Copyright (c) The Harvard Law Review Association 1994.
Harvard Law Review
107 Harv. L. Rev. 1765
LENGTH: 16231 words
POLITICAL LIBERALISM. By John
New York: Columbia University Press. 1993. Pp. xxxiv, 401. $ 29.95.
Reviewed by Michael J. Sandel
... Rare is the work of political philosophy that provokes sustained debate.
Political Liberalism defends the priority of right by detaching it from the Kantian conception of
the person, how convincing is its defense? As I shall try to argue,
Political Liberalism rescues the priority of the right from controversies about the nature of the
self only at the cost of rendering it vulnerable on other grounds. ... How,
political liberalism escape the company of Douglas and oppose slavery without presupposing some
comprehensive moral view? It might be replied that Douglas was wrong to seek
social peace at any price; not just any political agreement will do. ... The
restrictive character of liberal public reason can also be seen in the debate
about gay rights. ...
[*1765] Rare is the work of political philosophy that provokes sustained debate. It
is a measure of its greatness that John
A Theory of Justice
n3 inspired not one debate, but three.
The first, by now a starting point for students of moral and political
philosophy, is the argument between utilitarians and rights-oriented liberals.
Should justice be founded on utility, as Jeremy Bentham and John Stuart Mill
argue, or does respect for individual rights require a basis for justice
independent of utilitarian considerations, as Kant and
Rawls maintain? Before
Rawls wrote, utilitarianism was the dominant view within Anglo-American moral and
political philosophy. Since
A Theory of Justice, rights-oriented liberalism has come to predominate.
The second debate inspired by
Rawls's work is an argument within rights-oriented liberalism. If certain individual
rights are so important that even considerations of the general welfare cannot
override them, it remains to ask which rights these are. Libertarian liberals,
like Robert Nozick and Friedrich Hayek, argue that government should respect
basic civil and political liberties, and also the right to the fruits of our
labor as conferred by the market economy; redistributive policies that tax the
rich to help the poor thus violate our rights.
n5 Egalitarian liberals like
Rawls disagree. They argue that we cannot meaningfully exercise our civil and
political liberties without the provision of basic social and economic needs;
government should therefore assure each person, as a matter of right, a decent
level of such goods as education, income, housing, health care, and the like.
The debate between the libertarian and egalitarian versions of rights-oriented
liberalism, which flourished in the academy in the 1970s, corresponds roughly
to the debate in American politics, familiar since the New
[*1766] Deal, between defenders of the market economy and advocates of the welfare
The third debate prompted by
Rawls's work centers on an assumption shared by libertarian and egalitarian liberals
alike. This is the idea that government should be neutral among competing
conceptions of the good life. Despite their various accounts of what rights we
have, rights-oriented liberals agree that the principles of justice that
specify our rights should not depend for their justification on any particular
conception of the good life.
n6 This idea, central to the liberalism of Kant,
Rawls, and many contemporary liberals, is summed up in the claim that the right is
prior to the good.
I. CONTESTING THE PRIORITY OF THE RIGHT OVER THE GOOD
Rawls, as for Kant, the right is prior to the good in two respects, and it is
important to distinguish them. First, the right is prior to the good in the
sense that certain individual rights
"trump," or outweigh, considerations of the common good. Second, the right is prior to
the good in that the principles of justice that specify our rights do not
depend for their justification on any particular conception of the good life.
It is this second claim for the priority of the right that prompted the most
recent wave of debate about Rawlsian liberalism, an argument that has
flourished in the last decade under the somewhat misleading label of the
A number of political philosophers writing in the 1980s took issue with the
notion that justice can be detached from considerations of the good.
Challenges to contemporary rights-oriented liberalism found in the writings of
n8 Charles Taylor,
n9 Michael Walzer,
[*1767] and also in my own work,
n11 are sometimes described as the
"communitarian" critique of liberalism. The term
"communitarian" is misleading, however, insofar as it implies that rights should rest on the
values or preferences that prevail in any given community at any given time.
Few, if any, of those who have challenged the priority of the right are
communitarians in this sense.
n12 The question is not whether rights should be respected, but whether rights can
be identified and justified in a way that does not presuppose any particular
conception of the good. At issue in the third wave of debate about
Rawls's liberalism is not the relative weight of individual and communal claims, but
the terms of relation between the right and the good.
n13 Those who dispute the priority of the right argue that justice
[*1768] is relative to the good, not independent of it. As a philosophical matter,
our reflections about justice cannot reasonably be detached from our
reflections about the nature of the good life and the highest human ends. As a
political matter, our deliberations about justice and rights cannot proceed
without reference to the conceptions of the good that find expression in the
many cultures and traditions within which those deliberations take place.
Much of the debate about the priority of the right has focused on competing
conceptions of the person and of how we should understand our relation to our
ends. Are we, as moral agents, bound only by the ends and roles we choose for
ourselves, or can we sometimes be obligated to fulfill certain ends we have not
chosen -- ends given by nature or God, for example, or by our identities as
members of families, peoples, cultures, or traditions? In various ways, those
who have criticized the priority of right have resisted the notion that we can
make sense of our moral and political obligations in wholly voluntarist or
A Theory of Justice,
Rawls linked the priority of the right to a voluntarist, or broadly Kantian,
conception of the person. According to this conception, we are not simply
defined as the sum of our desires, as utilitarians assume, nor are we beings
whose perfection consists in realizing certain purposes or ends given by
nature, as Aristotle held. Rather, we are free and independent selves, unbound
by antecedent moral ties, capable of choosing our ends for ourselves. This is
the conception of the person that finds expression in the ideal of the state
[*1769] as a neutral framework. It is precisely because we are free and independent
selves, capable of choosing our own ends, that we need a framework of rights
that is neutral among ends. To base rights on some conception of the good
would impose on some the values of others and so fail to respect each person's
capacity to choose his or her own ends.
This conception of the person, and its link to the case for the priority of the
right, are expressed throughout
A Theory of Justice. Its most explicit statement comes toward the end of the book, in
Rawls's account of
"the good of justice." There
Rawls argues, following Kant, that teleological doctrines are
"radically misconceived" because they relate the right and the good in the wrong way:
We should not attempt to give form to our life by first looking to the good
independently defined. It is not our aims that primarily reveal our nature but
rather the principles that we would acknowledge to govern the background
conditions under which these aims are to be formed and the manner in which they
are to be pursued. For the self is prior to the ends which are affirmed by it;
even a dominant end must be chosen from among numerous possibilities. . . . We
should therefore reverse the relation between the right and the good proposed
by teleological doctrines and view the right as prior.
A Theory of Justice, the priority of the self to its ends supports the priority of the right to the
"[A] moral person is a subject with ends he has chosen, and his fundamental
preference is for conditions that enable him to frame a mode of life that
expresses his nature as a free and equal rational being as fully as
n15 The notion that we are free and independent selves, unclaimed by prior moral
ties, assures that considerations of justice will always outweigh other, more
particular aims. In an eloquent expression of Kantian liberalism,
Rawls explains the moral importance of the priority of the right in the following
[T]he desire to express our nature as a free and equal rational being can be
fulfilled only by acting on the principles of right and justice as having first
priority. . . . It is acting from this precedence that expresses our freedom
from contingency and happenstance. Therefore in order to realize our nature we
have no alternative but to plan to preserve our sense of justice as governing
our other aims. This sentiment cannot be fulfilled if it is compromised and
balanced against other ends as but one desire among the rest. . . . [H]ow far
we succeed in expressing our nature depends upon how consistently we act from
our sense of justice as finally regulative. What we cannot do is express our
nature by following a plan that views the sense of justice as but
[*1770] one desire to be weighed against others. For this sentiment reveals what the
person is, and to compromise it is not to achieve for the self free reign but
to give way to the contingencies and accidents of the world.
In different ways, those who disputed the priority of the right took issue with
Rawls's conception of the person as a free and independent self, unencumbered by prior
n17 They argued that a conception of the self given prior to its aims and
attachments could not make sense of certain important aspects of our moral and
political experience. Certain moral and political obligations that we commonly
recognize -- such as obligations of solidarity, for example, or religious
duties -- may claim us for reasons unrelated to a choice. Such obligations are
difficult to dismiss as merely confused, and yet difficult to account for if we
understand ourselves as free and independent selves, unbound by moral ties we
have not chosen.
II. DEFENDING THE PRIORITY OF THE RIGHT OVER THE GOOD
Political Liberalism, Rawls defends the priority of the right over the good. He sets aside, for the most
part, issues raised in the first two waves of debate, about utility versus
rights and libertarian versus egalitarian notions of distributive justice.
Political Liberalism focuses instead on issues posed by the third wave of debate, about the
priority of the right.
Given the controversy over the Kantian conception of the person that supports
the priority of the right, at least two lines of reply are possible. One is to
defend liberalism by defending the Kantian conception of the person; the other
is to defend liberalism by detaching it from the Kantian conception. In
Political Liberalism, Rawls adopts the second course. Rather than defend the Kantian conception of the
person as a moral ideal, he argues that liberalism as he conceives it does not
depend on that conception of the person after all. The priority of the right
over the good does not presuppose any particular conception of the person, not
even the one advanced in Part III of
A Theory of Justice.
A. Political Versus Comprehensive Liberalism
The case for liberalism,
Rawls now argues, is political, not philosophical or metaphysical, and so does not
depend on controversial claims about the nature of the self (pp. 29-35). The
priority of the right over the good is not the application to politics of
Kantian moral philosophy, but a practical response to the familiar fact that
people in modern democratic societies typically disagree about the good.
Because people's moral and religious convictions are unlikely to converge, it
is more reasonable to seek agreement on principles of justice that are neutral
with respect to those controversies (pp. xvi-xvii).
Rawls's revised view is the distinction between
political liberalism and liberalism as part of a comprehensive moral doctrine (pp. 154-58).
Comprehensive liberalism affirms liberal political arrangements in the name of
certain moral ideals, such as autonomy or individuality or self-reliance.
Examples of liberalism as a comprehensive moral doctrine include the liberal
visions of Kant and John Stuart Mill.
Rawls acknowledges, the version of liberalism presented in
A Theory of Justice is also an instance of comprehensive liberalism.
"An essential feature of a well-ordered society associated with justice as
fairness is that all its citizens endorse this conception on the basis of what
I now call a comprehensive philosophical doctrine" (p. xvi). It is this feature that
Rawls now revises, by recasting his theory as a
"political conception of justice" (p. xvi).
Unlike comprehensive liberalism,
political liberalism refuses to take sides in the moral and religious controversies that arise from
comprehensive doctrines, including controversies about conceptions of the self.
"Which moral judgments are true, all things considered, is not a matter for
political liberalism" (p. xx).
"To maintain impartiality between comprehensive doctrines, it does not
specifically address the moral topics on which those doctrines divide" (p. xxviii). Given the difficulty of securing agreement on any comprehensive
conception, it is unreasonable to expect that, even in a well-ordered society,
all people will support liberal institutions for the same reason -- as
expressing the priority of the self to its ends, for example.
Political liberalism abandons this hope as unrealistic and contrary to the aim of basing justice on
principles that adherents of various moral and religious conceptions can
accept. Rather than seek a philosophical foundation for principles of justice,
political liberalism seeks the support of an
"overlapping consensus" (p. 134). This means that different
[*1772] people can be persuaded to endorse liberal political arrangements, such as
equal basic liberties, for different reasons, reflecting the various
comprehensive moral and religious conceptions they espouse. Because
political liberalism does not depend for its justification on any one of those moral or religious
conceptions, it is presented as a
"freestanding" view; it
"applies the principle of toleration to philosophy itself" (p. 10).
political liberalism renounces reliance on the Kantian conception of the person, it does not do
without a conception of the person altogether. As
Rawls acknowledges, some such conception is necessary to the idea of the original
position, the hypothetical social contract that gives rise to the principles of
justice. The way to think about justice,
Rawls argued in
A Theory of Justice, is to ask which principles would be agreed to by persons who found themselves
gathered in an initial situation of equality, each in temporary ignorance of
his or her race and class, religion and gender, aims and attachments.
n20 But in order for this way of thinking about justice to carry weight, the
design of the original position must reflect something about the sort of
persons we actually are, or would be in a just society.
One way of justifying the design of the original position would be to appeal to
the Kantian conception of the person that
Rawls advanced in Part III of
A Theory of Justice. If our capacity to choose our ends is more fundamental to our nature as moral
persons than are the particular ends we choose, if
"[i]t is not our aims that primarily reveal our nature but rather the principles
that we would acknowledge to govern the background conditions under which these
aims are to be formed,"
"the self is prior to the ends which are affirmed by it,"
n22 then it makes sense to think about justice from the standpoint of persons
deliberating prior to any knowledge of the ends they will pursue. If
"a moral person is a subject with ends he has chosen, and his fundamental
preference is for conditions that enable him to frame a mode of life that
expresses his nature as a free and equal rational being as fully as
n23 then the original position can be justified as an expression of our moral
personality and the
"fundamental preference" that flows from it.
Rawls disavows reliance on the Kantian conception of the person, however, this way
of justifying the original position is no longer available. But this raises a
difficult question: what reason remains for insisting that our reflections
about justice should proceed without reference to our purposes and ends? Why
"bracket," or set aside, our moral and religious convictions, our conceptions of
[*1773] the good life? Why should we not base the principles of justice that govern
the basic structure of society on our best understanding of the highest human
B. The Political Conception of the Person
Political liberalism replies as follows: the reason we should think about justice from the
standpoint of persons who abstract from their ends is not that this procedure
expresses our nature as free and independent selves given prior to our ends.
Rather, this way of thinking about justice is warranted by the fact that, for
political purposes, though not necessarily for all moral purposes, we should think of
ourselves as free and independent citizens, unclaimed by prior duties or
obligations (pp. 29-35). For
political liberalism, what justifies the design of the original position is a
"political conception of the person" (p. 29). The political conception of the person embodied in the original
position closely parallels the Kantian conception of the person, with the
important difference that its scope is limited to our public identity, our
identity as citizens. Thus, for example, our freedom as citizens means that
our public identity is not claimed or defined by the ends we espouse at any
given time. As free persons, citizens view themselves
"as independent from and not identified with any particular such conception with
its scheme of final ends" (p. 30). Our public identity is not affected by changes over time in our
conceptions of the good.
In our personal or nonpublic identity,
Rawls allows, we may regard our
"ends and attachments very differently from the way the political conception
supposes" (p. 31). There, persons may find themselves claimed by loyalties and
"they believe they would not, indeed could and should not, stand apart from and
evaluate objectively. They may regard it as simply unthinkable to view
themselves apart from certain religious, philosophical, and moral convictions,
or from certain enduring attachments and loyalties" (p. 31). But however encumbered we may be in our personal identities, however
claimed by moral or religious convictions, we must bracket our encumbrances in
public, and regard ourselves,
qua public selves, as independent of any particular loyalties or attachments or
conceptions of the good (p. 31).
A related feature of the political conception of the person is that we are
"self-authenticating sources of valid claims" (p. 32). The claims we make as citizens carry weight, whatever they are,
simply by virtue of our making them (provided they are not unjust). That some
claims may reflect high moral or religious ideals, or notions of patriotism and
the common good, while others express mere interests or preferences, is not
relevant from the standpoint of
political liberalism. From a political point of view, claims founded on duties and obligations of
[*1774] citizenship or solidarity or religious faith are just things people want --
nothing more, nothing less. Their validity as political claims has nothing to
do with the moral importance of the goods they affirm, but consists solely in
the fact that someone asserts them. Even divine commandments and imperatives
of conscience count as
"self-authenticating" claims, politically speaking.
n24 This ensures that even those who regard themselves as claimed by moral or
religious or communal obligations are nonetheless, for political purposes,
This political conception of the person explains why, according to
political liberalism, we should reflect about justice as the original position invites us to do, in
abstraction from our ends. But this raises a further question: why should we
adopt the standpoint of the political conception of the person in the first
place? Why should our political identities not express the moral and religious
and communal convictions we affirm in our personal lives? Why insist on the
separation between our identity as citizens and our identity as moral persons
more broadly conceived? Why, in deliberating about justice, should we set
aside the moral judgments that inform the rest of our lives?
Rawls's answer is that this separation or
"dualism" between our identity as citizens and our identity as persons
"originates in the special nature of democratic political culture" (p. xxi). In traditional societies, people sought to shape political life in
the image of their comprehensive moral and religious ideals. But in a modern
democratic society like our own, marked as it is by a plurality of moral and
religious views, we typically distinguish between our public and personal
identities. Confident though I may be of the truth of the moral and religious
ideals I espouse, I do not insist that these ideals be reflected in the basic
structure of society. Like other aspects of
political liberalism, the political conception of the person as a free and independent self is
"implicit in the public political culture of a democratic society" (p. 13).
Rawls is right, and the liberal self-image he attributes to us is implicit in our
political culture. Would this provide sufficient grounds for affirming it, and
for adopting the conception of justice it supports? Some have read
Rawls's recent writings as suggesting that justice as fairness, being a political
conception of justice, requires no moral or philosophical justification apart
from an appeal to the shared understandings implicit in our political culture.
Rawls seemed to
[*1775] invite this interpretation when he wrote, in an article published after
A Theory of Justice but before
Political Liberalism, as follows:
What justifies a conception of justice is not its being true to an order
antecedent to and given to us, but its congruence with our deeper understanding
of ourselves and our aspirations, and our realization that, given our history
and the traditions embedded in our public life, it is the most reasonable
doctrine for us.
Richard Rorty, in an insightful article, interprets (and welcomes)
Rawls's revised view as
"thoroughly historicist and antiuniversalist."
A Theory of Justice seemed to base justice on a Kantian conception of the person, Rorty writes,
"no longer seems committed to a philosophical account of the human self, but
only to a historico-sociological description of the way we live now."
n27 On this view,
Rawls is not
"supplying philosophical foundations for democratic institutions, but simply
trying to systematize the principles and intuitions typical of American
n28 Rorty endorses what he takes to be
Rawls's pragmatic turn, a turn away from the notion that liberal political
arrangements require a philosophical justification, or
"extrapolitical grounding" in a theory of the human subject.
"[I]nsofar as justice becomes the first virtue of a society," Rorty writes,
"the need for such legitimation may gradually cease to be felt. Such a society
will become accustomed to the thought that social policy needs no more
authority than successful accommodation among individuals, individuals who find
themselves heir to the same historical traditions and faced with the same
Political Liberalism, Rawls pulls back from this purely pragmatic account. Although justice as fairness
"by looking to the public culture itself as the shared fund of implicitly
recognized basic ideas and principles" (p. 8), it does not affirm these principles simply on the grounds that they
are widely shared. Though
Rawls argues that his principles of justice could gain the support of an overlapping
consensus, the overlapping consensus he seeks
"is not a mere modus vivendi" (p. 147), or compromise among conflicting views. Adherents of different moral
and religious conceptions begin by endorsing the principles of justice for
reasons drawn from within their own conceptions. But, if all goes well, they
come to support those principles as expressing important political values. As
people learn to live in a pluralist society governed by liberal institutions,
they acquire virtues that strengthen their commitment to liberal principles.
[*1776] The virtues of political cooperation that make a constitutional regime
possible are . . . very great virtues. I mean, for example, the virtues of
tolerance and being ready to meet others halfway, and the virtue of
reasonableness and the sense of fairness. When these virtues are widespread in
society and sustain its political conception of justice, they constitute a very
great public good . . . (p. 157).
Rawls emphasizes that affirming liberal virtues as a great public good and
encouraging their cultivation is not the same as endorsing a perfectionist
state based on a comprehensive moral conception. It does not contradict the
priority of the right over the good. The reason is that
political liberalism affirms liberal virtues for political purposes only -- for their role in
supporting a constitutional regime that protects people's rights. Whether and
to what extent these virtues should figure in people's moral lives generally is
a question that
political liberalism does not claim to answer (pp. 194-95).
Political Liberalism defends the priority of right by detaching it from the Kantian conception of
the person, how convincing is its defense? As I shall try to argue,
Political Liberalism rescues the priority of the right from controversies about the nature of the
self only at the cost of rendering it vulnerable on other grounds.
Specifically, I shall try to show that liberalism conceived as a political
conception of justice is open to three objections.
First, notwithstanding the importance of the
"political values" to which
Rawls appeals, it is not always reasonable to bracket, or set aside for political
purposes, claims arising from within comprehensive moral and religious
doctrines. Where grave moral questions are concerned, whether it is reasonable
to bracket moral and religious controversies for the sake of political
agreement partly depends on which of the contending moral or religious
doctrines is true.
political liberalism, the case for the priority of the right over the good depends on the claim that
modern democratic societies are characterized by a
"fact of reasonable pluralism" about the good (p. xvii). Though it is certainly true that people in modern
democratic societies hold a variety of conflicting moral and religious views,
it cannot be said that there is a
"fact of reasonable pluralism" about morality and religion that does not also apply to questions of justice.
Third, according to the ideal of public reason advanced by
political liberalism, citizens may not legitimately discuss fundamental political and constitutional
questions with reference to their moral and religious ideals. But this is an
unduly severe restriction that would impoverish political discourse and rule
out important dimensions of public deliberation.
A. Bracketing Grave Moral Questions
Political liberalism insists on bracketing our comprehensive moral and religious ideals for
political purposes, and on separating our political from our personal
identities. The reason is this: in modern democratic societies like ours,
where people typically disagree about the good life, bracketing our moral and
religious convictions is necessary if we are to secure social cooperation on
the basis of mutual respect. But this raises a question that
political liberalism cannot answer within its own terms. Even granting the importance of securing
social cooperation on the basis of mutual respect, what is to guarantee that
this interest is always so important as to outweigh any competing interest that
could arise from within a comprehensive moral or religious view?
One way of assuring the priority of the political conception of justice (and
hence the priority of the right) is to deny that any of the moral or religious
conceptions it brackets could be true.
n30 But this would implicate
political liberalism in precisely the sort of philosophical claim it seeks to avoid. Time and
Rawls emphasizes that
political liberalism does not depend on skepticism about the claims of comprehensive moral and
religious doctrines. If
political liberalism therefore allows that some such doctrines might be true, then what is to
assure that none can generate values sufficiently compelling to burst the
brackets, so to speak, and morally outweigh the political values of toleration,
fairness, and social cooperation based on mutual respect?
One might reply that political values and values arising from within
comprehensive moral and religious doctrines address different subjects.
Political values, one might say, apply to the basic structure of society and to
constitutional essentials, whereas moral and religious values apply to the
conduct of personal life and voluntary associations. But if it were simply a
difference of subject matter, no conflict between political values and moral
and religious values could ever arise, and there would be no need to assert, as
Rawls repeatedly does, that in a constitutional democracy governed by
"political values normally outweigh whatever nonpolitical values conflict with
them" (p. 146).
The difficulty of asserting the priority of
"political values" without reference to the claims of morality and religion can be seen by
considering two political controversies that bear on grave moral and religious
questions. One is the contemporary debate about abortion rights. The other is
the famous debate between Abraham Lincoln and Stephen Douglas over popular
sovereignty and slavery.
[*1778] Given the intense disagreement over the moral permissibility of abortion, the
case for seeking a political solution that brackets the contending moral and
religious issues -- that is neutral with respect to them -- would seem
especially strong. But whether it is reasonable to bracket, for political
purposes, the comprehensive moral and religious doctrines at stake largely
depends on which of those doctrines is true. If the doctrine of the Catholic
Church is true, if human life in the relevant moral sense does begin at
conception, then bracketing the moral-theological question when human life
begins is far less reasonable than it would be on rival moral and religious
assumptions. The more confident we are that fetuses are, in the relevant moral
sense, different from babies, the more confident we can be in affirming a
political conception of justice that sets aside the controversy about the moral
status of fetuses.
The political liberal might reply that the political values of toleration and
equal citizenship for women are sufficient grounds for concluding that women
should be free to choose for themselves whether to have an abortion; government
should not take sides on the moral and religious controversy over when human
n31 But if the Catholic Church is right about the moral status of the fetus, if
abortion is morally tantamount to murder, then it is not clear why the
political values of toleration and women's equality, important though they are,
should prevail. If the Catholic doctrine is true, the political liberal's case
for the priority of political values must become an instance of just-war
theory; he or she would have to show why these values should prevail even at
the cost of some 1.5 million civilian deaths each year.
Of course, to suggest the impossibility of bracketing the moral-theological
question of when human life begins is not to argue against a right to abortion.
It is simply to show that the case for abortion rights cannot be neutral with
respect to that moral and religious controversy. It must engage rather than
avoid the comprehensive moral and religious doctrines at stake. Liberals often
resist this engagement because it violates the priority of the right over the
good. But the abortion debate shows that this priority cannot be sustained.
The case for respecting a woman's right to decide for herself whether to have
an abortion depends on showing -- as I believe can be shown -- that there is a
relevant moral difference between aborting a fetus at a relatively early stage
of development and killing a child.
A second illustration of the difficulty with a political conception of justice
that tries to bracket controversial moral questions is offered by the 1858
debates between Abraham Lincoln and Stephen Douglas.
[*1779] Douglas's argument for the doctrine of popular sovereignty is perhaps the most
famous case in American history for bracketing a controversial moral question
for the sake of political agreement. Because people were bound to disagree
about the morality of slavery, Douglas argued, national policy should be
neutral on that question. The doctrine of popular sovereignty he defended did
not judge slavery right or wrong, but left the people of each territory free to
make their own judgments.
"[T]o throw the weight of federal power into the scale, either in favor of the
free or the slave states," would violate the fundamental principles of the Constitution and run the risk
of civil war. The only hope of holding the country together, he argued, was to
agree to disagree, to bracket the moral controversy over slavery and respect
"the right of each state and each territory to decide these questions for
Lincoln argued against Douglas's case for a political conception of justice.
Policy should express rather than avoid a substantive moral judgment about
slavery. Though Lincoln was not an abolitionist, he believed government should
treat slavery as the moral wrong that it was, and prohibit its extension to the
"The real issue in this controversy -- the one pressing upon every mind -- is
the sentiment on the part of one class that looks upon the institution of
as a wrong, and of another class that
does not look upon it as a wrong."
n33 Lincoln and the Republican party viewed slavery as a wrong and insisted that
"be treated as a wrong, and one of the methods of treating it as a wrong is to
make provision that it shall grow no larger."
Whatever his personal moral views, Douglas claimed that, for political purposes
at least, he was agnostic on the question of slavery; he did not care whether
"voted up or voted down."
n35 Lincoln replied that it was reasonable to bracket the question of the morality
of slavery only on the assumption that it was not the moral evil he regarded it
to be. Any man can advocate political neutrality
who does not see anything wrong in slavery, but no man can logically say it who
does see a wrong in it; because no man can logically say he don't care whether
a wrong is voted up or voted down. He may say he don't care whether an
indifferent thing is voted up or down, but he must logically have a choice
between a right thing and a wrong thing. He contends that whatever community
wants slaves has a right to have them. So they have if it is not a wrong. But
if it is a wrong, he cannot say people have a right to do wrong.
[*1780] The debate between Lincoln and Douglas was not primarily about the morality of
slavery, but about whether to bracket a moral controversy for the sake of
political agreement. In this respect, their debate over popular sovereignty is
analogous to the contemporary debate over abortion rights. As some
contemporary liberals argue that government should not take a stand one way or
the other on the morality of abortion, but let each woman decide the question
for herself, so Douglas argued that national policy should not take a stand one
way or the other on the morality of slavery, but let each territory decide the
question for itself. There is of course the difference that in the case of
abortion rights, those who would bracket the substantive moral question
typically leave the choice to the individual, while in the case of slavery,
Douglas's way of bracketing was to leave the choice to the territories.
But Lincoln's argument against Douglas was an argument against bracketing as
such, at least where grave moral questions are at stake. Lincoln's point was
that the political conception of justice defended by Douglas depended for its
plausibility on a particular answer to the substantive moral question it
claimed to bracket. This point applies with equal force to those arguments for
abortion rights that claim to take no side in the controversy over the moral
status of the fetus. Even in the face of so dire a threat to social
cooperation as the prospect of civil war, Lincoln argued that it made neither
moral nor political sense to bracket the most divisive moral controversy of the
I say, where is the philosophy or the statesmanship based on the assumption
that we are to quit talking about it . . . and that the public mind is all at
once to cease being agitated by it? Yet this is the policy . . . that Douglas
is advocating -- that we are to care nothing about it! I ask you if it is not
a false philosophy? Is it not a false statesmanship that undertakes to build
up a system of policy upon the basis of caring nothing about
the very thing that every body does care the most about?
Present-day liberals will surely resist the company of Douglas and want
national policy to oppose slavery, presumably on the grounds that slavery
violates people's rights. The question is whether liberalism conceived as a
political conception of justice can make this claim consistent with its own
strictures against appeals to comprehensive moral ideals. For example, a
Kantian liberal can oppose slavery as a failure to treat persons as ends in
themselves, worthy of respect. But this argument, resting as it does on a
Kantian conception of the person, is unavailable to
political liberalism. Other historically important arguments against slavery are unavailable to
[*1781] for similar reasons. American abolitionists of the 1830s and 1840s, for
example, typically cast their arguments in religious terms, arguments that
political liberalism cannot invoke.
How, then, can
political liberalism escape the company of Douglas and oppose slavery without presupposing some
comprehensive moral view? It might be replied that Douglas was wrong to seek
social peace at any price; not just any political agreement will do. Even
conceived as a political conception, justice as fairness is not merely a modus
vivendi. Given the principles and self-understandings implicit in our
political culture, only an agreement on terms that treat persons fairly, as
free and equal citizens, can provide a reasonable basis for social cooperation.
For us twentieth-century Americans, at least, the rejection of slavery is a
settled matter. The historical demise of Douglas's position is by now a fact
of our political tradition that any political agreement must take as given.
This appeal to the conception of citizenship implicit in our political culture
might explain how
political liberalism can oppose slavery today; our present political culture was importantly
shaped, after all, by the Civil War, Reconstruction, the adoption of the
Thirteenth, Fourteenth, and Fifteenth Amendments,
Brown v. Board of Education,
n38 the civil rights movement, the Voting Rights Act,
n39 and so on. These experiences, and the shared understanding of racial equality
and equal citizenship they formed, provide ample grounds for holding that
slavery is at odds with American political and constitutional practice as it
has developed over the past century.
But this does not explain how
political liberalism could oppose slavery in 1858. The notions of equal citizenship implicit in
American political culture of the mid-nineteenth century were arguably
hospitable to the institution of slavery. The Declaration of Independence
proclaimed that all men are created equal, endowed by their Creator with
certain unalienable rights, but Douglas argued, not implausibly, that the
signers of the Declaration were asserting the right of the colonists to be free
of British rule, not the right of their Negro slaves to equal citizenship.
n40 The Constitution itself did not prohibit slavery, but to the contrary
accommodated it by allowing states to count three-fifths of their slave
population for apportionment purposes,
n41 providing that Congress could not prohibit the slave trade until 1808,
n42 and requiring the return of fugitive slaves.
n43 And in the notorious
[*1782] Scott case,
n44 the Supreme Court upheld the property rights of slaveholders in their slaves
and ruled that African-Americans were not citizens of the United States.
n45 To the extent that
political liberalism refuses to invoke comprehensive moral ideals and relies instead on notions of
citizenship implicit in the political culture, it would have a hard time
explaining, in 1858, why Lincoln was right and Douglas was wrong.
B. The Fact of Reasonable Pluralism
The abortion debate today and the Lincoln-Douglas debate of 1858 illustrate the
way a political conception of justice must presuppose some answer to the moral
questions it purports to bracket, at least where grave moral questions are
concerned. In cases such as these, the priority of the right over the good
cannot be sustained. A further difficulty with
political liberalism concerns the reason it gives for asserting the priority of the right over the
good in the first place. For Kantian liberalism, the asymmetry between the
right and the good arises from a certain conception of the person. Because we
must think of ourselves as moral subjects given prior to our aims and
attachments, we must regard the right as regulative with respect to the
particular ends we affirm; the right is prior to the good because the self is
prior to its ends.
political liberalism, the asymmetry between the right and the good is not based on a Kantian
conception of the person but instead on a certain feature of modern democratic
Rawls describes this feature as
"the fact of reasonable pluralism" (p. xvii).
"A modern democratic society is characterized not simply by a pluralism of
comprehensive religious, philosophical, and moral doctrines but by a pluralism
of incompatible yet reasonable comprehensive doctrines. No one of these
doctrines is affirmed by citizens generally" (p. xvi). Nor is it likely that sometime in the foreseeable future this
pluralism will cease to hold. Disagreement about moral and religious questions
is not a temporary condition but
"the normal result of the exercise of human reason" under free institutions (p. xvi).
"fact of reasonable pluralism," the problem is to find principles of justice that free and equal citizens can
affirm despite their moral, philosophical, and religious differences.
"This is a problem of political justice, not a problem about the highest good" (p. xxv). Whatever principles it generates, the solution to this problem must
be one that upholds the priority of the right over the good. Otherwise, it
will fail to provide a basis for social cooperation among adherents of
incompatible but reasonable moral and religious convictions.
[*1783] But here there arises a difficulty. For even if the fact of reasonable
pluralism is true, the asymmetry between the right and the good depends on a
further assumption. This is the assumption that, despite our disagreements
about morality and religion, we do not have, or on due reflection would not
have, similar disagreements about justice.
Political liberalism must assume not only that the exercise of human reason under conditions of
freedom will produce disagreements about the good life, but also that the
exercise of human reason under conditions of freedom will
not produce disagreements about justice. The
"fact of reasonable pluralism" about morality and religion only creates an asymmetry between the right and
the good when coupled with the further assumption that there is no comparable
"fact of reasonable pluralism" about justice.
It is not clear, however, that this further assumption is justified. We need
only look around us to see that modern democratic societies are teeming with
disagreements about justice. Consider, for example, contemporary debates about
affirmative action, income distribution and tax fairness, health care,
immigration, gay rights, free speech versus hate speech, and capital
punishment, to name just a few. Or consider the divided votes and conflicting
opinions of Supreme Court Justices in cases involving religious liberty,
freedom of speech, privacy rights, voting rights, the rights of the accused,
and so on. Do not these debates display a
"fact of reasonable pluralism" about justice? If so, how does the pluralism about justice that prevails in
modern democratic societies differ from the pluralism about morality and
religion? Is there reason to think that, sometime in the foreseeable future,
our disagreements about justice will dissolve even as our disagreements about
morality and religion persist?
The political liberal might reply by distinguishing two different kinds of
disagreement about justice. There are disagreements about what the principles
of justice should be and disagreements about how these principles should be
applied. Many of our disagreements about justice, it might be argued, are of
the second kind. Although we generally agree, for example, that freedom of
speech is among the basic rights and liberties, we disagree about whether the
right to free speech should protect racial epithets, or violent pornographic
depictions, or commercial advertising, or unlimited contributions to political
campaigns. These disagreements, vigorous and even intractable though they may
be, are consistent with our agreeing at the level of principle that a just
society includes a basic right to free speech.
Our disagreements about morality and religion, by contrast, might be seen as
more fundamental. They reflect incompatible conceptions of the good life, it
might be argued, not disagreements about how to put into practice a conception
of the good life that commands, or on reflection would command, widespread
agreement. If our controversies about justice concern the application of
principles we share or
[*1784] would share on due reflection, while our controversies about morality and
religion run deeper, then the asymmetry between the right and the good advanced
political liberalism would be vindicated.
But with what confidence can this contrast be asserted? Do all of our
disagreements about justice concern the application of principles we share or
would share on due reflection, rather than the principles themselves? What of
our debates about distributive justice? Here it would seem that our
disagreements are at the level of principle, not application. Some maintain,
Rawls's difference principle, that only those social and economic inequalities that
improve the condition of the least-advantaged members of society are just.
They argue, for example, that government must ensure the provision of certain
basic needs, such as income, education, health care, housing, and the like, so
that all citizens will be able meaningfully to exercise their basic liberties.
Others reject the difference principle. Libertarians argue, for example, that
it may be a good thing for people to help those less fortunate than themselves,
but that this should be a matter of charity, not entitlement. Government
should not use its coercive power to redistribute income and wealth, but should
respect people's rights to exercise their talents as they choose, and to reap
their rewards as defined by the market economy.
The debate between liberal egalitarians like
Rawls and libertarians like Robert Nozick and Milton Friedman is a prominent feature
of political argument in modern democratic societies. This debate reflects
disagreement about what the correct principle of distributive justice is, not
disagreement about how to apply the difference principle. But this would
suggest that there exists in democratic societies a
"fact of reasonable pluralism" about justice as well as about morality and religion. And if this is the
case, the asymmetry between the right and the good does not hold.
Political liberalism is not without a reply to this objection, but the reply it must make departs
from the spirit of toleration it otherwise evokes.
Rawls's reply must be that, although there is a fact of pluralism about distributive
justice, there is no fact of
n47 Unlike disagreements about morality and religion, disagreements about the
validity of the difference principle are not reasonable; libertarian theories
of distributive justice would not be sustained on
[*1785] due reflection. Our differences about distributive justice, unlike our
differences of morality and religion, are not the natural outcome of the
exercise of human reason under conditions of freedom.
At first glance, the claim that disagreements about distributive justice are
not reasonable may seem arbitrary, even harsh, at odds with
political liberalism's promise to apply
"the principle of toleration to philosophy itself" (p. 10). It contrasts sharply with
Rawls's apparent generosity toward differences of morality and religion. These
Rawls repeatedly writes, are a normal, indeed desirable feature of modern life, an
expression of human diversity that only the oppressive use of state power can
overcome (pp. 303-04). Where comprehensive moralities are concerned,
"it is not to be expected that conscientious persons with full powers of reason,
even after free discussion, will all arrive at the same conclusion" (p. 58). Since the exercise of human reason produces a plurality of
reasonable moral and religious doctrines,
"it is unreasonable or worse to want to use the sanctions of state power to
correct, or to punish, those who disagree with us" (p. 138). But this spirit of toleration does not extend to our disagreements
about justice. Because disagreements between, say, libertarians and advocates
of the difference principle do not reflect a reasonable pluralism, there is no
objection to using state power to implement the difference principle.
Intolerant though it may seem at first glance, the notion that theories of
distributive justice at odds with the difference principle are not reasonable,
or that libertarian theories of justice would not survive due reflection, is no
arbitrary claim. To the contrary, in
A Theory of Justice
Rawls offers a rich array of compelling arguments on behalf of the difference
principle and against libertarian conceptions: the distribution of talents and
assets that enables some to earn more and others less in the market economy is
arbitrary from a moral point of view; so is the fact that the market happens to
prize and reward, at any given moment, the talents you or I may have in
abundance; libertarians would agree that distributive shares should not be
based on social status or accident of birth (as in aristocratic or caste
societies), but the distribution of talents given by nature is no less
arbitrary; the notion of freedom that libertarians invoke can be meaningfully
exercised only if persons' basic social and economic needs are met; if people
deliberated about distributive justice without reference to their own
interests, or without prior knowledge of their talents and the value of those
talents in the market economy, they would agree that the natural distribution
of talents should not be the basis of distributive shares; and so on.
My point is not to rehearse
Rawls's argument for the difference principle, but only to recall the kind of reasons
he offers. Viewing
[*1786] justification as a process of mutual adjustment between principles and
considered judgments that aims at a
Rawls tries to show that the difference principle is more reasonable than the
alternative offered by libertarians. To the extent that his arguments are
convincing -- as I believe they are -- and to the extent they can be convincing
to citizens of a democratic society, the principles they support are properly
embodied in public policy and law. Disagreement will doubtless remain.
Libertarians will not fall silent or disappear. But their disagreement need
not be regarded as a
"fact of reasonable pluralism" in the face of which government must be neutral.
But this leads to a question that goes to the heart of
political liberalism's claim for the priority of the right over the good: if moral argument or
reflection of the kind
Rawls deploys enables us to conclude, despite the persistence of conflicting views,
that some principles of justice are more reasonable than others, what
guarantees that reflection of a similar kind is not possible in the case of
moral and religious controversy? If we can reason about controversial
principles of distributive justice by seeking a reflective equilibrium, why can
we not reason in the same way about conceptions of the good? If it can be
shown that some conceptions of the good are more reasonable than others, then
the persistence of disagreement would not necessarily amount to a
"fact of reasonable pluralism" that requires government to be neutral.
Consider, for example, the controversy in our public culture about the moral
status of homosexuality, a controversy based in comprehensive moral and
religious doctrines. Some maintain that homosexuality is sinful, or at least
morally impermissible; others argue that homosexuality is morally permissible,
and in some cases gives expression to important human goods.
Political liberalism insists that neither of these views about the morality of homosexuality should
play a role in public debates about justice or rights. Government must be
neutral with respect to them. This means that those who abhor homosexuality
may not seek to embody their view in law; it also means that proponents of gay
rights may not base their arguments on the notion that homosexuality is morally
defensible. From the standpoint of
political liberalism, each of these approaches would wrongly base the right on some conception of
the good; each would fail to respect the
"fact of reasonable pluralism" about comprehensive moralities.
But does the disagreement in our society about the moral status of
homosexuality constitute a
"fact of reasonable pluralism" any more than does the disagreement about distributive justice? According to
political liberalism, the libertarian's objection to the difference principle
[*1787] does not constitute a
"fact of reasonable pluralism" that requires government neutrality, because there are good reasons to
conclude, on due reflection, that the arguments for the difference principle
are more convincing than the ones that support libertarianism. But isn't it
possible to conclude, with equal or greater confidence, that on due reflection,
the arguments for the moral permissibility of homosexuality are more convincing
than the arguments against it? Consistent with the search for a reflective
equilibrium among principles and considered judgments, such reflection might
proceed by assessing the reasons advanced by those who assert the moral
inferiority of homosexual to heterosexual relations.
Those who consider homosexuality immoral often argue, for example, that
homosexuality cannot fulfill the highest end of human sexuality, the good of
n50 To this it might be replied that many heterosexual relations also do not
fulfill this end, such as contracepted sex, or sex among sterile couples, or
sex among partners beyond the age of reproduction. This might suggest that the
good of procreation, important though it is, is not necessary to the moral
worth of human sexual relations; the moral worth of sexuality might also
consist in the love and responsibility it expresses, and these goods are
possible in homosexual as well as heterosexual relations. Opponents might
reply that homosexuals are often promiscuous, and hence less likely to realize
the goods of love and responsibility. The reply to this claim might consist in
an empirical showing to the contrary, or in the observation that the existence
of promiscuity does not argue against the moral worth of homosexuality as such,
only against certain instances of it.
n51 Heterosexuals also engage in promiscuity and other practices at odds with the
goods that confer on sexuality its moral worth, but this fact does not lead us
to abhor heterosexuality as such. And so on.
My point is not to offer a full argument for the moral permissibility of
homosexuality, only to suggest the way such an argument might proceed. Like
Rawls's argument for the difference principle, it would proceed by seeking a
reflective equilibrium between our principles and considered judgments,
adjusting each in the light of the other. That the argument for the morality
of homosexuality, unlike the argument
[*1788] for the difference principle, explicitly addresses claims about human ends and
conceptions of the good does not mean that the same method of moral reasoning
cannot proceed. It is unlikely, of course, that such moral reasoning would
produce conclusive or irrefutable answers to moral and religious controversies.
Rawls acknowledges, such reasoning does not produce irrefutable answers to questions
of justice either; a more modest notion of justification is appropriate.
"[I]n philosophy questions at the most fundamental level are not usually settled
by conclusive argument," writes
Rawls, referring to arguments about justice.
"What is obvious to some persons and accepted as a basic idea is unintelligible
to others. The way to resolve the matter is to consider after due reflection
which view, when fully worked out, offers the most coherent and convincing
account" (p. 53). The same could be said of arguments about comprehensive moralities.
If it is possible to reason about the good as well as the right, then
political liberalism's claim for the asymmetry between the right and good is undermined. For
political liberalism, this asymmetry rests on the assumption that our moral and religious
disagreements reflect a
"fact of reasonable pluralism" that our disagreements about justice do not. What enables
Rawls to maintain that our disagreements about distributive justice do not amount to
"fact of reasonable pluralism" is the strength of the arguments he advances on behalf of the difference
principle and against libertarianism. But the same could be said of other
controversies, including, conceivably, some moral and religious controversies.
The public culture of democratic societies includes controversies about justice
and comprehensive moralities alike. If government can affirm the justice of
redistributive policies even in the face of disagreement by libertarians, why
cannot government affirm in law, say, the moral legitimacy of homosexuality,
even in the face of disagreement by those who regard homosexuality as sin?
n52 Is Milton Friedman's objection to redistributive policies a less
"reasonable pluralism" than Pat Robertson's objection to gay rights?
With morality as with justice, the mere fact of disagreement is no evidence of
"reasonable pluralism" that gives rise to the demand that government must be neutral. There is no
reason in principle why in any given case, we might not conclude that, on due
reflection, some moral or religious doctrines are more plausible than others.
In such cases, we would not expect all disagreement to disappear, nor would we
rule out the possibility that further deliberation might one day lead us to
reverse our view. But neither would we have grounds
[*1789] to insist that our deliberations about justice and rights may make no
reference to moral or religious ideals.
C. The Limits of Liberal Public Reason
Whether it is possible to reason our way to agreement on any given moral or
political controversy is not something we can know until we try. This is why
it cannot be said in advance that controversies about comprehensive moralities
"fact of reasonable pluralism" that controversies about justice do not. Whether a moral or political
controversy reflects reasonable but incompatible conceptions of the good, or
whether it can be resolved by due reflection and deliberation, can only be
determined by reflecting and deliberating. But this raises a further
political liberalism. For the political life it describes leaves little room for the kind of public
deliberation necessary to test the plausibility of contending comprehensive
moralities -- to persuade others of the merits of our moral ideals, to be
persuaded by others of the merits of theirs.
political liberalism upholds the right to freedom of speech, it severely limits the kinds of
arguments that are legitimate contributions to political debate, especially
debate about constitutional essentials and basic justice.
n53 This limitation reflects the priority of the right over the good. Not only
may government not endorse one or another conception of the good, but citizens
may not even introduce into political discourse their comprehensive moral or
religious convictions, at least when debating matters of justice and rights
Rawls maintains that this limitation is required by the
"ideal of public reason" (p. 218). According to this ideal, political discourse should be conducted
solely in terms of
"political values" that all citizens can reasonably be expected to accept. Because citizens of
democratic societies do not share comprehensive moral and religious
conceptions, public reason should not refer to such conceptions (pp. 216-20).
The limits of public reason do not apply,
Rawls allows, to our personal deliberations about political questions, or to the
discussions we may have as members of associations such as churches and
"religious, philosophical, and moral considerations" (p. 215) may properly play a role.
But the ideal of public reason does hold for citizens when they engage in
political advocacy in the public forum, and thus for members of
[*1790] political parties and for candidates in their campaigns and for other groups
who support them. It holds equally for how citizens are to vote in elections
when constitutional essentials and matters of basic justice are at stake.
Thus, the ideal of public reason not only governs the public disclosure of
elections insofar as the issues involve those fundamental questions, but also
how citizens are to cast their vote on these questions (p. 215).
How can we know whether our political arguments meet the requirements of public
reason, suitably shorn of any reliance on moral or religious convictions?
Rawls offers a novel test.
"To check whether we are following public reason we might ask: how would our
argument strike us presented in the form of a supreme court opinion?" (p. 254). For citizens of a democracy to allow their political discourse
about fundamental questions to be informed by moral and religious ideals is no
Rawls suggests, than for a judge to read his or her moral and religious beliefs into
The restrictive character of this notion of public reason can be seen by
considering the sorts of political arguments it would rule out. In the debate
about abortion rights, those who believe that the fetus is a person from the
moment of conception and that abortion is therefore murder could not seek to
persuade their fellow citizens of this view in open political debate. Nor
could they vote for a law that would restrict abortion on the basis of this
moral or religious conviction. Although adherents of the Catholic teaching on
abortion could discuss the issue of abortion rights in religious terms within
their church, they could not do so in a political campaign, or on the floor of
the state legislature, or in the halls of Congress. Nor for that matter could
opponents of the Catholic teaching on abortion argue their case in the
political arena. Relevant though it clearly is to the question of abortion
rights, Catholic moral doctrine cannot be debated in the political arena that
political liberalism defines.
The restrictive character of liberal public reason can also be seen in the
debate about gay rights. At first glance, these restrictions might seem a
service to toleration. Those who consider homosexuality immoral and therefore
unworthy of the privacy rights accorded heterosexual intimacy could not
legitimately voice their views in public debate. Nor could they act on their
belief by voting against laws that would protect gay men and lesbians from
discrimination. These beliefs reflect comprehensive moral and religious
convictions and so may not play a part in political discourse about matters of
But the demands of public reason also limit the arguments that can be advanced
in support of gay rights, and so restrict the range of reasons that can be
invoked on behalf of toleration. Those who oppose anti-sodomy laws of the kind
at issue in
Bowers v. Hardwick
[*1791] cannot argue that the moral judgments embodied in those laws are wrong, only
that the law is wrong to embody any moral judgments at all.
n56 Advocates of gay rights cannot contest the substantive moral judgment lying
behind anti-sodomy laws or seek, through open political debate, to persuade
their fellow citizens that homosexuality is morally permissible, for any such
argument would violate the canons of liberal public reason.
The restrictive character of liberal public reason is also illustrated by the
arguments offered by American abolitionists of the 1830s and 1840s. Rooted in
evangelical Protestantism, the abolitionist movement argued for the immediate
emancipation of the slaves on the grounds that slavery is a heinous sin.
n57 Like the argument of some present-day Catholics against abortion rights, the
abolitionist case against slavery was explicitly based on a comprehensive moral
and religious doctrine.
In a puzzling passage,
Rawls tries to argue that the abolitionist case against slavery, religious though it
was, did not violate the ideal of liberal public reason. If a society is not
well-ordered, he explains, it may be necessary to resort to comprehensive
moralities in order to bring about a society in which public discussion is
conducted solely in terms of
"political values" (p. 251 n.41). The religious arguments of the abolitionists can be justified
as hastening the day when religious arguments would no longer play a legitimate
role in public discourse. The abolitionists
"did not go against the ideal of public reason,"
"provided they thought, or on reflection would have thought (as they certainly
could have thought), that the comprehensive reasons they appealed to were
required to give sufficient strength to the political conception to be
subsequently realized" (p. 251).
It is difficult to know what to make of this argument. There is little reason
to suppose, and I do not think
Rawls means to suggest, that the abolitionists opposed slavery on secular political
grounds and simply used religious arguments to win popular support. Nor is
there reason to think that the abolitionists sought by their agitation to make
a world safe for secular political discourse. Nor can it be assumed that, even
in retrospect, the abolitionists would take pride in having contributed, by
their religious arguments against slavery, to the emergence of a society
inhospitable to religious argument in political debate. If anything the
opposite is more likely the case, that by advancing religious arguments against
so conspicuous an injustice as slavery, the evangelicals who inspired the
abolitionist movement were
[*1792] hoping to encourage Americans to view other political questions in moral and
religious terms as well. In any case, it is reasonable to suppose that the
abolitionists meant what they said, that slavery is wrong because it is
contrary to God's law, a heinous sin, and that this is the reason it should be
ended. Absent some extraordinary assumptions, it is difficult to interpret
their argument as consistent with the priority of the right over the good, or
with the ideal of public reason advanced by
The cases of abortion, gay rights, and abolitionism illustrate the severe
restrictions liberal public reason would impose on political debate.
Rawls argues that these restrictions are justified as essential to the maintenance
of a just society, in which citizens are governed by principles they may
reasonably be expected to endorse, even in the light of their conflicting
comprehensive moralities. Although public reason requires that citizens decide
fundamental political questions without reference
"to the whole truth as they see it" (p. 216), this restriction is justified by the political values, such as
civility and mutual respect, that it makes possible.
"[T]he political values realized by a well-ordered constitutional regime are
very great values and not easily overridden and the ideals they express are not
to be lightly abandoned" (p. 218).
Rawls compares his case for restrictive public reason with the case for restrictive
rules of evidence in criminal trials. There too we agree to decide without
reference to the whole truth as we know it -- through illegally obtained
evidence, for example -- in order to advance other goods (pp. 218-19).
The analogy between liberal public reason and restrictive rules of evidence is
instructive. Setting aside the whole truth as we know it carries moral and
political costs, for criminal trials and for public reason alike. Whether
those costs are worth incurring depends on how significant they are compared to
the goods they make possible, and whether those goods can be secured in some
other way. To assess restrictive rules of evidence, for example, we need to
know how many criminals go free as a result and whether less restrictive rules
would unduly burden innocent persons suspected of a crime, lead to undesirable
law enforcement practices, violate important ideals such as respect for privacy
(exclusionary rule) and spousal intimacy (spousal privilege), and so on. We
arrive at rules of evidence by weighing the importance of deciding in the light
of the whole truth against the importance of the ideals that would be
sacrificed if all evidence were admissible.
Similarly, to assess restrictive rules of public reason, we need to weigh their
moral and political costs against the political values they are said to make
possible; we must also ask whether these political values -- of toleration,
civility, and mutual respect -- could be achieved under less-restrictive rules
of public reason. Although
political liberalism refuses to weigh the political values it affirms against
[*1793] competing values that may arise from within comprehensive moralities, the case
for restrictive rules of public reason must presuppose some such comparison.
The costs of liberal public reason are of two kinds. The strictly moral costs
depend on the validity and importance of the moral and religious doctrines
liberal public reason requires us to set aside when deciding questions of
justice. These costs will necessarily vary from case to case. They will be at
their highest when a political conception of justice sanctions toleration of a
grave moral wrong, such as slavery in the case of Douglas's argument for
popular sovereignty. In the case of abortion, the moral cost of bracketing is
high if the Catholic doctrine is correct, otherwise much lower. This suggests
that, even given the moral and political importance of toleration, the argument
for tolerating a given practice must take some account of the moral status of
the practice, as well as the good of avoiding social conflict, letting people
decide for themselves, and so on.
This way of thinking about the moral cost of liberal public reason is
admittedly at odds with
political liberalism itself. Although
Rawls repeatedly states that a political conception of justice expresses values that
normally outweigh whatever other values conflict with them (pp. 138, 146, 156,
218), he also insists that this involves no substantive comparison of the
political values to the moral and religious values they override.
We need not consider the claims of political justice against the claims of this
or that comprehensive view; nor need we say that political values are
intrinsically more important than other values and that is why the latter are
overridden. Having to say that is just what we hope to avoid . . . (p. 157).
political liberalism allows that comprehensive moral and religious doctrines can be true, such
comparisons cannot reasonably be avoided.
Beyond the moral costs of liberal public reason are certain political costs.
These costs are becoming increasingly apparent in the politics of those
countries, notably the United States, whose public discourse most closely
approximates the ideal of public reason advanced by
political liberalism. With a few notable exceptions, such as the civil rights movement, American
political discourse in recent decades has come to reflect the liberal resolve
that government be neutral on moral and religious questions, that fundamental
questions of public policy be debated and decided without reference to any
particular conception of the good.
n58 But democratic politics cannot long abide a public life as abstract and
decorous, as detached from moral purposes, as Supreme
[*1794] Court opinions are supposed to be. A politics that brackets morality and
religion too completely soon generates its own disenchantment. Where political
discourse lacks moral resonance, the yearning for a public life of larger
meanings finds undesirable expressions. Groups like the Moral Majority seek to
clothe the naked public square with narrow, intolerant moralisms.
Fundamentalists rush in where liberals fear to tread. The disenchantment also
assumes more secular forms. Absent a political agenda that addresses the moral
dimension of public questions, public attention becomes riveted on the private
vices of public officials. Public discourse becomes increasingly preoccupied
with the scandalous, the sensational, and the confessional as purveyed by
tabloids, talk shows, and eventually the mainstream media as well.
It cannot be said that the public philosophy of
political liberalism is wholly responsible for these tendencies. But its vision of public reason
is too spare to contain the moral energies of a vital democratic life. It thus
creates a moral void that opens the way for the intolerant and the trivial and
other misguided moralisms.
If liberal public reason is too restrictive, it remains to ask whether a more
spacious public reason would sacrifice the ideals that
political liberalism seeks to promote, notably mutual respect among citizens who hold conflicting
moral and religious views. Here it is necessary to distinguish two conceptions
of mutual respect. On the liberal conception, we respect our fellow citizens'
moral and religious convictions by ignoring them (for political purposes), by
leaving them undisturbed, or by carrying on political debate without reference
to them. To admit moral and religious ideals into political debate about
justice would undermine mutual respect in this sense.
But this is not the only, or perhaps even the most plausible way of
understanding the mutual respect on which democratic citizenship depends. On a
different conception of respect -- call it the deliberative conception -- we
respect our fellow citizen's moral and religious convictions by engaging, or
attending to them -- sometimes by challenging and contesting them, sometimes by
listening and learning from them -- especially if those convictions bear on
important political questions. There is no guarantee that a deliberative mode
of respect will lead in any given case to agreement or even to appreciation for
the moral and religious convictions of others. It is always possible that
learning more about a moral or religious doctrine will lead us to like it less.
But the respect of deliberation and engagement affords a more spacious public
reason than liberalism allows. It is also a more suitable ideal for a
pluralist society. To the extent that our moral and religious disagreements
reflect the ultimate plurality of human goods, a deliberative mode of respect
will better enable us to appreciate the distinctive goods our different lives
n1 James Bryant Conant University Professor, Emeritus, Harvard University.
n2 Professor of Government, Harvard University. The author is grateful to Yochai
Joshua Cohen, Stephen Macedo, and J. Russell Muirhead for helpful comments and criticisms.
RAWLS, A THEORY OF JUSTICE (1971).
See, e.g., H.L.A. Hart,
Between Utility and Rights, in THE IDEA OF FREEDOM 77, 77 (Alan Ryan ed., 1979).
See FRIEDRICH A. HAYEK, THE CONSTITUTION OF LIBERTY (1960); ROBERT NOZICK,
ANARCHY, STATE, AND UTOPIA (1974).
See BRUCE A. ACKERMAN, SOCIAL JUSTICE IN THE LIBERAL STATE 349-78 (1980); RONALD
DWORKIN, TAKING RIGHTS SERIOUSLY 90-100, 168-77 (1977); CHARLES FRIED, RIGHT
AND WRONG 114-19 (1978); CHARLES E. LARMORE, PATTERNS OF MORAL COMPLEXITY 42-68
supra note 5, at 33;
supra note 3, at 30-32, 446-51, 560; Ronald Dworkin,
Liberalism, in PUBLIC AND PRIVATE MORALITY 113, 127-36 (Stuart Hampshire ed., 1978); Thomas
Moral Conflict and Political Legitimacy, 16 PHIL.
& PUB. AFF. 215, 227-37 (1987).
See IMMANUEL KANT, CRITIQUE OF PURE REASON (Norman K. Smith trans., St. Martin's
Press 1965) (1788); IMMANUEL KANT, GROUNDWORK OF THE METAPHYSIC OF MORALS (H.
J. Paton trans., Harper
& Row 3d ed. 1964) (1785); IMMANUEL KANT,
On the Common Saying:
"This May Be True in Theory, but It Does Not Apply in Practice," in KANT'S POLITICAL WRITINGS 61, 73-74 (Hans Reiss ed.
& H. B. Nisbet trans., 1970);
supra note 3, at 30-32, 446-51, 560.
See ALASDAIR MACINTYRE, AFTER VIRTUE (2d ed. 1984) [hereinafter MACINTYRE, AFTER
VIRTUE]; ALASDAIR MACINTYRE, IS PATRIOTISM A VIRTUE?: THE LINDLEY LECTURE
(1984) [hereinafter MACINTYRE, IS PATRIOTISM A VIRTUE?]; ALASDAIR MACINTYRE,
WHOSE JUSTICE? WHICH RATIONALITY? (1988).
See CHARLES TAYLOR,
The Nature and Scope of Distributive Justice, in PHILOSOPHY AND THE HUMAN SCIENCES, 2 PHILOSOPHICAL PAPERS 289 (1985); CHARLES
TAYLOR, SOURCES OF THE SELF: THE MAKING OF THE MODERN IDENTITY (1989)
[hereinafter TAYLOR, SOURCES OF THE SELF].
See MICHAEL WALZER, SPHERES OF JUSTICE: A DEFENSE OF PLURALISM AND EQUALITY (1983).
See MICHAEL J. SANDEL, LIBERALISM AND THE LIMITS OF JUSTICE (1982); Michael J.
The Procedural Republic and the Unencumbered Self, 12 POL. THEORY 81 (1984).
n12 Michael Walzer comes close to this view when he writes:
"Justice is relative to social meanings. . . . A given society is just if its
substantive life is lived . . . in a way faithful to the shared understandings
of the members." WALZER,
supra note 10, at 312-13. Walzer allows, however, that prevailing practices of
rights can be criticized from the standpoint of alternative interpretations of
a society's shared understandings.
See id. at 84-91.
n13 Much of the debate about liberal political philosophy in the last decade has
focused on the
"communitarian" critique of liberalism, or, more broadly, on the challenge to the priority of
the right over the good. The best overall account of this debate is STEPHEN
& ADAM SWIFT, LIBERALS AND COMMUNITARIANS (1992). Edited volumes on the subject
include COMMUNITARIANS AND INDIVIDUALISM (Shlomo Avineri
& Avner de-Shalit eds., 1992); LIBERALISM AND ITS CRITICS (Michael J. Sandel
ed., 1984); LIBERALISM AND THE GOOD (R. Bruce Douglass, Gerald M. Mara
& Henry S. Richardson eds., 1990); LIBERALISM AND THE MORAL LIFE (Nancy L.
Rosenblum ed., 1989); and UNIVERSALISM VS. COMMUNITARIANISM (David Rasmussen
ed., 1990). Notable book-length works include DANIEL BELL, COMMUNITARIANISM
AND ITS CRITICS (1993); WILL KYMLICKA, LIBERALISM, COMMUNITY AND CULTURE
(1989); CHARLES E. LARMORE, PATTERNS OF MORAL COMPLEXITY (1987); and STEPHEN
MACEDO, LIBERAL VIRTUES: CITIZENSHIP, VIRTUE, AND COMMUNITY IN LIBERAL
CONSTITUTIONALISM (1990). The vast literature on the subject includes, among
others: JEREMY WALDRON,
Particular Values and Critical Morality, in LIBERAL RIGHTS 168 (1993); C. Edwin Baker,
133 U. PA. L. REV. 895 (1985); Sheyla Benhabib,
Autonomy, Modernity and Community: Communitarianism and Critical Social Theory
in Dialogue, in ZWISCHENBETRACHTUNGEN IM PROZESS DER AUFKLAERUNG 373 (Axel Honneth, Thomas
McCarthy, Claus Offe
& Albrecht Welmer eds., 1989); Allen E. Buchanan,
Assessing the Communitarian Critique of Liberalism, 99 ETHICS 852 (1989); Gerald Doppelt,
Rawls's Kantian Liberalism Coherent and Defensible?, 99 ETHICS 815 (1989); Stephen A. Gardbaum,
Law, Politics, and the Claims of Community,
90 MICH. L. REV. 685 (1992); Emily R. Gill,
Goods, Virtues, and the Constitution of the Self, in LIBERALS ON LIBERALISM III (Alfonso J. Damico ed., 1986); Amy Gutmann,
Communitarian Critics of Liberalism, 14 PHIL.
& PUB. AFF. 308 (1985); H. N. Hirsch,
The Threnody of Liberalism, 14 POL. THEORY 423 (1986); Will Kymlicka,
Liberalism and Communitarianism, 18 CAN. J. PHIL. 181 (1988); Will Kymlicka,
Rawls on Teleology and Deontology, 17 PHIL.
& PUB. AFF. 173 (1988); Christopher Lasch,
The Communitarian Critique of Liberalism, 69 SOUNDINGS 60 (1986); David Miller,
In What Sense Must Socialism Be Communitarian?, 6 SOC. PHIL.
& POL. 57 (1989); Chantal Mouffe,
American Liberalism and Its Critics:
Rawls, Taylor, Sandel and Walzer, 8 PRAXIS INT'L 193 (1988); Patrick Neal,
A Liberal Theory of the Good?, 17 CAN. J. PHIL. 567 (1987); Jeffrey Paul
& Fred D. Miller Jr.,
Communitarian and Liberal Theories of the Good, 43 REV. METAPHYSICS 803 (1990); Milton C. Regan, Jr.,
Community and Justice in Constitutional Theory,
1985 WIS. L. REV. 1073; Richard Rorty,
The Priority of Democracy to Philosophy, in THE VIRGINIA STATUTE OF RELIGIOUS FREEDOM 257, 257-82 (Merrill D. Peterson
& Robert C. Vaughan eds., 1988); George Sher,
Three Grades of Social Involvement, 18 PHIL.
& PUB. AFF. 133 (1989); Tom Sorell,
Self, Society, and Kantian Impersonality, 74 MONIST 30 (1991); Symposium,
Law, Community, and Moral Reasoning,
77 CAL. L. REV. 475 (1989); Charles Taylor,
Cross-Purposes: The Liberal-Communitarian Debate, in LIBERALISM AND THE MORAL LIFE,
supra; Robert B. Thigpen
& Lyle A. Downing,
Liberalism and the Communitarian Critique, 31 AM. J. POL. SCI. 637 (1987); John Tomasi,
Individual Rights and Community Virtues, 101 ETHICS 521 (1991); John R. Wallach,
Liberals, Communitarians, and the Tasks of Political Theory, 15 POL. THEORY 581 (1987); Michael Walzer,
The Communitarian Critique of Liberalism, 18 POL. THEORY 6 (1990); Iris M. Young,
The Ideal of Community and the Politics of Difference, 12 SOC. THEORY
& PRAC. 1 (1986); and Joel Feinberg,
Liberalism, Community and Tradition, TIKKUN, May-June 1988, at 38. Prior to
Political Liberalism, Rawls addressed these issues in a number of essays, including
The Idea of an Overlapping Consensus, 7 OXFORD J. LEGAL STUD. 1 (1987);
Justice as Fairness: Political Not Metaphysical, 14 PHIL.
& PUB. AFF. 223 (1985); and
The Priority of Right and Ideas of the Good, 17 PHIL.
& PUB. AFF. 251 (1987). In
Political Liberalism, however, he states:
"The changes in the later essays are sometimes said to be replies to criticisms
raised by communitarians and others. I don't believe there is a basis for
saying this" (p. xvii).
supra note 3, at 560.
Id. at 561.
Id. at 574-75.
n17 The objection to the conception of the person presented in
A Theory of Justice does not depend on failing to see the original position as a device of
representation. It can be stated wholly in terms of the conception of the
person presented in Part III of
A Theory of Justice, which
Rawls now recasts as a political conception. Not only critics, but also defenders
Rawls's liberalism interpreted
A Theory of Justice as affirming a Kantian conception of the person.
See, e.g., LARMORE,
supra note 6, at 118-30.
See MACINTYRE, AFTER VIRTUE,
supra note 8, at 190-209; MACINTYRE, IS PATRIOTISM A VIRTUE?,
supra note 8,
supra note 11, at 175-83; TAYLOR, SOURCES OF THE SELF,
supra note 9, at 508.
n19 For contemporary examples of comprehensive liberalism, see GEORGE KATEB, THE
INNER OCEAN: INDIVIDUALISM AND DEMOCRATIC CULTURE (1992); and JOSEPH RAZ, THE
MORALITY OF FREEDOM (1986). Ronald Dworkin describes his view as a version of
comprehensive liberalism in
Foundations of Liberal Equality, in II THE TANNER LECTURES ON HUMAN VALUES 1 (Grethe B. Peterson ed., 1990).
supra note 3, at 11-12.
Id. at 560.
Id. at 561.
n24 The notion that we should regard our moral and religious duties as
"self-authenticating from a political point of view" (p. 33) accords with
Rawls's statement, in
A Theory of Justice, that
"from the standpoint of justice as fairness, these [moral and religious]
obligations are self-imposed."
supra note 3, at 206. But it is not clear what the justification can be on such a
view for according religious beliefs or claims of conscience a special respect
not accorded other preferences that people may hold with equal or greater
See id. at 205-11.
Kantian Constructivism in Moral Theory: Rational and Full Autonomy, 77 J. PHIL. 515, 519 (1980).
supra note 13, at 257, 262.
Id. at 265.
Id. at 268.
Id. at 264.
n30 Thomas Hobbes, who can be interpreted as advancing a political conception of
justice, ensured the priority of his political conception with respect to
claims arising from contending moral and religious conceptions by denying the
truth of those conceptions.
See THOMAS HOBBES, LEVIATHAN 168-83 (C.B. Macpherson ed., Penguin Books 1985)
Rawls seems to take this view in a footnote on abortion. But he does not explain
why political values should prevail even if the Catholic doctrine were true (p.
n32 CREATED EQUAL?: THE COMPLETE LINCOLN-DOUGLAS DEBATES OF 1858, at 369, 374
(Paul M. Angle ed., 1958) [hereinafter CREATED EQUAL?].
Id. at 390.
Id. at 392.
Id. at 388-89.
347 U.S. 483 (1954).
n39 Voting Rights Act of 1965,
42 U.S.C. §§ 1971, 1973 (1988).
See CREATED EQUAL?,
supra note 32, at 374.
See U.S. CONST. art. I,
§ 2, cl. 3.
See id. art. I,
§ 9, cl. 1.
See id. art. IV,
§ 2, cl. 3.
Scott v. Sandford, 60 U.S. (19 How.) 393 (1857).
id. at 404-05.
See MILTON FRIEDMAN, CAPITALISM AND FREEDOM 200 (1962); MILTON FRIEDMAN
& ROSE FRIEDMAN, FREE TO CHOOSE 134-36 (1980); HAYEK,
supra note 5, at 85-86, 99-100; NOZICK,
supra note 5, at 149, 167-74.
Rawls does not state this view explicitly, it is necessary in order to make sense of
"fact of reasonable pluralism" and the role it plays in supporting the priority of the right. He notes that
reasonable disagreements may arise over what policies fulfill the difference
principle, but adds,
"[t]his is not a difference about what are the correct principles but simply a
difference in the difficulty of seeing whether the principles are achieved" (p. 230).
supra note 3, at 72-75, 100-07, 136-42, 310-15.
See id. at 20-21, 48-51, 120, 577-87.
n50 In this paragraph, I draw on some of the arguments for and against the
morality of homosexuality that appear in John Finnis
& Martha Nussbaum,
Is Homosexual Conduct Wrong?: A Philosophical Exchange, NEW REPUBLIC, Nov. 15, 1993, at 12-13; Stephen Macedo,
The New Natural Lawyers, HARV. CRIMSON, Oct. 29, 1993, at 2; and Harvey C. Mansfield,
Saving Liberalism From Liberals, HARV. CRIMSON, Nov. 8, 1993, at 2.
n51 An alternative line of reply might undertake to defend promiscuity and to deny
that the goods of love and responsibility are necessary to the moral worth of
sexuality. From this point of view, the line of argument I suggest mistakenly
seeks to defend the moral legitimacy of homosexuality by way of an analogy with
See BONNIE HONIG, POLITICAL THEORY AND THE DISPLACEMENT OF POLITICS 186-95 (1993).
n52 It is possible to argue for certain gay rights on grounds that neither affirm
nor deny the morality of homosexuality. The question here is whether
government is justified in supporting laws or policies (such as gay marriage,
for example) on grounds that affirm the moral legitimacy of homosexuality.
Rawls states that the limits of public reason apply to all discussions involving
constitutional essentials and basic justice. As for other political questions,
he writes that
"it is usually highly desirable to settle political questions by invoking the
values of public reason. Yet this may not always be so" (pp. 214-15).
n54 This idea is repeated at several other points (pp. 215, 224, 254).
478 U.S. 186 (1986).
See Michael J. Sandel,
Moral Argument and Liberal Toleration: Abortion and Homosexuality,
77 CAL. L. REV. 521, 534-38 (1989).
See ERIC FONER, POLITICS AND IDEOLOGY IN THE AGE OF THE CIVIL WAR 72 (1980);
AILEEN S. KRADITOR, MEANS AND ENDS IN AMERICAN ABOLITIONISM 78, 91-92 (1967);
JAMES M. MCPHERSON, BATTLE CRY OF FREEDOM: THE CIVIL WAR ERA 7-8 (1988).
n58 I cannot elaborate this claim here, but will try to do so in a forthcoming
book, provisionally entitled
Liberal Democracy in America: In Search of a Public Philosophy.
Prepared: January 24, 2003 - 5:02:29 PM
Edited and Updated, January 25, 2003
Philosophy 111A Page