supra note –, at 26 n. 2.
42
privileges could easily be expected to legislate in the common interests of all -- particularly
to the extent that bourgeois interests were projected as being universal. Consistent with this,
all laws of Parliament, regardless of their outcome, became perceived as expressions of the
general will. Today, however, there seems no inherent reason for believing that parliamentary
democracy will necessarily give voice to the common good. And hence the need to balance
the État légal with the protections afforded by the État de droit.
Given the strong influences of Rousseau’s ideas, the constitutional tradition emanating
from the French Revolution conceived of the limitations of powers of government required by
constitutionalism in terms of parliamentary sovereignty – that is in terms of democracy itself97.
Moreover, parliamentary sovereignty and the État légal associated with it guaranteed
implementation of the rule of law -- at least in the narrow sense. Finally, the protection of
fundamental rights was guaranteed by the 1789 Declaration, but such protection was political
rather than legal. And it is precisely because of this lack of legal protection, that the État légal
had to be supplemented by the État de droit.
V. The Anglo American Common Law

and the Paradoxes of The Rule of Law
One of the factors that reinforced France’s commitment to the État légal was the
manifest distrust ever since the Revolution against judges -- a distrust rooted in the negative
97
Strictly speaking, parliamentary sovereignty arguably amounts to no limitation of the
powers of government. When placed in proper context, however, it does constitute a
limitation, both in terms of accountability to the people and of having to act through laws.
Moreover, these limitations become all the more important when contrasted to the absolute
powers of the French king.
43
role performed by judges during the Ancien Régime98. Although the British like the French
have a long tradition of parliamentary sovereignty, the British have developed a positive
attitude towards judicial power, which has enabled them to cast the judge as a protector of the
citizenry rather than as the enemy of the people 99. Unlike the United States, the United
Kingdom does not have a written constitution and its judges thus do not have as clear a
mandate as their American counterparts to provide a check against legislative powers.
Nevertheless, the Anglo-American tradition relying on the common law has developed a
strong sense of the rule of law. And as we have already seen and will now further investigate,
unlike their continental counterparts, the Anglo-American concept of the rule of law is not
exclusively dependent on the state as such, but rather functions as a buffer between the
interests of the state and those of its citizens.
In its American version, the rule of law is grounded on a written constitution designed,
among other things, to provide legal expression to pre-existing inalienable fundamental rights.
These are deeply steeped in a Lockean vision of natural rights as belonging to the individual
and as preexisting and transcending both the social contract and civil society100 . In
accordance with this vision, the individual agrees to the social contract and civil society in
order to secure better coordination in the enforcement of his or her rights. This, in turn,
98
See Rainer Grote, supra note ----, at 283.
99
Id., at 273-76.
100
For an extended discussion of Locke’s contractarianism in comparison to that of
Hobbes, Rousseau and Kant, see Michel Rosenfeld, “Contract and Justice: The Relation
Between Classical Contract Law and Social Contract Theory”, 70 Iowa L. Rev. 769 (1985).
44
imposes two essential duties on the state created pursuant to the social contract: the negative
duty to refrain from interfering with its citizens’ enjoyment of their inalienable rights; and the
positive duty to deter and/or punish private infringements of fellow citizens’ rights through the
provision of police protection and the enforcement of private contracts101. Strictly speaking
however, the latter state duty is positive only in a derivative sense, as its goal is not to confer
any right on the individual, but rather to insure that others are prevented from harming or
destroying already existing rights. Under the vision in question, therefore, the raison d’être of
the state is to safeguard its citizens negative rights through self-restraint and through restraint
of would-be (negative) rights infringers.
In the context of an idealized minimal state virtually exclusively concerned with better
securing pre-existing natural rights, the rule of law would by and large consist of the
deployment and enforcement of procedural safeguards. This presupposes that the legitimacy
of natural rights would remain beyond dispute, and that effective protection of such rights
would guarantee the welfare of society either because it would allow citizens to be self-
sufficient or because it would enable them to remedy any lack of in self-sufficiency through
private contracts. In other words, if natural rights were universally accepted and sufficient to
allow everyone to fulfill his or her welfare needs, then legal standards could be automatically
set and no room would be left for politics102. And accordingly, the rule of law would boil down
101
This much scaled down state derived from the Lockean vision is what Robert Nozick
has termed the “minimal night-watchman state”. See his ANARCHY, STATE AND UTOPIA
26-27 (1974).
102
Politics is understood here “as encompassing setting objectives for the polity and
devising means designed to further such objectives”. Michel Rosenfeld, JUST
45
to the deployment and maintenance of procedural safeguards which mediate between right
holders, the state, and potential or actual rights infringers. Depending on the circumstances,
right holder and state may be on the same side -- i.e., when the state acts to protect a right
holder against third parties or provides the means for a right holder to obtain legal redress for
private infringements -- or on opposite sides -- i.e, when the state exceeds its legitimate
authority and threatens to (or does) infringe on the right holder’s entitlement because of a
failure of self-restraint or because of excessive zeal in the protection of another right holder’s
entitlement.
It becomes clear from this rough outline how the rule of law can be invoked against the
state. As already mentioned this raises a paradox. Indeed, though the state may be morally
obligated to yield whenever it threatens natural rights, so long as it retains a monopoly in
lawmaking and law enforcement nothing short of revolution would seem capable of prompting
it to desist from a deliberate course of natural rights infringement103. As will become
apparent, however, the paradox in question is least troubling in the context of the Anglo-
American rule of law tradition embedded in the common law. On the other hand, the common
law itself gives rise to more serious paradoxes that pose far greater threats within the Anglo-
American tradition. Two of these raise serious questions about the existence and viability of
INTERPRETATIONS: LAW BETWEEN ETHICS AND POLITICS 75 (1998). Since in the

ideal minimal state society each individual endowed with inalienable natural rights sets his or
her own objectives and realizes them with the cooperation of others through private market
transactions, the economic realm presumably obviates any need for politics.
Cf. JOHN LOCKE, THE SECOND TREATISE OF GOVERNMENT § 222 (1714)

103
(asserting that it within the people’s right to rebel against a government that violates their
fundamental rights).
46
the rule of law: the paradox produced by the tension between the need for legal certainty and
predictability and the common law’s piecemeal, experimental and incremental approach; and
that generated as a consequence of the seeming clash between the need for binding and
transparent criteria of judicial application of relevant legal norms and the great latitude enjoyed
by common law judges which is prone to blurring the distinction between law making and
judicial interpretation.
Of the three paradoxes, the first seems to be the least troubling. Indeed, the
conception of the rule of law as being at once dependent on, and independent from, the state
becomes much more plausible once one realizes the unique position of judges and of the
judicial system in the Anglo-American tradition. Going back to feudal England, legal norms
have traditionally issued from multiple sources and the power of adjudication has remained
divided among different and often competing institutional actors104. Thus, the statutory law
made by Parliament has existed side by side with the judge made common law, and courts
of law were supplemented by courts of equity105. Moreover, the responsibilities entrusted to
the judicial function have been apportioned between judge and jury -- an institution brought to
England by William the Conqueror in the eleventh century and widely used as a check on the
monarchy’s judges by the seventeenth century106.
In as much as sources of law and judicial actors could be set against one another, parts
See, e.g., Frederick Pollock and Frederic Maitland THE HISTORY OF ENGLISH
104
LAW, vol. 1, 174ff. (1968).

105
See James Hazard, CIVIL PROCEDURE 14-15 (1985).
106
See Jack Friendenthal el al., CIVIL PROCEDURE 472 (1985).
47
of the state apparatus could be mobilized against others. Consistent with this although all law
ultimately depended on state backing, mobilization against others made it literally possible
to have the rule of law enforced against the state. Moreover, the United States not only carried
forward this English tradition but it also enshrined it in its Constitution through a system of
“checks and balances” and a sharp division between judicial power and its executive and
legislative counterparts107. Thus, federal constitutional norms can be invoked to invalidate
inconsistent legislation emanating from one of the states108;and, by setting the judicial branch
as independent from, and coequal to, the legislative branch and the executive branch, the
federal constitution enables judges to invalidate laws enacted by the Congress109.
The remaining two paradoxes, in contrast, seem much more troubling. Indeed, to the
extent that the common law is always changing, predictability is problematic, and it seems
difficult to conceive of the rule of law in a setting in which citizens may be unable to discover
ex ante the consequences of their acts. Furthermore, so long as the line between judicial
interpretation and judicial law making remains blurred, there seems to be no cogent way to
draw a plausible distinction between the rule of law and politics. As we shall see, these
difficulties may be surmounted if the lack of predictability associated with the common law
107
See U.S. Const. Articles I, II and III.
See, e.g., Brown v. Board of Education, 347 U.S. 483 (1954) (state of Kansas law
108
making racial segregation in public schools struck down as unconstitutional under federal
constitution’s equal protection clause).
109
The U.S. Supreme Court’s power to invalidate an unconstitutional law enacted by
the U.S. Congress was firmly established in its landmark decision in Marbury v. Madison, 5
U.S. 137 (1803).
48
could be tempered by procedural safeguards or if the dynamics of the common law system
could foster predictability in ways that are not dependent on rules; and if the realm of judicial
intervention, broad though it may be, could be ultimately constrained by principle.
V I Common Law, The Rule of Law, Constitutional Democracy and Politics:

Procedure, Substance and the Challenge
from Critical Legal Studies
In its purest form (which is but a counterfactual ideal), the common law is a case by
case judge made law that evolves through elaboration of precedents by means of a process
of accretion driven by a logic of induction. Set in the context of an adversary system, each
party to a controversy argues his or her position to a judge who must decide the matter after
having heard all the evidence and all the arguments advanced by the contestants. The first
judge ever confronted with the task of adjudicating such a controversy would presumably only
have his or her own experience, common sense, and his or her understanding of justice and
fairness to draw upon in order to reach a verdict. In all, subsequent cases, however, the judge
is supposed to take relevant precedents -- i.e. prior judicial determinations -- into account and
to resolve the matter at hand in a manner consistent with such precedents. In such a system,
legal rules are supposed to emerge gradually by stringing together a sufficient number of
successive precedents to circumscribe a distinct path. But unless all existing precedents
compel a determinate outcome -- which is often not the case -- the common law approach
cannot guarantee predictability. Indeed, if precedents can be equally reconciled in a way that
49
leads to imposing or to denying liability in the case under consideration, then it seems
impossible for the parties to eventual litigation to know ex ante the legal consequences of their
intended conduct. Consistent with this, moreover, strictly speaking, the legal rule applied in
a case -- as distinguished from the result in that case -- cannot be known until subsequent
judicial decisions in future cases have further specified its relevant contours.
To illustrate the workings of the common law approach, consider the following example
involving a legal rule that cannot be grasped until it becomes further elaborated in a future
judicial opinion. A landowner brings a lawsuit against his neighbor because the latter’s cat
has entered upon plaintiff land causing damage for which the latter seeks to be compensated.
Moreover, the only relevant precedent involves a case holding that the owner of a cow is liable
to his neighbor for the damage caused to the latter’s property by the cow following its
unauthorized entry upon the plaintiff’s property. Under those circumstances, the judge sitting
in the case concerning the cat can infer at least two different rules from the precedent involving
the cow. The first rule is that the owner of a large animal is liable for any damage caused by
the latter following unauthorized entry upon the owner’s neighbor’s property. The second rule,
on the other hand, is that an owner is thus liable for any such damage caused by any of his or
her domestic animals. Since a cat is a small domestic animal, the plaintiff will lose his case
if the judge infers the first rule from the precedent, but he will win if the judge infers instead the
second rule.
Now, suppose further that the judge in the case of the cat rules in favor of the plaintiff
after concluding that the situation involving the cat is in all relevant respects analogous to that

50
regarding the cow. But the judge leaves unclear the basis for the analogy she draws between
the case of the cow and that of the cat. Under those circumstances, it will be left to another
judge before whom the next case in the series will be brought at some further date, to infer
which legal rule might cover all three cases consistent with the results in the respective cases
of the cow and the cat. Thus, the judge before whom the third case will be brought may
decide, for example, that the rule to be inferred concerns all of an owner’s domestic animals,
or that it instead covers all animals, whether domestic or not, which usually live on the owner’s
property. The important point, however, is that no matter which of these two alternative legal
rules is eventually chosen, the legal rule that accounts for the result in the case of the cat
cannot become fully explicit until its further articulation in the course of the judicial resolution
of subsequent cases.
The inherent lack of predictability associated with the common law can be somewhat,
but not fully, alleviated through constitutional provisions and statutory laws. Many key
constitutional provisions, such as the “due process” and the “equal protection” guarantees
contained in the American Constitution110 are stated very generally and at a high level of
abstraction. This allows for a wide range of plausible interpretations, and the common law
trained judges who have dealt with such constitutional provisions have widely differed in their
interpretations making these provisions nearly as unpredictable as constantly evolving
110
See U.S. Const. Amendment 14 which provides in relevant part that no person “shall
be deprived of life, liberty or property without due process of law” or of “the equal protection
of the laws”.
51
common law standards111.
Statutes are usually less general and less abstract than many constitutional provisions,
but are nonetheless often subject to such differing interpretations as to defy predictability.
Indeed, as common law techniques are brought to bear on statutory interpretation, the latter
may well approximate the unpredictability of its constitutional counterpart. For example,
statutory language prohibiting employment discrimination against any individual on account
of his or her race led to a five to four U.S. Supreme Court decision, with the majority holding
that the statute sanctioned the use of affirmative action and the dissenters vehemently
maintaining that the individual’s right against discrimination clearly precluded the legitimacy
of affirmative action112.
Even if one concedes that common law unpredictability permeates the entire American
legal system, this does not necessarily preclude a successful deployment of the rule of law so
long as the latter is conceived primarily in procedural rather than substantive terms. Moreover,
a procedurally grounded rule of law would revolve around three essential components: the rule
of law in the narrow sense; the prevalence and maintenance fundamental due process
guarantees; and, institutionalization of the adversary system of justice as a means to channel
conflicts towards legal resolution rather than towards other possible outcomes.
The rule of law in the narrow sense is firmly established in the United States and can
Compare, e.g., Plessy v. Fergusson 163 U.S. 537 (1896) (racial apartheid does
111
not violate equal protection guarantees) with Brown v. Board of Education, 347 U.S. 483
(1954) (racial apartheid in public schools violates equal protection guarantee).
112
See United Steelworkers v. Weber, 443 U.S. 193 (1979).
52
be traced back to the Supreme Court’s 1803 landmark decision in Marbury v. Madison.
Thus, even if one considers common law trained judges unpredictable or at times perhaps
somewhat arbitrary, the rule of law in the narrow sense does insure some checks on the
exercise of the power of the state in the name of the law. Furthermore, to the extent that
judicial decisions must be made public and the reasons for such decisions revealed in
published opinions, the likelihood of blatant judicial abuses seems rather remote. In short, the
rule of law in the narrow sense appears to insure a significant amount of legality and the
promotion of legal norms that do not stray too far from the well of commonly accepted
values113.
Fundamental due process guarantees have been enshrined in the United States
Constitution since the adoption of the Bill of Rights in 1791114. Also, it has been argued that
the whole Bill of Rights and even the Constitution as a whole is overwhelmingly process
oriented115. According to this view, the function of the Constitution and of judicial review is
113
One may object that this last statement does not properly account for certain
important but very divisive constitutional decisions by the Supreme Court, such as those
involving abortion, see Roe v. Wade, 410 U.S.113 (1973) or flag burning, see Texas v.
Johnson. 491 U.S. 397 (1989) and United States v. Eichman 496 U.S. 310 (1990).
Nonetheless, the essential values relied upon by the Court in deciding these cases were
widely shared as most Americans cherish the right to privacy relied upon in the abortion
decision and the freedom of speech right involved in the flag burning cases. Arguably,
therefore, the bitter divisions were not over the values involved but rather over their proper
contours or applications.
See U.S. Const. Amendment V (1791) affording due process rights against the
114
federal government. Such rights, however, were not accorded against the states until the
conclusion of the Civil War. See U.S. Const. Amend XIV (1868).
This is the thrust of John Ely’s thesis in his celebrated book DEMOCRACY AND
115
DISTRUST: A THEORY OF JUDICIAL REVIEW (1980).

53
to provide the necessary legal basis for a well functioning democracy. Consistent with this,
besides protecting democracy from its traditional enemies, the Constitution is meant insulate
the democratically generated legal order against majoritarian excesses and pathologies. In
this context, process based guarantees become part and parcel of the rule of law through
imposition of essentially procedural constraints on the generation and application of majority
based legal rules.
While the just described view need not necessarily go hand in hand with Lockean
presuppositions such as those already discussed above, it fits particularly well with such
presuppositions. Focus on this fit, moreover, is particularly useful as it highlights a very
serious problem with conceptions of the Constitution as overwhelmingly process based which,
in turn, undermines the importance of the nexus between procedural safeguards as such and
the rule of law. This problem is best highlighted through a brief consideration of the
controversy surrounding the Due Process Clause of the U.S. Constitution116.
This controversy centers on whether due process should be understood as conferring
purely procedural rights or whether it also includes a “substantive” component. The U.S.
Supreme Court has gone back and forth or this issue without reaching any definitive or
unanimous solution. In one of its most criticized decisions, the Court embraced substantive
due process and constitutionalized private property and freedom of contract in its 1905
Lochner decision117. The Lochner doctrine enshrined laissez-faire, and for approximately
See U.S. Const. Amend. V (due process protection against the federal government)
116
and Amend. XIV (against state governments).

117
Lochner v. New York, 198 U.S. 45.
54
thirty years led the Court to strike down state economic and social laws, such as minimum
wage maximum hours laws, designed to promote the general welfare, and often supported
by sizeable democratic majorities. Moreover, although the Lochner doctrine was repudiated
during the New Deal118, substantive due process has been reinvigorated in more recent times
and used to constitutionalize privacy and personal liberty rights rather than property and
economic liberty rights119.
The justification for substantive due process is that process based and procedural
rights can only be persuasively justified, if they are understood as part of a set of fundamental
norms inextricably tied to certain crucial substantive values. But to the extent that these
substantive values are not universally embraced within the polity -- as amply demonstrated by
the critics of Lochner or those of the abortion decision in Roe v. Wade -- the
substance/process dichotomy cannot be legitimately used to vindicate the rule of law in the
context of significant unpredictability concerning legal outcomes. More generally, this criticism
can be extended beyond due process itself as critics of Ely’s process based theory of the
United States Constitution have convincingly demonstrated120.
See Nebbia v. New York, 291 U.S. 502 (1934); West Coast Hotel Co. v. Parrish,
118
300 U.S. 379 (1937).
See e.g., Griswold v. Connecticut, 381 U.S. 479 (1965) (marital privacy rights), and
119
Roe v. Wade, 410 U.S. 113 (1973) (abortion rights).

120
See in particular the excellent refutation of Ely’s thesis in Lawrence Tribe, “The
puzzling Persistence of Process-Based Constitutional Theories”, 59 Yale L. J. 1063 (1980)
(arguing that not only are many of the provision of the Bill of Rights substantive in nature, but
also that those that are explicitly procedural cannot be cogently separated from their
substantive underpinnings).
55
The third component of a procedurally grounded rule of law is the adversary system of
justice. The adversary system blends naturally with the common law approach described
above and complements the rule of law in the narrow sense and the notion that no one is
above the law. Ideally, the adversary system allows each contending party to argue his or her
case to an open-minded and disinterested judge who will only reach a decision after having
heard and properly weighed all the relevant evidence presented as well as after having duly
considered the conflicting interpretations of relevant legal precedents advanced by each of
the contenders121. Because of his or her passive role during the trial and because of his or
her obligation to remain open-minded until all the arguments and proofs are in, the adversary
system’s judge looms as disinterested and impartial. At the very least, therefore, such a judge
promotes the rule of law by reaching an unbiased (in the sense that he or she has no reason
to favor any party before the court over any other) legally grounded and procedurally fair
decision that by and large should make dispute resolution through law preferable to other
alternatives for a vast majority of the citizenry.
In sum, two of the three essential components of a procedurally grounded rule of law
do not appear pose any insurmountable obstacles. These are the rule of law in the narrow
sense and the adversary system. The third component, however, has proven much more
problematic as process-based guarantees are seemingly inextricably tied to contestable
substantive norms. In other words, since different substantive commitments often lead to
different procedural constraints, process-based guarantees may not only be arbitrary but also
121
See Geoffrey Hazard, ETHICS IN THE PRACTICE OF LAW 121-23 (1978).
56
unpredictable.
The problems confronting the project of relying on a heavily procedurally oriented rule
of law would greatly diminish in importance if it were possible to draw a cogent and principled
line between law and politics in the context of common law adjudication. Indeed, if procedural
fairness could be guaranteed without the need to appeal to any contestable substantive
norms122, then procedural regularity and predictability would suffice to buttress the rule of law.
Everyone would enjoy basic legal guarantees against the government and fellow citizens and
would be assured that the arena left to political competition would operate under fair and
orderly generally applicable rules. In the absence of fair universally acceptable pure
procedural legal safeguards, however, the rule of law would have to depend on the
implementation of substantive law. But if the latter remains thoroughly political in its creation,
interpretation and application, then law would seen bound to collapse into politics. In short,
all law would be but politics by another name.
That all law is ultimately but politics is a position elaborated by the Critical Legal
Studies (CSL) movement123. In the broadest terms, the core of the CLS critique is that
common law judges are ultimately unconstrained by the legal materials which they must
For an extended argument against such possibility, see Michel Rosenfeld, “A

122
Pluralist Critique of Contractarian Proceduralism”, 11 Ratio Juris 291 (1998).
The CSL literature is both vast and diverse. For prominent samples, see Duncan
123
Kennedy, A CRITIQUE OF ADJUDICATION (FIN DE SIÈCLE) (1997); Mark Kelman,

“Interpretive Construction in the Substantive Criminal Law”, 33 Stanford L. Rev.591 (1981);
Duncan Kennedy, “The Structure of Blackstone’s Commentaries” 28 Buffalo L. Rev. 205
(1979); and Roberto Mangabeira Unger, “The Critical Legal Studies Movement,” 98 Harvard
L. Rev. 561 (1983).
57
interpret, and that therefore their decisions are essentially purely political. Typically, the
constitutional, statutory and common law materials with which judges must deal is made up
of widely overlapping rules and exceptions, conflicting principles and standards, and open
ended directives susceptible of contradictory interpretations. Accordingly, judges always have
a choice among various plausible alternatives with differing political consequences. Because
of this the judicial decision is as political as the legislative or executive one. But it is couched
in legal rather than political terms, which often allows it to conceal its politics, thus frequently
escaping the adverse reactions that typically confront controversial legislative enactments or
executive policies.
If, as CLS contends, law is indeterminate and judicial decisions both unpredictable and
political, then the only significant difference between a rule of law regime and one that is not
relates to how politics are actually conducted and implemented. This difference is by no
means trivial as a one person dictatorship without any pretense to state adherence to law
seems clearly less desirable than a state in which politics are apportioned among three
distinct branches of government. Nonetheless, so long as the rule of law cannot at all rise
above politics, then the benefits it might yield would seem meager indeed.
CSL’s critique has focused mainly on the United States where in their eyes the
business interests of advanced capitalism are largely if not exclusively dominant. Accordingly,
judge made law may be reducible to politics, but not all politics. Indeed, in CLS’s view
American judges are predominantly issued from, and sympathetic to, the ruling elites, and thus
prone to making use of the inherent indeterminacy of law to produce outcomes that benefit

58
dominant interests.
If CLS is right about this, however, it undermines its own position, at least in part. If
judges’ conclusions can be generally expected to be consistent with dominant pro-business
interests, then though law may be indeterminate and political, it would still, by and large, be
predictable -- at least in those cases in which representatives of the dominant elites are pitted
against adversaries with incompatible interests or ideologies. Consistent with this, moreover,
judges would be constrained not by the legal rules they must apply nor by the intent of the
legislator, but by the dictates of their own ideology. And, particularly in a common law system
where judges elaborate the law piecemeal by stringing together precedents a commonly
shared ideology may infuse judicial interpretation and judicial law making with a sufficient set
of directives (about which judges may be in part conscious, in part unconscious, or both) to
yield outcomes which, if not altogether predictable, would nevertheless fall within a fairly
narrow range of expectations. As we shall see below, this insight derived from CLS’s critique
can be extricated from their own conclusions and used to buttress a conception of a working
rule of law regime relying on a common law type of adjudication.
Another critique of CLS accepts the insight that all law -- including judge made law --
is political, but rejects the conclusion that all law is but politics. This latter critique zeroes in
on CLS’s negative assessment of rights, and is particularly telling and stinging, because it has
been articulated by proponents of critical race theory rather than by mainstream or
conservative critics who might easily be cast as spokespersons for the dominant elites124.
124
For a forceful and eloquent statement of the argument against CLS’s “trashing” of
rights, see Kimberle Creshaw, “Race, Reform, and Retrenchment: Transformation and
59
In a nutshell, CSL’s critique echoes the Marxist conclusion, that liberal rights, which purport to
be universal in scope and equally applicable to all, are in practice but thinly veiled
prescriptions designed to advance bourgeois interests125. In the eyes of proponents of critical
race theory, however, rights may be skewed in favor of the powerful, but have nonetheless
allowed oppressed racial minorities to experience significant progress. In other words, taking
into account the changes that occurred since the Supreme Court decreed apartheid
unconstitutional in 1954, African-Americans in the United States are at least somewhat better
off as a consequence of the spread of civil rights than they would have been without their
deployment.
Two important conclusions emerge from the standpoint of establishing the possibility
of the rule of law. First, even if law is but politics that does not entail that it is necessarily
unpredictable. And, second, even if all law is political, that does not foreclose that it contains
some residue rising above mere politics. These conclusions may seem relatively modest, but
as we shall now see, they provide a solid launching pad for the elaboration of a cogent
conception of the rule of law in a contemporary constitutional democracy steeped in the
common law such as the United States.
Legitimation in Antidiscrimination Law”, 101 Harvard L. Rev.1331 (1988).

125
Among the most systematic CLS critiques of rights along these lines is Mark
Tushnet, ”An Essay on Rights”, 62 Texas L. Rev. 1363 (1984).
60
VII The Common Law and the Tension

Between Predictability and Fairness
If we return to the ideal of the common law on its pristine form, and recall the example
concerning various animals discussed above 126, we can see that the common law proceeds
in a way that is in some sense diametrically opposed to the way civil law is supposed to be
implemented in the context of the Rechtsstaat or the État légal. In the civil law setting, the
judge is supposed to apply a previously enacted law to a set of facts in a deductive process
modeled on the syllogism. The legal rule figures as the major premise; the facts of the case,
as the minor premise; and the syllogistically derived judicial decision, as the conclusion. In
contrast, the common law not only involves an inductive process as already noted, but also a
future oriented act of law making grounded in the very process of adjudicating a present
dispute concerning past acts. Referring back to the example of the trespassing cow
discussed above 127, no rule appraising the parties involved of likely legal consequences
existed at the time of the act giving rise to litigation, this rule (although then incomplete) being
announced only at the moment of adjudication. On the other hand, the very act of adjudication
constitutes an announcement to all cow owners of how like cases will be adjudicated in the
future. Consistent with this, moreover, the conscientious common law judge should be at least
equally concerned -- if not more so -- with the future effect of his or her ruling than with
achieving backward looking justice in the dispute at hand. And to the extent that this is so, the
common law, in contrast to the civil law, involves both an act of (judicial) legislation and an act
126
See, supra at —
127
See, supra, at ----
61
of adjudication which is not sufficiently constrained by any pre-existing rule of have been
logically predictable.
Consistent with theses observations, one would have to conclude that a pure common
law system leads to the very opposite of the rule of law were it not predictable in its own way
as suggested by the above examination of CLS’s critique. Moreover, ideally, the common law
would be significantly predictable not because it is the handmaiden of the dominant elite’s
ideology, but because it is grounded in a common well of values, a widely shared sense of
justice and fairness, and dedication to elaborating a pragmatically oriented empirically based
working legal order that insures stability through steadfast adherence to core principles. In
other words, the common law can be compatible with the rule of law so long as it can satisfy
well settled expectations at the level of values and principles, even if not at that of particular
outcomes. Furthermore, as the common law evolves and precedents accumulate, presumably
fluctuations between expectations and particular outcomes tend by and large to narrow. In the
last analysis, the civil law cannot in practice remain a perfect deductive syllogistic system, as
rules proliferate, become riddled with exceptions and are not entirely consistent with one
another. Under such circumstances, in an increasing number of complex cases the actual
outcomes of judicial adjudication cannot be fully predicted. Accordingly, though they proceed
differently, in the real world common law and civil law would seem poised to converge at least
in terms of predictability.
As already mentioned, stabilizing expectations in an increasingly complex world is one

62
of the paramount functions of contemporary law128. Upon first impression it would seem that
a regime of fairly rigid civil law type rules would be far preferable for this purpose than a set
of evolving common law standards -- or loosely construed statutory provisions interpreted in
accordance with broad common law canons. Upon further consideration, however, common
law methodology may be ultimately better attuned to the dual task of stabilizing expectations
and meeting evolving needs in a rapidly changing economic environment. Indeed, so long as
the interests of legislators, judges and private parties converge, expectations may be kept
relatively stable even in the face of frequent adaptation to novel conditions. While prediction
of actual outcomes may be more difficult, confidence in the maintenance of accepted
standards concurrently with adaptation to rapidly evolving needs may well provide the best
possible means toward stabilizing expectations in the polities at the forefront of economic
development.
Thus, the convergence of interests decried by CLS can furnish a key element of
stability to any rule of law project grounded in the common law. But this stability cannot extend
to those situations in which divergence rather than convergence becomes the norm. In the
latter case, the rule of law must turn to protection of fundamental rights for sustenance.
Moreover in periods in which the scope of fundamental rights is in flux129, focus on rights may
generate a significant degree of unpredictability. On the other hand, as the core fundamental
128
See supra ----
129
For example, the U.S. Supreme Court during the period in which Earl Warren was
Chief Justice (1953 - 1969) greatly expanded the scope and breath of fundamental
constitutional rights. See generally, Thomas Walker & Lee Epstein, The Supreme Court of
the United States: An Introduction (1993).
63
rights tend to become universal in nature and scope--through transnational norms such as
those enshrined in the United Nation’s Human Rights covenants or the European Convention
on Human Rights -- the protection of fundamental rights seems poised to become more
predictable. And, to the extent that both the requirements imposed by the protection of
fundamental rights and those necessitated by the stabilization of expectations were to
become fairly stable, even the conflicts among the two might become largely predictable.
In any pluralist society with diverging conceptions of the good, there may be as wide
a lack of consensus concerning what ought to count as a fundamental right as regarding what
would constitute a fair means to stabilize legal expectations. Moreover, the more
disagreement there is on both those issues, the more existing legal norms are likely to be
perceived as a function of politics rather than of the institutionalization of the rule of law. More
generally, in terms of perception, the rule of law would seem to go hand in hand with a
relatively high level of integration among diverse perspectives, while the rule of politics would
seem tied to significant fragmentation within the polity. It is worth noting, however, that the
perceptions in question have little to do, strictly speaking, with whether legal outcomes are in
fact unpredictable. It may be that in a badly fragmented society, most are alienated from the
law but can nonetheless predict legal outcomes or even consider them as inevitable as they
are unfair. In contrast, in a polity rapidly moving towards greater integration, it may be that
widely supported constantly evolving fundamental rights upset expectations but promote
greater fairness and legitimacy. In short, if the rule of law depends on fairness then it should
be open to a large measure of unpredictability. If on the other hand, it is based above all on

64
predictability, than it should be compatible with widespread unfairness.
In the last analysis, the Anglo-American conception of the rule of law seems capable
of integrating a certain number of formal elements, a workable amount of predictability, a
certain sense of fairness, and certain contextual elements such as relative integration as
opposed to unbridgeable fragmentation and a cultural predilection for a procedurally based
array of individual rights available against the state and legislative majorities. Moreover, this
conception of the rule of law depends on the maintenance of a division of labor between law
and politics, but not on one that purges law of all politics. Instead, the division in question is
a contextual one that is triggered by a shift in perspective.
As I have argued at greater length elsewhere,130 the distinction between law and
politics turns on whether a decision concerning some past occurrence can be reconstructed
in terms of the application of standards that (upon reconstruction) plausibly cast the eventual
result as ex ante just and predictable. If such reconstruction is possible, then the decision
institutes legal justice and is compatible with the rule of law. Otherwise, the decision may
promote political justice, but cannot transcend the realm of politics. Ultimately, the distinction
between law and politics reflects above all a difference in perspective. From law’s
perspective -- and hence from that of the rule of law -- “legal argument and legal discourse
occupy the foreground, with political arguments and values receding to the background.
Consistent with this, insurmountable ruptures in legal discourse and irreparable breakdown
130
See Michel Rosenfeld, JUST INTERPRETATIONS, supra, at 74 - 82.
65
in legal argumentation signal a failure to come within the ambit of justice according to law”.131
From the perspective of politics, in contrast, political concerns and values come to the
forefront and law recedes to the background. Political decisions may by cast in the discourse
of law, but so long as their political elements remain predominant and upset logic and
continuity, they cannot qualify as genuinely legal or as properly falling within the ambit of law.
A division of labor between political and legal decisions is certainly compatible with the rule
of law, but not one in which all decisions legal in form turn out to be political in perspective.
VIII Conclusion:
Constitutional Democracy and the Limits
of The Rule of Law
Notwithstanding that it raises certain paradoxes and that it is often prone to blur the
divide between law and politics, the American legal system based on a written constitution
and a well entrenched common law approach can certainly satisfy the requirements of a rule
of law regime. Because of its apparent greater flexibility than its civil law counterparts, the
American system seems better suited to deploy a coherent rule of law regime provided there
is a high degree of consensus on core values and objectives, on a sense of fairness, and on
an essential bundle of constitutional rights. Conversely, in the context of a serious breakdown
of consensus, the American system seems less suited to maintenance of a rule of law regime
131
Id., at 82 - 83.
66
than a traditional civil law system based on a more rigid conception of legality132.
At least under certain propitious circumstances therefore, the rule of law can promote
both predictability and fairness, and this seems equally possible in an Anglo-American
common law setting and in a continental civil law system. Beyond that, however, it is not clear
whether the link between the rule of law and predictability and fairness is an intrinsic or an
extrinsic one. In other words, does the rule of law in certain circumstances make for
predictability and fairness? Or does it merely coincide with them?
Consistent with the modified Habermasian consensus-based criterion of legitimacy
described above 133, predictability is crucial for purposes of assessing the normative weight
attributable to ex-ante acquiescence. Fairness, on the other hand, breaks down into a
subjective and an objective component. If a citizen implicitly or explicitly endorses a law or
legal regime, the latter can be considered subjectively fair. Conversely, in the absence of such
endorsement, if consistent with a citizen’s conception of the good (or regardless of such
conception), it would be reasonable for such citizen to endorse a law or legal regime, then the
132
Indeed, a fairly rigid civil law system based on a deductive reasoning model would
be more prone to being predictable than a common law system in the hands of judges divided
by ideology, political aim, and judicial philosophy. Moreover, under such circumstances, a
common law regime would not be likely to being widely perceived as promoting fairness. It
should be emphasized, however, that contemporary civil law constitutional regimes are far
less likely to differ sharply from common law regimes than their more traditional predecessors.
See, e.g. , Dominique Rousseau, “The Constitutional Judge: Master or Slave of the
Constitution?” In CONSTITUTIONALISM, IDENTITY AND DIFFERENCE 261, supra note ....
(discussing role of constitutional judge in France) and Bernhard Schlink, “German
Constitutional Culture in Transition,” in id. at 197 (discussing broad powers of German
Constitutional Court).
133
See supra, at -----.
67
latter can be said to be objectively fair134 .
In accordance with the modified Habermasian criterion of legitimacy elaborated above,
existence of an extrinsic link between the rule of law and fairness would only seem possible
in the context of a broad consensus on extra-legal norms and values, or in other words, in the
case of a strong convergence of conceptions of the good135.
In a pluralist society with competing conceptions of the good, however, it is difficult to
imagine the prevalence of a solid extrinsic link between the rule of laws and both predictability
and fairness. Indeed, given significant divergences stemming from competing conceptions
of the good, it seems inevitable that particular laws and even legal regimes taken as a whole
would be bound to promote certain contested interests or conceptions of the good at the
expense of others. Accordingly, in the context of a pluralist society, satisfaction of the
modified Habermasian criterion of legitimacy would seem dependent on forgoing an intrinsic
link between the rule of law and both predictability and fairness.
The rule of law as developed in each of the three traditions examined above ultimately
falls short from the standpoint of establishing the requisite intrinsic link. The Rechtsstaat fails
to secure the requisite intrinsic link in all three of its versions discussed above. As we have
seen, the Kantian Rechtsstaat aspires to rise above the realm of interests, but ends up being
134
In other words, subjective fairness is measured from a participant’s perspective
while objective fairness is determined from an observer’s perceptive that properly accounts
for the relevant participants’ conceptions of the good.
13 5
Under such circumstances, the law’s (subjective) fairness would derive from the
substantive vision behind it, and its predictability may be sufficiently grounded on an
expectation of fair and orderly outcomes regardless of whether a legal contest could be
confidently predicted prior to adjudication.
68
either impossible or biased. The positivistic Rechsstaat, on the other hand, pins its hopes
on strict adherence to the principle of legality and thus promotes predictability, but gives no
assurance of fairness. Finally, the contemporary Rechsstaat or Verfassungsstaat aims at
fairness through constitutionalization of substantive norms and values, such as human dignity,
but fails to promote consensus in as much as it endorses certain contested values at the
expense of others.
The French Rousseauian État légal and the État de droit also fail to satisfy the
requisite intrinsic link, and thus fall short of the modified Habermasian criterion of legitimacy.
The État Légal and its mission to promote the general will can certainly contribute to the
satisfaction of the above criterion through promotion of democratic self-government which may
lend legitimacy to a number of legal norms operating at the infra-constitutional level. However,
to the extent that the general will remains too abstract or shrouded in mystery, and that self-
government by itself cannot guarantee fairness, then the État Légal standing alone cannot be
relied upon to produce the required legitimacy. This was understood by the proponents of the
État de droit , but supplementing the État Légal with the État de droit does not solve the
problem of legitimacy in as much as the resulting constitutional regime is no more inherently
capable than its American or German counterparts to overcome reliance on either insufficient
formalistic criteria – such as those based on Kantian or Lockean conceptions of rights -- or
on contested substantive values such as those embedded in the German Basic Law. Finally,
as already mentioned, in the absence of a consensus concerning a sufficient core of common
values, the American common law based constitutionally shaped rule of law regime cannot

69
alone guarantee maintenance of an intrinsic link between implementation of the rule of law and
of predictability and fairness.
There can be little question that in the absence of commitment to legality, fundamental
rights guarantees, and genuine opportunities for citizen participation in the political process,
satisfaction of the modified Habermasian criterion of legitimacy would be altogether
impossible. Accordingly, adherence to the rule of law -- in one of the three traditions
considered above -- emerges as a necessary but insufficient means to legitimacy. Moreover,
as we have seen in the course of examining these three rule of law regimes, each of them is
much more readily justified in terms of what it stands against rather than in terms of what it
purports to stand for. The challenge, therefore, is to find a rule law regime that could be used
to counter the tendency of existing liberal democratic constitutional rule of law regimes to
frustrate recognition and accommodation of identity related claims in multicultural and
multiethnic polities. What this regime might look like is still difficult to say. But it would have
to involve more than procedural safeguards and traditional liberal rights.
It seems clear that the gains brought by the implementation of the rule of law regimes
discussed above should not be squandered. It is an open question, however, what the next
step should be. To the extent that the regimes in question seem insufficient to properly
accommodate a plurality of conceptions of the good, the most important task would be to
devise better means to accommodate pluralism. It is not clear how much of this could be
accomplished through transformation of the rule of law and how much through establishment
of institutions and practices beyond the rule of law. But given the failure of the rule of law to

70
secure the requisite internal links between law, predictability and fairness, it seems
reasonable to assume that at least some of the necessary changes will have to take place
beyond the realm of law.

Rule of Law

Uploaded by

Copyright:

Available Formats

Rule of Law

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

Rule of Law

Uploaded by

Copyright:

Available Formats

1

Cardozo Law School

Working Paper Series No. 36

THE RULE OF LAW, AND THE

This paper can be downloaded without charge from the

a: RULE OF LAW.01 March 27, 2001

THE RULE OF LAW, AND THE

The rule of law is a cornerstone of contemporary constitutional democracy1 as

totalitarian regimes2 to constitutional democracy in Eastern Europe and elsewhere3. In the

and Diversity” in CONSTITUTIONALISM, IDENTITY, DIFFERENCE AND LEGITIMACY:

Poland: The Emergence of a Rechtsstaat?”, 31 Stanford J. of Intl. L. 35 (1995); “Special

democracy as an indispensable ingredient of constitutionalism. Beyond that, however, it is

constitutional democracy or what specific role it must assume to ensure a working

constitutional democracy, or how it might ultimately contribute to the latter’s legitimacy.

enforceable and as the highest law of the land).

maintenance of a prescribed constitutional order8.

disagreements concerning the relevant prescriptive standards at stake10. Consistent with

See Richard H. Fallon, Jr., “The Rule of Law“ as a Concept in Constitutional

Discourse”, 97 Colum. L. Rev. 1, 7 (1997).

be largely, if not exclusively, procedural or substantive, or whether it should be primarily

institutional framework necessary for constitutional democracy, and of implementation of the

often against the citizen12. In contrast, in connection with protection of fundamental

the legitimacy of constitutional democracy in a contemporary pluralistic society, it is necessary

likely fail to live up to the requirements of the rule of law.

II. Background Presuppositions:

occur in a vacuum. As already mentioned, constitutional democracy looms as virtually

indispensable when contrasted to the totalitarian or authoritarian regimes which it replaced

democracy could be superfluous, or even undesirable. For example, in a close knit

appropriate than a constitutional democracy. Furthermore, in such a society instructions and

the rule of law.

In contrast, in heterogeneous societies with various competing conceptions of the

for purposes of achieving political cohesion with a minimum of oppression13. Such

heterogenous societies, moreover, can be characterized as being pluralistic - in-fact14. A

society is pluralistic-in-fact if it is divided along ethnic, religious, linguistic, cultural or

states qualify as being pluralistic-in-fact.

pluralistic-in-fact, the legitimacy of fundamental political institutions must ultimately depend on

theory of the justification of law20.

Leaving aside differences among various consent-based theories, legitimation based

unmistakable attractiveness when contrasted with authoritarian or totalitarian regimes,

objectives with which such minorities may strongly

See Jürgen Habermas, BETWEEN FACTS AND NORMS: CONTRIBUTIONS TO

enforcement of constitutional rights would seem to lead to a significant amount of coercion22.

challenged infra-constitutional laws.

be experienced as coercive would not necessarily impair their legitimacy.

a: RULE OF LAW.01 March 27, 2001

satisfy the prescriptions of the “rule of law in the narrow sense”.

insufficient for purposes of satisfying the minimum requirements of a legitimate constitutional

See Richard H. Fallon, Jr. “The Rule of Law” as a Concept in Constitutional

as it is entirely compatible with legal regimes predicated on slavery, apartheid or countless

the narrow sense.

To become legitimate the rule of law would seem to have to be democratically

these requirements may be necessary without being sufficient to produce legitimacy.

particularly in pluralist settings marked by clashing conceptions of the good.

a constitutional democracy. Moreover, even if we assume that as shields against ordinary

to parochial school, the Supreme Court’s interpretation of the Establishment Clause as

advantageous or acceptable by some and oppressive or coercive by others.

narrowly antimajoritarian constitutional constraints ought to be construed without questioning

concerning certain fundamental constitutional constraints and a shared commitment to