Families, Associations, and Political Pluralism
Families, Associations, and Political Pluralism
William A. Galston 0 1
0 Thi s Article is brought to you for free and open access by FLASH: The F ordham Law Archive of Scholarship and History. It has been accepted for inclusion in Fordham Law Review by an authorized editor of FLASH: The F ordham Law Archive of Scholarship and History. For more information , please contact
1 William A. Galston , Families, Associations, and Political Pluralism, 75 Fordham L. Rev. 815 (2006). Available at:
INTRODUCTION: A SKETCH OF LIBERAL PLURALISM
We often use the phrase "liberal democracy," but we do not always think
about it very carefully. The noun points to a particular structure of politics
in which decisions are made, directly or indirectly, by the people as a
whole; more broadly, it points to an understanding of politics in which all
legitimate power flows from the people. The adjective points to a particular
understanding of the scope of politics, in which the domain of legitimate
political decision making is seen as inherently limited. Liberal governance
acknowledges that important spheres of human life are wholly or partly
outside the purview of political power. It stands as a barrier against all
forms of total power, including the power of democratic majorities.
The question then arises: How are we to understand the nature and
extent of limits on government? The signers of the Declaration of
Independence appealed to the self-evidence of certain truths, among them
the concept of individuals as bearers of rights that both orient and restrict
governmental power. Today, individual rights represent an
importantsome would say excessive-part of our moral vocabulary. The question is
whether they are sufficient to explain and justify the full range of
constraints we may wish to impose on the exercise of public power-for
example, limits on government's right to intervene in the internal affairs of
In two recent books, 1 I argue that we must develop a more complex
theory of the limits to government. In this endeavor, three concepts are of
special importance. The first is "political pluralism," an understanding of
social life that comprises multiple sources of authority-individuals,
* William A. Galston is Senior Fellow in Governance Studies at the Brookings Institution
and College Park Professor at the University of Maryland.
1. William A. Galston, Liberal Pluralism: The Implications of Value Pluralism for
Political Theory and Practice (2002) [hereinafter Galston, Liberal Pluralism]; William A.
Galston, The Practice of Liberal Pluralism (2005) [hereinafter Galston, Practice].
parents, civil associations, faith-based institutions, and the state, among
others-no one of which is dominant in all spheres, for all purposes, or on
all occasions. Because so many types of human association possess an
identity not wholly derived from the state, pluralist politics does not
presume that the inner structure and principles of every sphere must mirror
those of basic political institutions. For example, in filling positions of
religious authority, faith communities may use, without state interference,
gender-based norms that would be forbidden in businesses and public
The second key concept is "value pluralism," made prominent by the late
British philosopher Isaiah Berlin. This concept offers an account of the
moral world we inhabit: While the distinction between good and bad is
objective, there are multiple goods that differ qualitatively from one another
and which cannot be rank-ordered. If this is the case, there is no single way
of life, based on a singular ordering of values, that is the highest and best
for all individuals. This has important implications for politics. While
states may legitimately act to prevent the great evils of human existence,
they may not seek to force their citizens into one-size-fits-all patterns of
desirable human lives. Any public policy that relies upon, promotes, or
commands a single conception of human good or excellence as equally
valid for all individuals is on its face illegitimate.
The modes of argument that lead me toward these two forms of pluralism
are similar: An unblinkered inspection of the way each of these domains
happens to be structured points toward basic plurality. It might have been
otherwise (logically or metaphysically), but it is not. When we attend to the
way moral arguments go in specific cases, we reach a point where the
assertion that value "A" dominates all others in all situations comes to
appear forced and implausible-the sort of claim one would make only if
one is determined to maintain one's position at all costs. 2 Likewise, at
some point the contention that claims flowing from political ends or
necessities trump all others goes against the grain of normal human
responses. I shall argue, for example, that the pressure from political
authorities on children to betray their parents, or on sisters to ignore family
ties and treat brothers simply as enemies of the state, runs against the grain
of our humanity. Again, it might have been otherwise, but it is not. I am
not trying to do Kantian-style political or moral philosophy valid for "all
rational beings" (whatever that might mean) but rather for the kind of
beings we are, in the circumstances in which we are placed.
The third key concept in my account of limited government is
"expressive liberty." This is a presumption in favor of individuals and
groups leading their lives as they see fit, within the broad range of
legitimate variation defined by value pluralism and in accordance with their
own understandings of what gives life meaning and value. Expressive
liberty may be understood as an extension of the free exercise of religion,
2. For more on this, see Galston, Practice, supra note 1, at 15-27.
generalized to cover comprehensive conceptions of human life that rest on
both nonreligious and religious claims. This extension applies to groups as
well as individuals. Some shared cultural understandings go just as deep as,
and define identity just as much as, shared religious beliefs. They are
equally entitled to deference-again, within limits. 3 There is a presumption
in favor of this expanded conception of free exercise, and a liberal pluralist
state must discharge a burden of proof whenever it seeks to restrict it.
A plausible objection: Does this presumption not elevate expressive
liberty above other values, and is that not inconsistent with the basic
structure of value pluralism? No. Value pluralists need not (and in my
view should not) assert that all goods and principles are on a par, arranged
purely horizontally. There may be good reasons for giving some greater
weight than others or for ascribing presumptive force to a subset of values. 4
Value pluralists need only deny that one or more values override all others,
without regard to circumstances or to the magnitudes of the tradeoffs and
sacrifices that the once-and-for-all elevation of some values over others
Why a presumption in favor of expressive liberty? Because having a
vision of the way you want your life to go is not a theoretical matter.
Having that vision, as we all do, means wanting to live in one way rather
than others. So expressive liberty is a value, the implementation of which
enables individuals to realize a wide range of other values. To restrict
individuals' expressive liberty is to deprive them of what they cannot help
regarding as a very great good. There may be reasons to do so, but these
reasons must be weighty; hence the presumption.
This standard for state action is demanding, but hardly impossible to
meet. While expressive liberty is a very important good, it is not the only
good, and it is not unlimited. First, the social space within which differing
visions of the good are pursued must be organized and sustained through
the exercise of public power; to solve inevitable problems of coordination
among divergent individuals and groups, the rules constituting this space
will inevitably limit in some respects their ability to act as they see fit.
Second, there are some core evils of the human condition that states have
the right (indeed the duty) to prevent; to do so, they may rightly restrict the
actions of individuals and groups. Third, the state cannot sustain a free
social space if its very existence is jeopardized by internal or external
threats; within broad limits it may do what is necessary to defend itself
against destruction, even if self-defense restricts valuable liberties of
individuals and groups. A free society is not a suicide pact.
Liberal pluralists, then, endorse the essential conditions of public order,
such as the rule of law and a public authority with the capacity to enforce it.
3. For a detailed account of the meaning of taking culture seriously, see id. at 176-82
(discussing Brian Barry, Culture and Equality: An Egalitarian Critique of Multiculturalism
4. For more on how presumptions operate within value pluralism, see Galston, Liberal
Pluralism, supra note 1,at 69-78.
They also endorse what may be called a "minimal universalism," or the
moral and practical necessity of organizing public life so as to ward off, to
the greatest extent possible, the great evils of the human condition, such as
tyranny, genocide, mass starvation, and deadly epidemics. 5 (I call the
human condition characterized by the absence of the great evils one of basic
decency.) This minimal universalism overlaps with contemporary
movements for universal human rights and the provision of basic needs.
So understood, a liberal pluralist government is robust within its
appropriate sphere. In securing the cultural conditions of its survival and
perpetuation, for example, it may legitimately engage in civic education,
carefully restricted to the public essentials-the virtues and competences
that citizens will need to fulfill diverse roles in a liberal pluralist economy,
society, and polity. Limits on the scope of democratic authority do not
imply limits on the scope of democratic participation. Liberal pluralists
may favor representative institutions or more direct self-government, strong
executives or parliamentary supremacy. Democracy can be "strong"
without being "total." Indeed, if the reasons in favor of liberal pluralism are
compelling, then in practice self-government can be strong only if it does
not strive to regulate every aspect of our lives.
One thing above all is clear: Because the likely result of liberal pluralist
institutions and practices will be a highly diverse society, the virtue of
tolerance will be a core attribute of liberal pluralist citizenship. This type of
tolerance does not mean wishy-washiness or the propensity to doubt one's
own position, the sort of thing Robert Frost had in mind when he defined a
liberal as someone who cannot take his own side in an argument. 6 It neither
implies, nor requires, an easy relativism about the human good; indeed, it is
compatible with engaged moral criticism of those with whom one differs.
Toleration rightly understood means the principled refusal to use coercive
state power to impose one's views on others, and therefore a commitment to
moral competition through recruitment and persuasion alone.
Liberal pluralism is, in the terms John Rawls made familiar, a
"comprehensive" rather than "political" theory. 7 I believe it makes sense to
connect what one believes to be the best account of public life with
comparably persuasive accounts of morality, human psychology, and the
natural world. As a practical matter, of course, it makes sense to seek
overlapping consensus. Politics as we know it would come to a halt if
cooperation required agreement, not only on conclusions, but on premises
as well. But philosophical argument, even concerning politics, need not
mirror the structure of public life. A political philosopher may assert that
"X" is true and foundational for a particular understanding of a good,
decent, or just society, without demanding that all citizens affirm the truth
5. For development of this point, see Stuart Hampshire, Justice Is Conflict (2000).
6. While this remark is widely attributed to Robert Frost, it does not appear in his
published poetry or interviews.
7. See John Rawls, Justice as Fairness: A Restatement 14 (Erin Kelly ed., 2001).
of "X." Indeed, the founders of a political regime may publicly proclaim
what they take to be moral, metaphysical, or religious truths as the basis of
that regime without insisting that all citizens assent to those truths. In the
United States, naturalizing citizens affirm their loyalty to the Constitution,
not the Declaration of Independence, and all citizens pledge allegiance to
the republic for which the flag stands, not John Locke or Francis
Hutcheson. So I disagree with Martha Nussbaum when she suggests that
making public claims about foundational truths somehow signals disrespect
for those who dissent. 8 Disrespect requires something more-namely, the
use of coercive state power to silence and repress dissenters. Respect
requires not parsimony in declaring truth, but restraint in the exercise of
power. By limiting the scope of legitimate public power, liberal pluralism
does all that is necessary to secure the theoretical and institutional bases of
I. WHY POLITICAL PLURALISM?
I turn now to the main focus of this essay-political pluralism,
understood as a distinctive empirical account of human society that also
supports a specific normative account of political authority.
A liberal democracy is, among other things, an invitation to struggle over
the control of individual and associational activities. State versus society
debates have recurred over the past century of U.S. history, frequently
generating landmark U.S. Supreme Court cases. While the specific issues
vary, the general form is the same. On one side are general public
principles that the state seeks to enforce; on the other are specific beliefs
and practices that individuals and associations seek to protect. Boy Scouts
of America v. Dale9 is a recent chapter in what will no doubt be a
Within the U.S. constitutional context, these issues are often debated
using terms such as free exercise of religion, freedom of association, or the
individual liberty broadly protected under the Fourteenth Amendment.
Rich and illuminating as it is, this constitutional discourse does not go deep
enough. It is necessary to reconsider the understanding of politics that
pervades much contemporary discussion (especially among political
theorists), an understanding that tacitly views public institutions as
plenipotentiary and civil society as a political construction possessing only
those liberties that the polity chooses to grant and modify or revoke at will.
This understanding of politics makes it all but impossible to give serious
weight to the "liberal" dimension of liberal democracy.
The most useful point of departure for the reconsideration of politics I am
urging is found in the writings of the British political pluralists and pluralist
8. Martha C. Nussbaum, PoliticalObjectivity, 32 New Literary Hist. 883, 888 (2001).
9. 530 U.S. 640 (2000).
thinkers working in the Calvinist tradition. 10 This pluralist movement
began to take shape in the nineteenth century as a reaction to the growing
tendency to see state institutions as plenipotentiary. This tendency took
various practical forms in different countries: French anticlerical
republicanism, British parliamentary supremacy, and the drive for national
unification in Germany and Italy against subordinate political and social
powers. Following Stephen Macedo (though disagreeing with him in other
respects), I shall call this idea of the plenipotentiary state "civic totalism."' "I
Historically, one can discern at least three distinct secular-theoretical
arguments for civic totalism. 12 The first is the idea, traced back to
Aristotle, that politics enjoys general authority over subordinate activities
and institutions because it aims at the highest and most comprehensive good
for human beings. The Politics virtually begins with the proposition that
"all partnerships aim at some good, and ...the partnership that is most
authoritative of all and embraces all the others does so particularly, and
aims at the most authoritative good of all. That is what is called ...the
political partnership." 13
Thomas Hobbes offered a second kind of justification for civic totalism:
Any less robust form of politics would, in practice, countenance divided
sovereignty-the dreaded imperium in imperio, an open invitation to civic
conflict and war. 14 Sovereignty cannot be divided, even between civil and
spiritual authorities. In Hobbes's view, undivided sovereign authority has
unlimited power to decide whether, and under what conditions, individuals
and associations would enjoy liberty of action. No entity, individual or
collective can assert rights against the public authority. Indeed, civil law
may rightfully prohibit even the teaching of truth, if it is contrary to the
requirements of civil peace.
A third argument for civic totalism was inspired by Jean-Jacques
Rousseau: Civic health and morality cannot be achieved without citizens'
wholehearted devotion to the common good. Loyalties divided between the
republic and other ties, whether to civil associations or to revealed religious
truth, are bound to dilute civic spirit. And the liberal appeal to private life
as against public life will only legitimate selfishness at the expense of the
spirit of contribution and sacrifice without which the polity cannot endure.
Representing this tradition, Emile Combes, a turn-of-the-century premier in
the French Third Republic, declared that "[t]here are, there can be no rights
except the right of the State, and there [is], and there can be no other
authority than the authority of the Republic."' 15
These three traditions may seem far removed from the mainstream of
contemporary views. Does the "liberal" strand of liberal democracy not
qualify and limit the legitimate power of the state? Is this not the entering
wedge for a set of fundamental freedoms that can stand against the claims
of state power?
The standard history of liberalism lends support to this view. The rise of
revealed religion created a division of authority and challenged the
comprehensive primacy of politics. The early modem wars of religion
sparked new understandings of the relation between religion and politics,
between individual conscience and public order, and between unity and
diversity. As politics came to be understood as limited rather than total, the
possibility emerged that the principles constituting individual lives and civil
associations might not be congruent with the principles constituting public
institutions. The point of liberal constitutionalism, and of liberal
statesmanship, was not to abolish these differences but rather, so far as
possible, to help them serve the cause of ordered liberty.
II. THE TOTALIST TEMPTATION
Despite this history, many contemporary theorists, including some who
think of themselves as working within the liberal tradition, embrace
propositions that draw them away from the idea of limited government and
toward civic totalism, perhaps against their intention. Some come close to
arguing that if state power is exercised properly-that is, democratically-it
need not be limited by any considerations other than those required by
Jurgen Habermas offers the clearest example of this tendency. He insists
that once obsolete metaphysical doctrines are set aside, "there is no longer
any fixed point outside the democratic procedure itself."' 16 But this is no
cause for worry or regret: Whatever is normatively defensible in liberal
rights is contained in the discourse-rights of "sovereign [democratic]
citizens." 17 The residual rights not so contained constitute not bulwarks
against oppression, but rather the illegitimate insulation of "private"
practices from public scrutiny.
An eminent American democratic theorist, Robert A. Dahl, is tempted by
Habermas's stance. He characterizes as "reasonable" and "attractive" the
15. See J.N. Figgis, The Great Leviathan, in The Pluralist Theory of the State, supra
note 10, at 112 (quoting Emile Combes).
16. John Rawls, Political Liberalism 379 (1996) (quoting and discussing Habermas's
17. Id. at 379 n.10.
view that members of political communities have no fundamental interests,
rights, or claims other than those integral to the democratic process or
needed for its preservation. 18 The only limits to the legitimate scope of
democratic power are the requisites of democracy itself. Put simply, a
demos that observes the norms of democratic decision making may do what
it wants. 19
Unlike Habermas, Dahl is not entirely comfortable with restricting the
domain of rights to the conditions of democracy. He concedes that this
proposal raises "disturbing" questions: What about interests, rights, and
claims that cannot be adequately understood as aspects of the democratic
process but which nonetheless seem important and defensible? What about
fair trials, or freedom of religion and conscience? Without definitively
answering these questions, Dahl examines the various ways in which the
defense of rights may be institutionalized, concluding that those who would
temper democratic majorities with "guardian" structures such as courts bear
a heavy burden of proof that they rarely-if ever-discharge successfully. 20
The most reliable cure for the ills of democracy is more democracy; the
resort to non-majoritarian protections risks undermining the people's
capacity to govern itself.2 1
John Rawls presents the most complex case of the phenomenon I call the
"totalist temptation." He asserts that "the values of the special domain of
the political.., normally outweigh whatever values may conflict with
them[.] ' 22 Why is this the case? Rawls offers two reasons. First, political
values are very important; they determine social life and make fair
cooperation possible. Second, conflict between political and nonpolitical
values can usually be avoided, so long as political values are appropriately
Rawls famously maintains that justice is the preeminent political value,
the "first virtue of social institutions" and that "laws and institutions no
matter how efficient and well-arranged must be reformed or abolished if
they are unjust. '24 Nonetheless, he asserts, consistent with the liberal
tradition, that the principles of justice do not directly regulate institutions
and associations-such as churches and families-within society.2 5
The difficulty is to explain why, within the structure of Rawls's theory,
the principles regulating the basic structure of society should not be applied
directly to institutions such as church and the family. Taken literally, many
of these background principles would seem to warrant such interventions.
18. Robert A. Dahl, Democracy and Its Critics 182-83 (1989).
20. Id. at 187-91.
21. Id. at 183-92.
22. Rawls, supranote 16, at 139, 157.
23. See id. at 139-40.
24. John Rawls, A Theory of Justice 3 (1971).
25. Rawls, supra note 7, at 10. Principles of justice do affect these institutions
indirectly, via the influence of just background institutions.
For example, imbalances in parenting responsibilities can affect women's
"fair equality of opportunity. '26 Does this mean, as Rawls seems to
suggest, that "special provisions... [of] family law" should prevent or
rectify this imbalance? 27 If the family is part of the basic structure of
society, as Rawls now claims, why does he judge it "hardly sensible" that
parents be required to treat their children in accordance with the principles
directly governing the basic structure? 28
The ambiguous status of the family reflects a deeper structural problem
in Rawls's account. At one point, he offers a formulation that seems
promising: We distinguish between the point of view of citizens and of
members of associations. As citizens, we endorse the constraints of
principles of justice; as association members, we want to limit those
constraints so that the inner life of associations can flourish. This generates
a "division of labor" that treats the basic structure and civil association as
being, so to speak, on a par with one another.29
But on closer inspection, it turns out that there is a hierarchical relation
after all, with the principles of justice serving as trumps. Otherwise put, the
basic structure constitutes the end, and the various associations are, in part,
means to that end. So, for example, "the treatment of children must be such
as to support the family's role in upholding a constitutional regime." 30 But
what if religious free exercise includes teachings and practices that do not
do this? (Imagine a religious group that has no intention of altering the
public structure of equal political rights for women but teaches its own
members that women should not participate in public life.)
Rawls is certain (quite sensibly in my view) that "[w]e wouldn't want
political principles of justice to apply directly to the internal life of the
family." 3 1 The reasoning appears to be that various associations have inner
lives that differ qualitatively from that of the political realm, so that
political principles would be "out of place." 32 This then raises questions:
Why are political and nonpolitical associations not understood as related
horizontally rather than vertically? Why can nonpolitical associations not
be seen as limiting the scope of politics at the same time that the basic
structure of politics constrains associations?
Rawls's apparent answer runs as follows: The domain of the nonpolitical
has no independent existence or definition but is simply the result (or
residuum) of how the principles of justice are applied directly and
indirectly. In particular, the principle of equal citizenship applies
In one sense, this is clearly true. If an association uses coercion to
prevent some of its members from exercising their equal political rights, the
state must step in to enforce those rights. But the more usual case is that the
association organizes itself according to norms of membership or activity
that are inconsistent with principles of equal citizenship. What is the state's
legitimate power in the face of these dissenting practices? Is it so obvious
that the legitimate activities of nonpolitical associations should be defined
relative, not to the inner life of those associations, but to the principles of
the public sphere? Can we not say something important about the
distinctive natures of individual conscience, friendship, families,
communities of faith, or inquiry? Should not those primary features of our
social life have an effect on the scope of political principles, not just vice
versa? Even if justice is the "first virtue" of public institutions and enjoys
lexical priority over other goods of the public realm (a debatable
proposition), does it follow that the public realm enjoys comprehensive
lexical priority over the other forms of human activity and association? 33
III. THE PLURALIST ALTERNATIVE
It is in the context of questions such as these that political pluralism
emerges as an alternative to all forms of civic totalism. Political pluralism,
to begin, rejects efforts to understand individuals, families, and associations
simply as parts within and of a political whole. Relatedly, pluralism rejects
the instrumental/teleological argument that individuals, families, and
associations are adequately understood as "for the sake of' some political
purpose. Religion is not (only) civil and in some circumstances may be in
tension with civil requirements. As Linda McClain properly reminds us, at
their best, families not only help develop good citizens but also serve as
sites of diversity limiting public power. 34
This is not to say that political communities must be understood as
without common purposes. The political order is not simply a framework
within which individuals, families, and associations may pursue their own
purposes. As we have seen, there are core evils that only decent politics can
minimize, and core goods that only sensible politics can promote. But civic
purposes are not comprehensive and do not necessarily trump the purposes
of individuals and groups.
Political pluralism understands human life as consisting in a multiplicity
of spheres, some overlapping, with distinct natures and/or inner norms.
Each sphere enjoys a limited but real autonomy. Political pluralism rejects
any account of political community that creates a unidimensional
hierarchical ordering among these spheres of life. Rather, different forms
33. For more on civic totalism, see Galston, Practice, supranote 1, at 23-44.
34. Linda C. McClain, The Place of Families: Fostering Capacity, Equality, and
Responsibility 20 (2006). The political pluralism I present in this essay represents an
unorthodox account of what McClain calls the 'jurisdictional rationale" for delimiting public
power. See id. at 30.
of association and activity are complexly interrelated. There may be local
or partial hierarchies among subsets of spheres in specific contexts, but
there are no comprehensive lexical orderings among categories of human
For these reasons, among others, political pluralism does not seek to
overcome, but rather endorses, the post-pagan division of human loyalty
and political authority created by the rise of revealed religion. That this
creates problems of practical governance cannot be denied. But pluralists
refuse to resolve these problems, like Hobbes, by allowing public
authorities to determine the substance and scope of allowable belief or, like
Rousseau, by reducing faith to civil religion and elevating devotion to the
common civic good as the highest human value. Fundamental tensions
rooted in the deep structure of human existence cannot be abolished in a
stroke but rather must be acknowledged, negotiated, and adjudicated with
due regard to the contours of specific cases and controversies.
Pluralist politics is a politics of recognition rather than of construction. It
respects the diverse spheres of human activity; it does not understand itself
as creating or constituting those activities ex nihilo. Families are shaped by
public law, but that does not mean that they are wholly socially constructed.
There are complex relations of mutual impact between public law and faith
communities, but it is preposterous to claim that the public sphere creates
these communities. Do environmental laws create air and water?
Pluralist politics is, however, responsible for coordinating other spheres
of activity, and especially for adjudicating the inevitable overlaps and
disputes among them. This form of politics evidently requires the mutual
limitation of some freedoms, individual and associational. It monopolizes
the legitimate use of force, except in cases of self-defense when the polity
cannot or does not protect us. It understands that group tyranny is possible
and therefore protects individuals against some associational abuses. But
pluralist politics presumes that the enforcement of basic rights of
citizenship and of exit rights, suitably understood, will usually suffice.
Associational integrity requires a broad-though not unlimited-right of
groups to define their own membership, to exclude as well as include, and a
pluralist polity will respect that right.35
35. Granted, "exit rights" is a general concept that admits of a range of possible
specifications. I have defended a relatively robust conception, one that warrants a greater
measure of state interference in religious and cultural communities than many
dyed-in-thewool pluralists can accept. See Galston, Liberal Pluralism, supra note 1, at 122-23. That has
not prevented fervent egalitarians from attacking my conception as insufficiently attentive to
the plight of vulnerable minorities within minority communities. See Susan Moller Okin,
"Mistresses of Their Own Destiny": Group Rights, Gender, and Realistic Rights of Exit,
112 Ethics 205 (2002). For my response to Okin on this point, see Galston, Practice, supra
note 1, at 182-85.
FORDHAM LAW RE VIEW
IV. VARIETIES OF ARGUMENTS FOR POLITICAL PLURALISM
The core of my thesis is that different forms of human activity and
association generate different kinds of claims, both to liberty and authority,
and that no single ensemble of claims dominates the rest for all purposes or
in all circumstances. As a distinctive form of activity and association,
politics both makes claims and is limited in various ways by claims
deriving from other sources. The question is how to justify this thesis.
Joseph Raz offers an argument for limited political authority based on
two premises: the idea of responsibility as the core of human agency and
an understanding of political authority as derived from the consent of
individual agents. Raz contends that while individuals can legitimately
transfer responsibility for some aspects of their lives to others, they cannot
abandon responsibility altogether without undermining agency. Thus, he
concludes, "[o]nly limited government can be legitimate. ' '36
Limited how, exactly? Raz argues that a doctrine specifying the content
of limited government would have two parts. The first is instrumental:
Delegating responsibility to government lacks a rationale in areas where
individuals can act for themselves as efficiently and effectively as others
can act for them. The second principle of limitation is avowedly
noninstrumental: It consists of all those matters regarding which "it is more
important to act independently than to succeed in doing the best."' 37 Raz
remarks, laconically, that he "feel[s] the need for a substantive account of
this category," an account he does not provide. 38 Any effort to do so would
quickly encounter, inter alia, well-worn debates about the justification of
paternalism-that is, the category of interventions in which considerations
of an agent's well-being trump those of responsibility or autonomy.
Much liberal thought grounds limits to government on some conception
of individual rights. While this line of argument has considerable merit, it
is also incomplete. It is not easy to move from the concept of rights to a
specific conception. Most Americans can recite the Declaration's famous
litany of life, liberty, and the pursuit of happiness; few realize that the
words "among these are" precede this triad, 39 raising but not answering the
question concerning the content of these unnamed additional rights.
Similarly, the Ninth Amendment cautions that "[t]he enumeration in the
Constitution of certain rights, shall not be construed to deny or disparage
others retained by the people." 40 But what exactly are these retained rights?
The language of individual rights goes astray, moreover, not when it
claims to be a source of limits on governmental authority, but when it
presents itself as the sole source. I take seriously the idea of individual
inviolability, rightly understood, and I defend in detail one aspect of
36. Joseph Raz, Introductionto Authority 1, 12 (Joseph Raz ed., 1990).
37. Id. at 13.
39. The Declaration of Independence para. 2 (U.S. 1776).
40. U.S. Const. amend. IX.
inviolability-namely, freedom of conscience. 4 1 But I want to suggest that
we may do better to proceed more empirically, by considering the diverse
forms of human sociability and association.
For example, family obligations can limit the scope of political authority.
Sophocles' Antigone revolves around primordial imperatives of kinship that
stand opposed to the imperatives of patriotic loyalty. The fact that one of
Antigone's brothers was slain in battle against his own city does not per se
justify Creon's effort to prevent her from burying him. To be sure,
Antigone is as deaf to Creon's legitimate concerns for his city as he was to
her family ties. One might (just) make out a case for Creon's stance. Still,
the playwright presents the disaster that befalls Creon as the result of his
extension of political authority beyond its rightful limits. Whether we
ultimately agree with Sophocles or not, the basic point is that political
leaders cannot rightly assign a weight of zero to nonpolitical values.
A. Family Ties
American constitutional law endorses the proposition that family ties
limit political authority. In the first place, the status of parenthood
generates a sphere of authority. For example, in the famous case of Pierce
v. Society of Sisters,42 the Supreme Court rejected the right of a public
authority (in this case, the state of Oregon) to require all parents within its
jurisdiction to send their children to public schools. In justifying its stance,
the Court declared,
The fundamental theory of liberty upon which all governments in this
Union repose excludes any general power of the State to standardize its
children by forcing them to accept instruction from public teachers only.
The child is not the mere creature of the State; those who nurture him and
direct his destiny have the right, coupled with the high duty, to recognize
and prepare him for additional obligations.4 3
This does not mean, of course, that formative family activities are wholly
immunized from political regulation. The Pierce Court explicitly
recognized a substantial degree of legitimate governmental regulation of
families' educational decisions. For example, the theory of liberty on
which the Court relied allows the state to require "that all children of proper
age attend some school," to regulate all schools "reasonably," and to ensure
that "certain studies plainly essential to good citizenship" are taught and
that those "manifestly inimical to the public welfare" are not.4 4
Other aspects of family life also enjoy at least partial protection from
state power. For example, there is a strong presumption that confidential
communications between spouses cannot be introduced into evidence in
either criminal or civil cases over the objection of either spouse. As Justice
41. See Galston, Practice, supranote 1, at 45-71.
42. 268 U.S. 510 (1925).
43. Id. at 535.
44. Id. at 534 (emphasis added).
Stone stated in Wolfle v. United States,45 this presumption rests on the
belief that the protection of marital confidences is "so essential to the
preservation of the marriage relationship as to outweigh the disadvantages
to the administration of justice which the privilege entails." 46 Justice
Harlan Fiske Stone regarded it as unnecessary to state the implied
philosophical argument: The intimate good of marital trust is so important
as to outweigh the civic good of legal justice. Pluralism, in short, is the
underlying theory of established practice.
This theory can be understood as either empirical or normative sociology.
The third part of Dean Wigmore's famous four-part test for the recognition
of confidential communications privileges is that the relationship protected
by the privilege is "one which the public believes should be fostered. 47
But suppose the public came to believe that relations between husbands and
wives, doctors and patients, or clergy and penitents are not so important
after all, are perhaps even detrimental ,to key human and social goods.
From Wigmore's empirical standpoint, the case for the privilege would be
gravely weakened. Not so from the standpoint of pluralist theory, because
part of the point of that theory is to identify important goods that publics
and legal systems ought to take seriously.48 Otherwise, the case for liberal
pluralism would depend on the vagaries of public opinion.
This point is strengthened when we broaden our inquiry. The relations
between family-based and political authority, delicate enough within liberal
democracies, become fraught with danger in oppressive regimes. Families
that cooperate with schools and civil associations to shape their children
into good citizens of bad governments will almost certainly imperil
children's human and moral development and may distort family life as
well. (Can parents share intimate thoughts within the household if they fear
that their own children may feel duty-bound to report them to the
authorities?) The worse the regime, the more parents must tilt childrearing
toward resistance rather than civic integration; but the worse the regime, the
more dire the consequences for resisting families, including their children,
are likely to be. By historical standards, families are fortunate if their
circumstances do not transform the inherent tensions of their civic role into
outright contradictions, with tragic consequences.
United States law and jurisprudence limit the sway of public authority
over religious associations. I have already made reference to the
commonplace notion that these associations may establish their own criteria
for their religious offices, general public norms of nondiscrimination to the
contrary notwithstanding. But these limits go even deeper. As Laurence
Tribe observes, courts and other agencies of the U.S. government "may not
inquire into pervasively religious issues." 49 The rationale for this restriction
goes beyond prudential fears of entanglement and political divisiveness; it
reflects the belief that doctrinal and scriptural interpretations are beyond the
competence and rightful authority of political power.
Democratic polities have not always acknowledged these limits. Well
into the twentieth century, the British House of Lords operated on the
premise that property was contributed to religious bodies pursuant to an
"implied trust" framed by the doctrines and practices of those bodies
prevailing at the time of the donation. When doctrinal disputes and schisms
occurred, the Lords did not hesitate to adjudicate property claims on the
basis of their own interpretation of the litigants' fidelity to those doctrines
and practices, an intrusion against which British pluralists such as J.N.
Figgis protested bitterly but to no avail. 50
As early as 1872, however, U.S. courts abandoned the implied trust
doctrine in favor of the rule that whenever the ordinary principles of
contract and property law did not resolve disputes within religious
associations, courts should defer either to the majority in congregational
churches or to the highest authority in hierarchical churches. The Court's
argument for this position was rooted in principle as well as prudence:
The law knows no heresy, and is committed to the support of no dogma,
the establishment of no sect. The right to organize voluntary religious
associations to assist in the expression and dissemination of any religious
doctrine, and to create tribunals for the decision of controverted questions
of faith within the association, and for the ecclesiastical government of all
the individual members, congregations, and officers within the general
association, is unquestioned. All who unite themselves to such a body do
so with an implied consent to this government, and are bound to submit to
it. But it would be a vain consent and would lead to the total subversion
of such religious bodies, if any one aggrieved by one of their decisions
could appeal to the secular courts and have them reversed. It is of the
essence of these religious unions, and of their right to establish tribunals
for the decisions of questions arising among themselves, that those
decisions should be binding in all cases of ecclesiastical cognizance,
subject only to such appeals as the organism itself provides for. 5 1
There is, in short, a sphere of religious authority distinct from, and limiting,
the scope of rightful political authority. 52
This is not exactly a new idea. In the familiar New Testament story of
the silver coin, Jesus responds to a politically charged inquiry about paying
taxes to Rome with the maxim, "Give to Caesar what is Caesar's, and to
God what is God's. ' 53 The task of tracing the line between God's realm
and Caesar's has challenged political thought, secular and theological, for
two millennia. But the difficulty of this task does not obviate its necessity,
once we acknowledge the independent force of religious claims.
CONCLUSION: PLURALIST THEORY AND PLURALIST CASUISTRY
This line of argument is bound to leave many readers unsatisfied. No,
religious communities may not engage in human sacrifice, and if they try,
political authorities have the right and duty to stop them. Yes, religious
communities may assign positions of religious authority on grounds that
would not be permitted in the public domain, and it would be wrong for
political authorities to prevent them. But is there not a broad middle ground
of issues between these extremes, which liberal pluralist theory does not
help us resolve?
This is a fair objection, as far as it goes, but it does not go as far as those
who urge it think it does. It reflects, I would suggest, more of a
disagreement about the possibilities of theory than a critique of liberal
pluralism in particular. In my view, theory mostly provides a template of
considerations that the sound exercise of practical reason must take into
account. It will yield very few bright lines that neatly resolve entire
categories of controversies. Rather, it structures the productive conduct of
the form of practical reason known as casuistry. It yields results that look
much more like common law than the clear rules of law that many students
of jurisprudence crave. Broader patterns are inductive and emergent, not
laid down in advance.
This common law approach does not mean that liberal pluralism cannot
reach clear conclusions in specific cases. I have argued to the contrary; for
example, it vindicates famous Supreme Court decisions such as Pierce v.
Society of Sisters54 and Wisconsin v. Yoder,55 and it warrants more
accommodation to the plaintiff parents than the courts ultimately accorded
them in Mozert v. Hawkins County Board of Education.56 But liberal
pluralists must always remain open to the possibility that circumstances will
override presumptions. It is impossible to expunge the need for the exercise
of judgment that cannot be reduced to rules.
One need not be a pluralist to embrace some version of this approach.
For example, McClain points out that while (in the language of Pierce)the
"child is not the 'mere creature of the state,' so a child is not the mere
53. Matthew 22:15-22; Mark 12:13-17; Luke 20:20-26 (New International Version).
54. 268 U.S. 510 (1925); see supranotes 42-44 and accompanying text.
55. 406 U.S. 205 (1972).
56. 827 F.2d 1058 (6th Cir. 1987); see Galston, Liberal Pluralism, supra note 1, at
creature of his or her parents." 57 This implies some limits on parental
authority, including their formative authority. I agree, as everyone should
(and just about everyone does). The question is whether it is possible to
move from three broad baskets of interests-parents, children, state-to
specific conceptions of these interests that enable theorists, judges, and
policy makers to draw bright-line distinctions. For cases other than those
that are intuitively clear and therefore strongly consensual, I doubt it. In
practice, I suspect we will be forced to resort to some kind of balancing test,
with all the attendant uncertainty and contestability. Taken together,
political and value pluralism offer the best account I have found so far as to
why intuitionist balancing is an inescapable feature of social life.
57. McClain, supra note 34, at 42.
Notes & Observations
10. For the British tradition, see generally The Pluralist Theory of the State: Selected Writings of G.D.H. Cole , J.N. Figgis , and H.J. Laski (Paul Q. Hirst ed., 1989 ) [hereinafter The Pluralist Theory of the State]. For the Calvinist tradition, see generally Political Order and the Plural Structure of Society (James W . Skillen & Rockne M. McCarthy eds., 1991 ).
11. For the full account of our agreement and (mainly) disagreement, see my review of Stephen Macedo's Diversity and Distrust: Civic Education in a Multicultural Society ( 2000 ), in 112 Ethics 386 ( 2002 ).
12. Theological arguments, which raise a different set of issues, are beyond the scope of these comments .
13. Aristotle , The Politics 35 (Carnes Lord trans ., University of Chicago Press 1984). For present purposes, whether this statement is an adequate representation of Aristotle's full view is a matter we may set aside .
14. For discussion of this and the following points , see Thomas Hobbes, Leviathan 82- 84 , 147 - 53 , 298 - 308 (London, George Routledge and Sons, 2d ed. 1886 ).
45. 291 U.S. 7 ( 1934 ).
46. Id . at 14.
47. Katherine 0. Eldred , "Every Spouse's Evidence": Availability of the Adverse Spousal Testimonial Privilege in FederalCivil Trials , 69 U. Chi . L. Rev. 1319 , 1340 ( 2002 ) (discussing Wigmore's four-part test).
48. For an example of how this deeper argument might work in the case of spousal relations, see generally Milton C. Regan , Jr., Spousal Privilege and the Meanings of Marriage , 81 Va. L. Rev . 2045 ( 1995 ).
49. Laurence H. Tribe , American Constitutional Law 1227 (2d ed. 1988 ).
50. See generallyJ. N. Figgis, supra note 15.
51. Watson v. Jones , 80 U.S. ( 13 Wall.) 679 , 728 - 29 ( 1872 ). While this case was decided on the basis of common law, eighty years later the Court rendered a parallel decision based on the First Amendment . See Kedroff v. Saint Nicholas Cathedral , 344 U.S. 94 ( 1952 ).
52. For a parallel discussion , see Nancy L. Rosenblum, Membership and Morals: The Personal Uses of Pluralism in America 80-83 ( 1998 ).