I. Introduction.

 In this paper I argue that a state-sponsored and state-regulated compulsory moral
education is an essential institutional requirement of all just states, even those only minimally
just.   The argument proceeds in three steps.  The first aims to establish that justice demands the
widespread if not universal distribution among citizens of certain moral capacities, skills,
dispositions and knowledge.  The second aims to show that as a matter of institutional design the
only reliable way to secure such a distribution is through a system of state-sponsored and state-
regulated compulsory moral education.  The third step aims to assess the force of parentalist
objections to state-governed compulsory moral education.  Parents often claim that they have a
prior right to determine the moral education of their children, a right which, if asserted, justifies
immunizing or excusing their children from any regime of compulsory moral education.  Thus,
before I can conclude that all just states, even those only minimally just, will be tutelary states
within which the state plays a central and proper role in the moral education of citizens, I must
show that parents either don’t have this right, or that it is a right insufficient to justify
immunizing or excusing their children from an otherwise justified regime of compulsory moral
education.

II.  Justice and Compulsory Moral Education:  Citizenship and Reciprocity

 A just state is, at a minimum, a constitutional republic within which familiar basic rights
and liberties are secured equally for all, including the right to participate meaningfully in the
democratic selection of legislative officials.  A just state also is, at a minimum, one in which
equality of opportunity is realized to some significant degree; it is a state within which no
individual’s life prospects are determined by the social class into which she is born or by morally
indefensible racist, sexist or similar social prejudices.  Finally, a just state is, at a minimum, one
in which no class of citizens is allowed to fall below an appropriate social minimum (in terms of
economic resources) and in which economic institutions are arranged so as generally to advance
the well-being of all social classes over time (even if they permit an ever-increasing inequality
between classes).  With the foregoing, liberals and conservatives largely agree.

 They also largely agree that no just state can long endure unless citizens possess a certain
range of moral virtues and share a public political culture that reflects and supports those virtues.
But what are these virtues?  Here liberals and conservatives typically march off in different
directions.

 Yet, there is one virtue that liberals and conservatives both affirm as necessary to any just
polity, even one only minimally just.  And that is the virtue possessed by those who honor the
ideal of reciprocity, or justice as reciprocity.  This ideal is implicitly affirmed by everyone who
regards fundamental social and political institutions and forms of authority as conventional and
thus in need of being justified to those subject to them.  And while there are some who do not
hold such a view, most all contemporary conservatives and liberals do; dissenters from this view
stand fully outside the American constitutional tradition.  It is hard to see how anyone could
reject this assumption and still think that coercive political force is legitimate only if there are
reasons for it that it would be unreasonable for those subject to it to reject.  Yet, this is what all,
or virtually all, conservatives and liberals think about coercive political force.

 According to the ideal of reciprocity persons are morally obligated to insure that they
always act toward others in a manner that they are prepared to justify on terms they (the others)
could not reject while remaining committed to the ideal of reciprocity themselves.  Persons who
affirm and honor the ideal of reciprocity are reasonable persons.  Thus, to put the point about the
ideal of reciprocity another way, reasonable persons seek always to act toward others in a manner
no reasonable person could reasonably reject.  Honoring the ideal of reciprocity, then, is the first
and the constitutive virtue of reasonable persons.

 As a virtue of social and political institutions, justice is that which reasonable persons
secure for themselves when they honor the ideal of reciprocity with respect to the full range of
basic social relations and social forces, including coercive political force, into which they are
born and within which they live out their lives.  Reasonable persons, then, affirm and honor the
ideal of justice as reciprocity.  Of course, justice as reciprocity may be to some degree
indeterminate in terms of substantive principles of institutional design.  So, reasonable persons
may reasonably disagree over the details of what exactly justice as reciprocity requires when it
comes to social and political institutions.

 Central to the ideal of reciprocity, and thus also of the reasonable person, is the idea that
moral justification is always addressed to another and advanced from a common or shared moral
point of view, or at least a point of view that a reasonable person could not reasonably reject.
The inclination and ability so to deliberate over and debate the moral justification of basic social
institutions and political arrangements must count, then, as the first virtue, or among the first
virtues, of citizenship.  Conservatives and liberals ought both to agree to this.  Without citizens
inclined and able so to think about and publicly discuss the moral justification of their basic
social structure, no just state, not even a minimally just state, will be able to endure.

 This is an admittedly abstract requirement.  Yet, it does have significant substantive
content.  Each individual member of each successive generation of citizens must come to regard
herself and her fellow citizens as mutually bound to support only those social and political
institutions each is prepared sincerely to justify to others from a shared or common moral point
of view.  The foregoing requires, of course, that each successive generation of citizens acquire
the ability to identify and then reason from a shared moral point of view, or a point of view no
reasonable person could reasonably reject.  But what is the content of this point of view?

 Matters would be simple were it the case that the content of this point of view could be
read off of some universal moral consensus among citizens.  But such is not the case.  Beyond
claims regarding the minimum institutional requirements of any just state and a few additional
noncontroversial moral claims, citizens are likely to disagree (and will for any foreseeable future
disagree) over what is, or ought to be, the content of their shared moral point of view from which
any justificatory undertaking must proceed.  Because of such disagreement, it may seem that the
demands of reciprocity either cannot be met (if there is no shared moral point of view) or are
realizable in only a narrow range of cases (if there is only a very thin shared moral point of view).
The temptation, then, is to set the ideal of justice as reciprocity aside as yet another utopian
political pipe-dream.  This is a temptation to which we need and ought not yield.

 Some disagreements, call them simple disagreements, arise solely out of failures of logic
or inference, dogmatic or ideological refusals to examine evidence or arguments, or the unhappy
influence of irrational bias,  prejudice or naked self-interest.  Simple disagreements, inevitable as
they are, constitute but temporary and surmountable hurdles on the path to a shared moral point
of view from which to justify our conduct and social institutions to one another.  Of course, to
clear the hurdle of simple disagreement, people need good critical reasoning skills.  But provided
they possess and exercise such skills, they will find themselves collectively able in due course to
overcome simple disagreements as they arise.  From the fact that persons often are divided by
simple disagreement when it comes to the shared moral point of view from which basic social
relations are to be justified, it does not follow that justice as reciprocity demands the impossible.
What justice as reciprocity demands is the development and exercise of the sort of critical
reasoning skills needed to identify and overcome simple disagreements.

 Other disagreements, call them reasonable disagreements, arise even among highly
intelligent persons of manifest good will.  Like simple disagreements they are inevitable.  But
unlike simple disagreements they are ineliminable through the free, collective exercise of human
reason.  No two persons ever have the same life experiences; often it is very difficult to
determine which of many competing conceptions of an abstract concept is best; often the right
resolution of an issue turns on a balance of reasons so close as to preclude certainty in judgment.
These are among the irreproachable sources of disagreement likely to challenge every attempt to
articulate a shared moral point of view from which to justify our conduct to one another.   But
what follows from the fact that persons are often divided by reasonable disagreement when it
comes to the shared moral point of view from which basic social relations are to be justified?

 That justice as reciprocity expresses a utopian political pipe-dream would follow only if
no shared moral point of view capable of justifying determinate conclusions regarding social
relations could be articulated independent of all reasonable disagreements.  Whether or not this is
so cannot be known a priori.  In many areas of human inquiry (e.g., much of science), reasonable
disagreements can be accommodated and a shared point of view (neutral with respect to the
reasonable disagreement) still articulated from which to justify the resolution of one or another
issue.  Whether this will prove true with respect to issues of justice is an open question.  But it is
a question the answer to which cannot be known in advance of attempting to resolve issues of
justice, consistent with the demands of reciprocity, notwithstanding various reasonable
disagreements (at levels of both concrete substantive claims and the more general claims
constitutive of a shared moral point of view).  Here it is worth remembering that if one or another
approach to democratic decision-making may be justified from, and thereby included within, a
shared moral point of view as a reasonable way to resolve reasonable disagreements over
particular substantive issues, then it may be possible to reason (collectively, through democratic
voting) from such a shared moral point of view to a determinate resolution of many if not all
substantive issues over which citizens reasonably disagree.

 It is, of course, logically possible that once all reasonable disagreements have been
accommodated, it will prove impossible to articulate the shared moral point of view presupposed
by the ideal of justice as reciprocity, or at least impossible to articulate anything more than a
shared but greatly impoverished point of view capable of doing no real justificatory work.  But
the jury is still out on this question, and while it is, it is premature to conclude that the ideal of
justice as reciprocity demands the impossible.  For the foreseeable future, justice as reciprocity
requires persons to do their best to justify basic social and political institutions to one another
from a shared moral point of view that accommodates, or is neutral with respect to, reasonable
disagreements.

 To realize justice as reciprocity, persons must be inclined and able, first, to distinguish
simple from reasonable disagreements, second, collectively to employ the requisite critical
reasoning skills to resolve simple disagreements, and, third, to undertake to articulate and reason
from a shared moral point of view that accommodates or is neutral with respect to reasonable
disagreements.  A just state, then, presupposes a citizenry generally possessed of the capacities,
skills, knowledge, habits and dispositions – in short the virtue or virtues – necessary to both the
constant identification and elimination of simple disagreements and the constant identification
and accommodation of reasonable disagreements.  Citizens must be inclined and able to look
critically, imaginatively and sympathetically at the commitments and claims of both others and
themselves, all with an eye toward honoring the ideal of reciprocity.

 Absent such a citizenry and public political culture, there is little hope for securing and
preserving the minimum essentials of a just state, let alone securing social justice more fully and
adequately.  Victims of injustice will find themselves powerless to attack as unjust policies or
institutions either wrongly endorsed by others purely out of self-interest or wrongly believed or
claimed by others to be justified from a shared moral point of view.  They will lack the
knowledge, skills and inclinations necessary either to demand that the defenders justify
themselves from a shared moral point of view, or to demonstrate publicly either that defenders
are simply wrong about what follows from a genuinely shared moral point of view or that the
moral point of view from which they reason is non-neutral with respect to one or another
reasonable disagreement and thus not shared.

 The capacities, skills, knowledge, habits and dispositions demanded by justice as
reciprocity are both moral and substantial.  They include the capacity for and skills necessary to
sympathetic identification with others, knowledge of a wide range of moral ideals and modes of
moral reasoning, an understanding of the structure of various common moral disagreements, the
habits of patience and critical self-examination, and a disposition to act in social life from moral
reasons.  Taken together, the capacities, skills, knowledge, habits and dispositions – the virtue or
virtues – necessary to justice as reciprocity render untenable the often invoked distinction
between moral education and civic education or education for citizenship more narrowly
construed.

 One need not appeal to more than the demands justice as reciprocity place on citizens
with respect to fundamental political issues to justify the claim that citizens must come to possess
the foregoing capacities, skills, knowledge, habits and dispositions.  But once these are acquired
it is hard to see how they could be limited purely to claims regarding fundamental matters of
political justice, the design of basic social and political institutions, especially those through
which coercive political force is exercised.  First, it is likely impossible to disentangle political
commitments and claims from the wider categories of theoretical or doctrinal moral, religious,
philosophical and scientific commitments and claims.  The latter inevitably shape the former for
each and every person.  And even if this weren’t true, there still will always be some political
issues that will prove indeterminate from any moral point of view, shared or otherwise, that does
not draw more broadly from moral, religious, philosophical, and scientific theory or doctrine in
fairly robust ways.  So, honoring the ideal of reciprocity with respect to fundamental political
issues will inevitably involve bringing the sort of critical, self-reflective eye justice as reciprocity
demands to a wide range of moral, religious and philosophical claims and disagreements.
Disagreements over these sorts of claims are sometimes simple and sometimes reasonable.  So,
citizens must be inclined and able to think critically, imaginatively and sympathetically about
them, with the aim of distinguishing simple from reasonable disagreements and treating each
accordingly.  There is no avoiding the task of examining in this spirit disagreements over, for
example, the truth or rationality of theism, the moral status of a fetus, or the morality of
homosexuality.

 Another reason why the distinction between general moral education and civic education
or education for citizenship collapses once we attend to the virtue or virtues essential to justice as
reciprocity is that a commitment to justice as reciprocity is a commitment to insure that the
demands of justice are met by all basic social or political relations, forces, institutions or forms of
authority into which persons are born and within which they live their lives, not just by the
institutions through which coercive political force is most obviously exercised.  Justice as
reciprocity governs the institutional nature and structure of the family, the institutional forces
shaping gender roles, racial identities, and preference and consumption patterns, and much else
besides.  While citizens may be able publicly to debate and resolve basic issues of tax policy, for
example, without being inclined or able critically, sympathetically and imaginatively to examine
a wide range of doctrinal and theoretical moral, religious, philosophical and scientific claims and
disagreements with an eye toward honoring the moral demands of reciprocity, it is hard to see
how, consistent with those demands, they could debate and resolve issues regarding the justice of
the traditional family or traditional gender roles without subjecting many such claims and
disagreements to careful, critical, sympathetic and imaginative examination.

 Justice as reciprocity requires a citizenry inclined and able to examine, in a certain spirit
and with a certain moral end in mind, the full range of their beliefs and values bearing on social
life – including doctrinal or theoretical moral, religious, philosophical and scientific beliefs.
From the point of view of justice as reciprocity it is necessary but not sufficient for citizens to
live up to the demands of reciprocity with respect only to the more obvious cases of coercive
state action.  Liberals are right to insist that they must do this, but feminists and other liberal
critics are right to insist that they must also live up to the demands of reciprocity with respect to
the structure of the family, the formation of gender and racial identities, the nature of preference
and consumption patterns and the like.

 And while conservatives are right to argue that law-abidingness, patriotic loyalty, self-
reliance and constitutional competence are political virtues, they are wrong to overlook or to
allow for only a truncated version of the more basic virtue or virtues of reasonableness or
reciprocity.  The virtue or virtues of reasonableness or reciprocity are analytically and
normatively prior to those of law-abidingness, patriotic loyalty and the like.  Indeed, the proper
limits of the latter can be articulated only through appeal to the former.  Moreover, there are no
good reasons for conservatives to deny that the virtue or virtues of reasonableness or reciprocity
must include the capacity and disposition to examine critically, sympathetically and
imaginatively one’s own beliefs and the beliefs of others across a very wide range of human
thought – from politics to morality to philosophy to religion to science.

 Many conservatives resist the claim that justice requires the virtue or virtues of
reciprocity just described.  Often they argue that social institutions aimed at cultivating in
citizens the virtue or virtues of justice as reciprocity will prove unduly corrosive to certain
traditional beliefs or ways of life and thus to the psychological integrity of certain persons.  This
claim is, I think, an overstatement.  But I shall not argue the point here.  Instead, what I want to
argue here is that these same conservatives typically tacitly appeal (as, of course, do liberals) to
the ideal of justice as reciprocity when it comes to public arguments over the demands of justice.
They are thus not in a position to resist the ideal of justice as reciprocity.  To take recent
examples close to the concerns of this paper, Stephen Gilles, a conservative critic of much that is
said in contemporary debates over compulsory education, argues that the disagreement between
theists and atheists (as well as the disagreement between feminists and traditionalists and any
number of other socially prominent disagreements) is not a simple disagreement rooted in faulty
logic or obvious and reproachable bias on one side or the other, but is rather a reasonable
disagreement.  As such, he argues, it must be accommodated in public political discourse in ways
a simple disagreement would not have to be.  So, when it comes to both constitutional and
statutory law defining compulsory education and the scope of parental rights, on Gilles’s view,
the arguments made and justifications tendered must be neutral as between theist or atheist (or
feminist or traditionalist, etc.) presuppositions.

 Gilles is no doubt right about this (at least with respect to the disagreement between
theists and atheists).  But the point is that in making this argument, he implicitly affirms the ideal
of justice as reciprocity, the distinction between simple and reasonable disagreements, and the
moral demand that the latter be accommodated when it comes time to justify ourselves to others.
Having affirmed all this, he cannot avoid drawing the conclusion that one way or another, a just
state must insure that each successive generation of citizens come to acquire and make use of the
capacities, skills, knowledge, habits and dispositions necessary to justice as reciprocity.

 Michael McConnell, another conservative critic of much that is said in contemporary
debates over compulsory education, is vulnerable on the same front.  He argues that because
there is reasonable (rather than simple) disagreement over what sort of moral content ought to be
taught in any compulsory moral curriculum (e.g., the morality of homosexuality, etc.), the state
must stay out of the moral education business and leave that task to parents and churches.  But
McConnell’s argument presupposes a citizenry committed to the virtue of reciprocity, and so
absent an argument showing that parents and churches can be relied on to cultivate such a
citizenry, McConnell’s conclusion remains unjustified.  Further, his claim that there is reasonable
disagreement over the sort of moral content that ought to be included in any compulsory moral
curriculum is overstated insofar as he presupposes that it extends to and includes reasonable
disagreement over the need to include the virtue or virtues of reciprocity in such a curriculum
(whether taught by parents and churches, or by schools).

 Those who argue for the teaching of creationism generally fall prey to the same trap.
There is, they argue, a reasonable rather than a simple disagreement over the truth of
evolutionary theory.  Thus, arguments over compulsory science curricula ought to be neutral with
respect to the disagreement between creationists and evolutionary theorists.  Like Gilles’s and
McConnell’s arguments, these arguments implicitly affirm the multiple lines of mutual
dependency between justice, reciprocity, and the capacity and inclination both to distinguish
reasonable from simple disagreements and to accommodate the former in ways different from the
latter.  From this implicit affirmation it follows that justice requires more than the sort of limited
civic education for law-abidingness, patriotic loyalty, self-reliance, and constitutional
competence conservatives are willing to accept.  It requires citizens ready and able to examine
critically, sympathetically and imaginatively a wide range of deep commitments – political,
moral, religious, philosophic and scientific – whether held by themselves or others, and to do so
with the aim of honoring the moral demands of reciprocity.

III.  Compulsory Moral Education and the State.

 What remains to be shown is that not only is it unreasonable to suppose that simply
growing up, or growing up in a constitutional and democratic republic, will lead most citizens to
acquire the virtue or virtues of reciprocity in due course, but that it is also unreasonable to assign
the task of morally so educating citizens primarily to families, churches or other non-state civic
associations.  To make these showings would be to make a prima facie case for assigning the
state the task of insuring that all citizens receive the required moral education.  What would
remain, then, would be to rebut objections to this prima facie case.

 The proper aim of compulsory moral education must be to produce citizens inclined and
able to think in a certain way about the design and operation of social and political institutions:
in the spirit of fidelity to the ideal of justice as reciprocity, with a readiness to examine all
relevant arguments and claims, both their own and others, critically, imaginatively, respectfully,
and nondogmatically; always searching for the truth and ready to identify and resolve simple
disagreements, but always also hesitant ever to claim for themselves more than a provisional or
tentative truth; always searching for a shared moral point of view from which to justify mutually
acceptable principles to govern social life, but also always ready to revise or expand that point of
view to accommodate reasonable disagreements with others.

 Now, it is evident that persons do not naturally or inevitably come to acquire such
capacities, dispositions and the like as simple matter of biological development.  Nor do they
come to acquire them naturally or inevitably under a wide range of social and political
institutional arrangements.  The question, then, is whether there is reason to think they will
acquire them naturally or inevitably growing up within a constitutional and democratic republic
that takes no concrete and purposeful institutional steps to cultivate them through moral
education.

 Suppose a minimally just constitutional and democratic republic.  Suppose further what
was assumed earlier, namely that a necessary institutional feature of any such society will be
some form of state-sponsored and state-regulated compulsory non-moral education for literacy,
numeracy, facility with various technologies and the like.  Must such a society also include as a
matter of institutional design a system of compulsory moral education?

 If the answer to this question is no, then it must be the case that under normal
circumstances citizens will be adequately socialized into the virtues of reciprocity through their
experience of growing up and living within such a society.  But what socializing forces will
citizens be exposed to simply by growing up and living within such a society?  They will be
exposed, first, to public political institutions (the legislature, executive, judiciary, agencies, etc.),
second, to the public political culture to which such institutions give rise (political parties and
conventions, newspaper editorials, public debates, etc.), third, to the many voluntary associations
constitutive of civil society and not part of the public political culture (families, churches, art
societies, businesses, clubs, etc.), and finally to a regime of compulsory but non-moral education
(for literacy, numeracy, and perhaps even basic constitutional competence).

 Now, of these various socializing forces, the first two – public political institutions and
the public political culture to which they give rise – must be heavily discounted since children
will be largely unaffected by them, and they will have an effect only on those young adults
already predisposed to participate in them, namely those already committed to the notion that
basic social and political issues ought to be resolved democratically consistent with the moral
demands of reciprocity.  And the last – education for literacy, numeracy, technical knowledge,
perhaps even constitutional competence – must also be heavily discounted since there is no
reason to think that by itself it will significantly contribute to the development of the virtue or
virtues of reciprocity in particular.  Thus, if it is plausible to suppose that successive generations
of citizens will acquire the virtue or virtues of reciprocity merely by growing up and living within
a minimally just constitutional and democratic republic, it must be because  growing up and
participating in families, churches and various voluntary associations will bring most members of
each generation to a point where they are so inclined and able to honor the demands of
reciprocity that as young adults they will be predisposed to participate in public political
institutions and the public political culture to which they give rise, and thereby to develop more
fully and adequately the virtue or virtues of reciprocity.

 It is within the family, of course, that children receive their first lessons in the morality of
reciprocity.  They learn to distinguish diverse roles and responsibilities and imaginatively to take
up points of view other than their own.  And they learn to distinguish their own good from the
common good.  But for many reasons the family is not well-suited to providing children the
necessary lessons in the morality of reciprocity.  And it is hard to see how any alteration that did
not turn it into a fundamentally different form of human association could enable it to
accomplish this end.

 Children properly look to parents for (and parents properly offer their children) a point of
entry into the world, a set of values, beliefs, explanations, and stories capable of orienting and
giving direction to a young life.  And parents generally if not universally take great satisfaction in
passing along much of what they value, believe and the like to their children.  There is nothing
objectionable in all this; indeed, these are valuable aspects of the parent-child relationship and of
family life more generally.  But they are aspects that suggest the parent-child relationship or
family life more generally is unlikely by itself to provide a context conducive to developing in
children to the requisite degree the capacities, skills, knowledge, habits and dispositions essential
to justice as reciprocity.  It is not plausible to suppose, without demanding fundamental changes
in the nature of the family, that parents will regularly train their children to examine critically the
very values, beliefs, explanations and stories they provide their children as a point of entry into
the world.

 Moreover, families are rarely egalitarian democracies.  They are rather closer to what
Rawls, referring to illiberal but nevertheless decent states, calls “consultative hierarchies.”
While parents are obligated to make decisions which advance what they regard as the common
good of the family, and to consult with and take into account the views of their children when so
doing, they are not obligated to be able to defend their decisions from a moral point of view
neutral with respect to any reasonable disagreements they may have with their children.  As full
and fully developed moral equals, parents are obligated to be able to so defend their decisions to
each other in such a manner.  And this will surely have an impact on the moral development of
children.  But it does not change the fact that the parent/child relationship is fundamentally
different from the parent/parent relationship.

 Finally, it is not an altogether simple matter to learn, let alone to teach others how, to
distinguish simple from reasonable disagreements and to accommodate the latter when
undertaking to offer others a moral justification for some action or proposal.  Indeed, many
parents are unlikely to be able to teach their children these matters without some significant
training in how to proceed.  Thus, even if parents could in principle provide their children with
the sort of moral education demanded by justice as reciprocity, there are good reasons of
efficiency to explore alternative institutional mechanisms for achieving the same result.

 The only available alternative, other than to rely on a state-sponsored and state-regulated
system of compulsory moral education, is to assign the task of compulsory moral education to
civil society, or more particularly, to the various voluntary associations constitutive of civil
society.  It is, of course, true that membership and participation in voluntary civil associations
very often tends to encourage the development and deployment of many of the capacities,
dispositions and the like essential to justice as reciprocity.  It is also true that in a diverse civil
society most persons will experience membership in at least a few voluntary associations prior to
reaching young adulthood.  It is even true that the nature of the association is often only
marginally relevant to whether membership and participation tends to cultivate the virtue or
virtues of reciprocity.  What is more relevant, generally, is whether members regard themselves
as engaged in a voluntary, collective and cooperative endeavor on terms all must accept.

 All this is true, but it is not enough to make the case for leaving the task of compulsory
moral education to the voluntary associations of civil society.  In any free society many voluntary
associations will be dogmatically and ideologically committed to claims with which there can be
no reasonable disagreement, or to a general withdrawal from political life or earthly affairs.
Those who primarily belong to and participate in these associations are not likely to be brought
closer to the virtue or virtues of justice as reciprocity through their experiences in them.

 If neither families nor voluntary associations can be plausibly relied on to insure
successive generations of citizens inclined and able to honor the ideal of reciprocity, then it is not
unreasonable to assume that the same is true for families and voluntary associations working
together.  If this is right, then, a  prima facie case exists for assigning the state the task of
compulsory moral education.  Assigning the state such a task, however, carries with it grave
risks.  Specifically, any system of state-sponsored and state-regulated compulsory moral
education may appear to threaten liberty of conscience (by constituting a sort of brainwashing by
the state), to disrupt social stability (by forcing particular social groups to undergo an experience
they would just as soon forego), and to eliminate social diversity (by leading all citizens to the
same moral outlook).   But these risks ought not be overestimated.

 The sort of compulsory moral education defended here poses no threat to liberty of
conscience.  The right to liberty of conscience is not a right to a life immune from experiences
that might cause one to rethink or critically examine fundamental commitments.  It is a right to
be free from a confessional state, a right to draw and affirm one’s own conclusions with respect
to life’s most pressing questions.  And the compulsory moral education defended here poses no
threat to this right.  It is true, of course, that it is inconsistent with the belief that basic social and
political institutions and forms of authority need not be justified to those subject to them, or at
least from premises no reasonable person could reject.  But this is a matter about which a just
state need and ought not be neutral.  Furthermore, it does not follow from a commitment to
justice as reciprocity that those who come to reject it, their moral education notwithstanding,
ought to be forced to confess (although they may be forced to act) otherwise.

 The second alleged risk of empowering the state to undertake compulsory moral
education is that it will exacerbate social conflict rather than limit it.  Diverse social groups will
compete for the power of the state so as to use compulsory moral education as an instrument for
establishing their preferred doctrine as officially orthodox.  Furthermore, since there is no way
for any compulsory curriculum to give equal time to all points of view, some groups will always
feel as if the curriculum treats their beliefs unfairly, regardless of whether the power of the state
has in fact been captured by any particular group.  Given time and other constraints, it is not
possible to design a curriculum that all will perceive as fair.  Better, the argument goes, to
exclude moral education from any system of state-run or state-regulated compulsory education.

 But the risk of civil strife here is less pronounced than it might otherwise seem, since the
moral education required here is not an indoctrination into any particular moral doctrine, but
rather into a way of thinking about and resolving publicly a certain range of moral issues, namely
those concerning social and political justice.   Thus, so long as the goal of compulsory moral
education is widely understood, affirmed and honored, competing social groups will have little to
gain by gaining the power of the state with respect to educational institutions, and little to
complain of if their particular substantive moral, religious or philosophical doctrine receives less
attention within the curriculum than other doctrines.

 Still, insofar as the state is in the compulsory moral education business, social groups
may be tempted to try to seize state power over compulsory moral education so as to use it to
their own advantage.  And where there is temptation, sin is sure to follow, at least among
humans.  Thus, any system of state-run or state-regulated compulsory moral education is likely to
generate some social strife or division.  The question is whether it is likely to generate more strife
than would leaving moral education fully to families or civil society.  In the latter circumstance
there is no reason to expect widespread success at cultivating among successive generations the
capacities and inclinations essential to justice as reciprocity.  And where these capacities and
inclinations are not widespread and deeply rooted, there can be little hope that any state will
remain even minimally just for the long term.  Thus, while there are reasons to be cautious with
respect to a state-sponsored system of compulsory moral education, there are greater reasons to
move forward.

 Finally, the third risk alleged to attend state-sponsored compulsory moral education is the
homogenization of civil society.  Social diversity constitutes a bulwark against tyranny as well as
a seed bed of public virtue.  And as the collective expression of the free development of human
potential, it is also, arguably, a great collective good.  Thus, social institutions corrosive of social
diversity ought to be looked upon with a wary eye.  But what reasons are there for thinking that a
compulsory moral education into the virtue or virtues of reciprocity will prove corrosive of social
diversity?

 There is little reason to think that the sort of compulsory moral education defended here
will in any harmful or dangerous way homogenize civil society or reduce social diversity.  It
poses a threat only to dogmatic, ideological groups unwilling critically to reflect with fellow
citizens on terms of mutual respect when it comes to beliefs and commitments fundamental to
social life.  And these groups are neither necessary to nor supportive of a social diversity capable
of guarding against tyranny or cultivating public virtue in civil society.

 One might argue that a compulsory moral education into the virtue or virtues of
reciprocity will lead to a citizenry without any firm moral, religious, or philosophical
commitments, and that such a citizenry, given the institutional backdrop of a minimally just state,
will inevitably subscribe to materialist and market values alone.  But this argument simply
presupposes that the virtue or virtues of reciprocity are not compatible with firm, foundational
commitments.  Of course, the presupposition may be true, but if so that must be demonstrated
and demonstrated empirically.  Here it is not enough simply to cite the rise of materialist and
market values in the United States and other liberal democracies, since there is no reason to think
the cause of that rise the widespread realization among citizens of the virtue or virtues of justice
as reciprocity.

 There have been in the United States periods of state-sponsored and state-regulated
compulsory moral education wrongly aimed at an objectionable assimilationist goal.  During the
early decades of the 20th century in the United States, educational institutions were deployed to
advance an overtly anti-Catholic and anti-immigrant social policy aimed at bringing diverse
citizens to a consensus over a substantive and deeply Protestant moral orientation.  Early 20th
century educational policy was in fact socially divisive and, where successful, corrosive of social
diversity.  And it contributed significantly to the growth of today’s system of separate Catholic
and private schools.

 In more recent years there has been a movement, partly in response to the perceived
failures of the “values clarification” agenda of the 70's and early 80's, to reintroduce a moral
component to the compulsory curriculum.  And this movement has itself often proved socially
divisive.  In general, two sorts of initiatives have been born of this movement.  Civic virtue
initiatives aim at reintroducing or revitalizing civic education, usually understood as aiming at
the cultivation of the virtues of law-abidingness, patriotic loyalty, constitutional competence, and
sometimes social tolerance and civic or political participation.  Common moral values initiatives
aim at reintroducing what is sometimes called traditional moral education, usually understood as
aiming at generating a shared commitment to such common moral values or principles as
honesty, respect for nature, non-violence, charity and the like.

 Civic virtue initiatives have proved divisive primarily when they include an education
into the virtue or virtues of civic and social tolerance but not the ideal of justice as reciprocity.
Where civic and social tolerance is taught as the limited political virtue of sometimes refraining
from actively seeking to suppress activities that one abhors, many parents rightly object that
justice and stability in any minimally just state demands more.  Yet where it is taught as anything
more than such a limited political virtue, but without any connection to the ideal of justice as
reciprocity, many parents committed to one or another traditionalist doctrine or view object.  The
trouble here is that the virtue or virtues of civic and social tolerance cannot effectively be taught
in a just state apart from the ideal of justice as reciprocity, for it is the latter that grounds and
determines the proper limits of the former.

 Common moral values initiatives have proved socially divisive primarily because parents
very often reasonably disagree over the meaning or application of common moral values and
principles in concrete cases.  And it is very difficult to teach common moral values and principles
to students without discussing their meaning or application in concrete cases.  Respect for the
dignity of human life is no doubt a common moral value.  But what it means for concrete cases
of euthanasia or abortion is very often an open question over which parents will quite often
disagree.  Because the common moral values initiative is rarely grounded in and framed by the
ideal of justice as reciprocity, it inevitably falters at the point of such disagreements.  Teachers
find themselves forced either to treat all disagreements as reasonable disagreements and to speak
solely in terms of vague platitudes or to treat no disagreements as reasonable disagreements and
to lead students to concrete determinate resolutions of particular issues (resolutions their parents
will often reasonably reject).  Again, the trouble here is that common moral values cannot be
cultivated publicly in a just state apart from the ideal of reciprocity, for it is the latter that
grounds and determines the limits of the former.

 A compulsory moral education aimed primarily at cultivating the virtue or virtues of
reciprocity would likely generate far less resistance from parents than current moral education
initiatives.  Teachers would aim to cultivate in their students the inclination and capacity to
discuss important moral, religious, philosophical and political beliefs in a critical, sympathetic
and imaginative way, with an eye toward determining what can be justified on terms no
reasonable person could reject.  Discussions of common moral values, including social and civic
tolerance in one or another form, would be undertaken in this spirit, so that students determine
for themselves together the content and limits of their shared moral point of view.  Specific
moral issues, for example, the morality of homosexuality, would be taken up in the same spirit.

 Of course, the result for some students will be the discovery that their views on, say, the
morality of homosexuality are simply mistaken.  But if it turns out that the morality of
homosexuality is something over which there can be only simple disagreement, then there is no
loss or harm in those who hold demonstrably false or unreasonable views being brought to see
the error of their ways.  And if it turns out that the morality of homosexuality is a matter over
which persons can reasonably disagree, then the result for all will be a deeper understanding of
their own views and a more complete ability publicly to defend them.

 To be sure, the sort of compulsory moral education defended here carries with it certain
risks.  Perhaps the most significant is that one will discover that from a shared public moral point
of view one’s views on some important matter, such as the morality of homosexuality for
example, are simply wrong.  In light of this risk, one might expect citizens to be able to endorse
the ideal of justice as reciprocity in only a partial or incomplete way.  For example, one might
expect certain conservatives to refuse to embrace the ideal as applied to the morality of
homosexuality, since they cannot possibly be wrong in their views about this matter.  But such a
refusal, while psychologically understandable, would be logically problematic.  Neither
conservatives nor liberals can know in advance which if any of their views will be shown to be
fully indefensible from a shared public moral point of view.  And neither is in a position publicly
to demonstrate that certain issues of public morality are simply beyond the reach of the ideal of
justice as reciprocity.  Both run the same risks of discovering that on certain important matters
they are simply in error.  Thus, neither conservatives nor liberals can logically take the view that
certain issues are simply beyond the reach of justice as reciprocity.  As a matter of logic there can
be no half-hearted or partial affirmation of the ideal.

IV.  The Rights of Parents.

 Suppose the foregoing constitutes an adequate case for assigning the state a central role in
distributing the sort of compulsory moral education essential to justice as reciprocity.  Do not
parents have a right, indeed a prior right, to shield their children from such an education if they
think it in the best interests of their children to do so?  After all, isn’t moral education first and
foremost the responsibility of parents?  Parentalist objectors to compulsory moral education (or
compulsory education generally) answer both of these questions in the affirmative.

 Courts in the United States and elsewhere have generally affirmed the existence of a
fundamental right on the part of parents to control or direct the education of their children, and
such a right is affirmed also by international human rights documents.  Yet, courts in the United
States and elsewhere have said little about the foundation or justification of this right.  Some
recent commentators have argued that courts have often reached erroneous results in particular
cases because they have largely misunderstood the nature and grounds of the right of parents to
determine the education of their children.  Courts in the United States and elsewhere have also
said little about whether the rights of parents trump the rights of children.  Indeed, they have in
particular cases typically asked only whether there is a compelling state interest sufficient to
justify limiting the rights of the parents seeking to control the education of their children.  The
rights of the children, if any, are not discussed at all.  Again, some recent commentators have
argued that courts have often reached erroneous results in particular cases because they have
failed to attend in any way to the rights of the child or children at issue.  These latter rights, they
argue, will at least sometimes outweigh any rights asserted by the parents.

 Now, if the argument of the previous section of this paper is correct, then as a matter of
political morality, if not constitutional law, it is not hard to see that children have a right not only
to an education, but to a compulsory education with a significant moral component aimed at
cultivating the virtue or virtues of reciprocity.  A minimally just state is not possible in the
absence of some such system of compulsory education, and there are compelling reasons to
assign the state the task, or duty, of providing it to each successive generation of citizens.  To say
that the state is under a duty to provide citizens generally with such an education is just to say
that individual citizens have a fundamental claim against the state for, or a right to, it.  While
this result does not square with existing federal constitutional law in the United States, it is
consistent with both constitutional and statutory law at the state level in the United States, with
international human rights law, and with the views of many contemporary commentators.

 Suppose, then, for the sake of the argument, that children have a fundamental right to an
education that satisfies both the non-moral and moral curricular requirements imposed by justice.
Recognizing this right would, in itself, fundamentally change the structure of judicial resolutions
in cases in which parents assert their own rights as grounds for immunizing their children from
one or another aspect of compulsory education.  Instead of simply asking whether or not the
asserted parental rights are fundamental, and if so whether or not the aspect of compulsory
education objected to advances a compelling state interest in a manner least restrictive of the
asserted right, courts would have also to ask whether granting the parental request would infringe
on the right of the child or children in question to an appropriate education, and if so, whether
that right might outweigh the right or rights asserted by the parents as the ground for their
request.  Thus, even if parents can show that one or another aspect of compulsory education
infringes upon their fundamental rights without advancing a compelling state interest, or without
advancing such an interest in the manner least restrictive of the infringed upon right, it might still
be the case that courts should refuse to honor parental requests to immunize a child or children
from the relevant aspect of compulsory education because the child has or children have a right
to that aspect of compulsory education sufficient to outweigh the infringed upon parental right.

 To resolve a conflict of rights through balancing, it is necessary first to understand the
content and weight of the relevant rights.  Presumably, the right of each child is to a publicly
funded and appropriate education, both non-moral and moral, sufficient to meet the demands of
justice.  And presumably, this is a very weighty right.  To say this is to say more than that such an
education is in the best interests of each child.  Such an education is that, of course; but it is also
much more than that.  The trouble with treating the right to an appropriate compulsory education
as primarily a matter of determining and advancing the best interests of particular children is that
it creates the false impression that everything turns on determining who is best suited to judge the
best interests of the children in question: the parents or state officials.  The right to an appropriate
compulsory education is better regarded as an enforceable claim, possessed by each child
regardless of his or her particular identity, to a particular sort of education which the state, or
citizens collectively, have a duty to provide.  Of course, some cases may indeed genuinely raise
the issue of whether or not the best interests of some particular child or children argue in favor of
an exemption from one or another aspect of compulsory education.  And these cases will raise
also the issue of who is best suited to determining the child’s or children’s best interests.  But
most cases will not genuinely raise these issues.  Most cases will genuinely raise only the issue of
whether the rights of a particular child or children ought to yield to the rights of his or her
parents, or vice versa.  And there is no doubt as who is best suited to determining this sort of
issue: judges and courts.

 Parents have a fundamental right to religious liberty.  But it is hard to see how any aspect
of the familiar compulsory education system could infringe upon, let alone violate, this right.
(To be sure, it is easy to imagine how a compulsory education regime that might infringe upon or
violate the rights of students to religious liberty.)  The right to religious liberty is a right to non-
interference with respect to one’s own religious beliefs, and a wide (but not unlimited) range of
religious practices.  It is not a right to determine the religious beliefs of others, not even one’s
own children.  Of course, we do allow parents to force or forbid their children, upon pain of
punishment, to attend church, synagogue, mosque or Sunday school and the like.  And we do so
rightly.  But it does not follow that we do so because we think it a matter of respecting their (the
parents’) right to religious liberty.  Indeed, we do so because, or at least in part because, we think
an intimate and secure parent-child relationship so important to the full and adequate
development of children as citizens and persons that unless there are compelling reasons of
justice to interfere with that relationship, the assumption is that parents are free to advance the
best interests of their children as they see fit.  In most familiar cases of parentally required or
forbidden religious training and experience, there are no clear and compelling reasons of justice
to interfere with the parent-child relationship – no clear threat to the full and adequate
development of the children as citizens or persons.  Thus, even apart from any considerations of
religious liberty, there are good reasons in such cases for courts properly to uphold the freedom
of parents to require or forbid religious training or experience as they see fit.

 If parents have a fundamental right at stake when they object to one or another aspect of
the compulsory education of their child or children, it must be because they have a right, of one
sort or another, to a parent-child relationship free of certain sorts of state interference (rather than
a right to religious liberty that includes determining the religious beliefs of their children).
Now, there are three possibilities here.  The first is to deny that parents have any such right at all.
Of course, such a claim is simply not plausible.  It runs contrary to well-established principles of
American constitutional law as well as a widespread, enduring and plausible American consensus
over the primacy of the family both within social life and with respect to the full and adequate
moral development of persons as citizens.

 The second possibility is to allow that parents have a right to a parent-child relationship
free of certain sorts of state interference, but to argue that this right is derivative, instrumental
and indirect, affirmed only because as a matter of general social policy affording parents such a
right best secures the rights and advances the interests of children generally.  On this account,
parents enjoy a right to advance the best interests of their children as they see fit – to assert the
interests of their children as a proxy, so to speak – not because they necessarily best know their
children’s best interests in any particular matter, but because affording them this right insures
that their children generally will enjoy the sort of spontaneous, intimate and caring parent-child
relationship necessary to their full and adequate development as both citizens and persons and
thus in their best interests generally.

 If this is the right asserted by parents who object to one or another aspect of a compulsory
education regime, then courts ought to accommodate parental objections only where the harm
caused to the child by enforcing the compulsory education regime (the harm of a parent-child
relationship governed, at least in part, by the external coercive force of the state) is greater than
the harm caused to the child by upholding the parental objection (the harm of being denied one or
another aspect of the compulsory education to which he or she is entitled).  While courts
committed to working from this account may frame their deliberations over parental objections to
compulsory education in terms of the rights of parents and the rights of children, what they are
really singly aiming at is a resolution of the issue that minimizes the harm done to the child or
children in question.

 The third possibility is to allow that parents have a non-derivative, non-instrumental,
direct right to a parent-child relationship free of certain sorts of state interference.  Even if
parents have a right of this sort, of course, the state might still justifiably interfere with the
parent-child relationship.  It might do so, indeed ought to do so, when parents act within the
parent-child relationship in ways that conflict with weightier rights of others, including the rights
of their children.  So, if parents have a direct, non-derivative, non-instrumental right to be free
from various sorts of state interference with the parent-child relationship, then when parents
object to one or another aspect of a compulsory education regime, courts must weigh and balance
the conflicting rights.  This will prove a complex undertaking, requiring courts to determine not
only whether one right is more fundamental than the other from the point of view of justice, but
also how deeply vindicating either right cuts to the core of the other.   This task is not analytically
reducible to trying simply to determine how best to minimize harm to the child or children in
question.  It is a more complex task involving the weighing and balancing of rights.

 By treating the rights of parents as indirect, derivative and instrumental, the second
account above leaves courts with a more manageable and less complex task when it comes to
adjudicating parental objections to compulsory education.  And there is no doubt that simplicity
is a virtue when it comes to adjudication.  Moreover, there is something politically appealing
about making the best interests of children the north star by which courts should ultimately
navigate in their reasoning.  Thus, it is perhaps not surprising that many commentators have
come to favor the second account.  Nevertheless, the third is the better account.

 Suppose the second account is the right account.  On this account parents have no self-
standing, independent interest in a parent-child relationship free from state interference, or at
least no such interest sufficient to ground a direct, non-instrumental, non-derivative right to such
a relationship.  But if this were true, then there would be no reason for thinking it unjust or
wrong for the state to redistribute children shortly after birth, taking them away from their natural
parents and giving them to others judicially determined more suitable to parent the children in
question.  If such redistributions were made only very early in life, and generally worked to put
children into more competent or better-suited parental hands, then no significant harm would
befall the child.  He or she would still grow up within a spontaneous, secure and intimate parent-
child relationship free from various sorts of state interference.  And he or she would have the
advantage of parents more talented at or inclined toward parenting than those to whom he or she
was naturally born.

 Of course, the parents to whom he or she was naturally born might object to the court’s
determination that others were more competent or better-suited to serve as parents.  But surely it
is sometimes true that those who give birth to a child are less competent and less well-suited to
serve as parents than any number of other persons willing to so serve.  And surely it is not
beyond the ability of courts to make this judgment.  So long as courts adhere to principles of
fairness in adjudication, including fair and reliable evidentiary rules, and so long as there is
appropriate appellate review, there is no reason here to prevent state redistributions of children
across the board.  Lacking any direct, non-derivative, non-instrumental right to a parent-child
relationship free of this sort of state interference, parents would have no grounds for objecting.

 But surely this is a deeply counter-intuitive and highly unacceptable result.  Since the
third account does not lead to this result, it is preferable in this respect.  But why think parents
have the sort of direct, non-derivative and non-instrumental right posited by the third account?
Surely, it can’t be because children “belong” to their natural parents.  Nor can it be because
parents have some sort of “natural” right to such a relationship.

 Parenting or child rearing is a basic life activity in which parents have a fundamental
expressive interest.  The right to a parent-child relationship free from certain sorts of state
interference is a right to engage in parenting as a basic life activity, and to express oneself or
one’s identity thereby.  It is a right belonging to the family of rights constitutive of personal
sovereignty.  I have a right to devote my life to painting even though all the world knows that I
am not and never will be a very good painter and that ultimately I and others would be happier
were I to make some other choice.  Similarly, I have a right to have and parent children, my own
children, even though all the world knows that I am not and never will be a very good parent and
that ultimately I and others, including my child, would be happier were I to make some other
choice.  It is this right that is violated when the state redistributes children early in life so as to
assign them to the parents best suited or most inclined to take good care of them.

 If this is right, then parents have a right to express themselves through their parenting or
child rearing decisions.  And this right includes the right to parent their children (within limits) in
ways that the state regards (correctly, let us suppose) as suboptimal.  When parents object to one
or another otherwise justifiable aspect of a compulsory education regime, it is this right that they
are asserting.  To defeat such parental objections, then, it is insufficient to show merely that it
would be suboptimal in terms of advancing the best interests of the child or children in question
to uphold the parental objection.  Parents have a right to make bad decisions with respect to the
rearing of their children.

 But that right is not unlimited.  Indeed, it is limited by the basic rights of children.  Thus,
parents have no right to make bad parenting decisions that violate or unjustifiably infringe upon
their children’s basic rights.  Now, every child has a basic right to a compulsory education
sufficient to meet the demands of justice (one aimed at literacy, numeracy, technical facility and
the like, as well as the virtue or virtues of reciprocity).  Parents who seek to shelter their children
from aspects of such an education seek to deny their children a basic right.  Courts should uphold
such a parental objection only if they find that not so doing would result in a worse violation of
rights than so doing.  This will be true in only two cases.  One is when the parental right at stake
is simply much weightier than the competing rights, so that even a minor infringement on it for
the sake of lesser rights will prove unjustified.  The other is when the competing rights are of
roughly equal weight but upholding the parental rights does not seriously threaten the
fundamental core of the other rights at stake but upholding those other rights does seriously
threaten the fundamental core of the parental rights.  When either of these conditions obtain, and
other things being equal, courts ought to uphold the parental objections.

 But neither of these conditions is likely ever to obtain with respect to any aspect (moral or
non-moral) of a compulsory educational regime demonstrably necessary to realize the minimum
institutional demands of justice.  Take, for example, a parental objection to a compulsory reading
list and subsequent in-class discussion covering diverse religious traditions.  This is the sort of
curricular requirement essential to an appropriate compulsory moral education into the virtue or
virtues of reciprocity.  Children have a basic right to such an education.  To deny them such an
education is to deny them an essential social and psychological condition to living actively within
a stable and at least minimally just constitutional democratic republic.  To deny the parental
objection, on the other hand, is not to deny the parents an essential condition to living within
such a state.  Constraining the freedom of parents when it comes to certain aspects of the moral
education of their children is compatible with parents still enjoying a great deal of freedom with
respect to the moral education of their children, with respect to the parent-child relationship more
generally, and with respect to their own projects, commitments and ends (the zone of personal
sovereignty).  Moreover, while the general right to personal sovereignty is surely as fundamental
a right as any there is, it is on no account an unlimited or unconstrained right.  The social and
psychological conditions necessary to the realization of justice mark its limit or constraint.  Thus,
it is appropriate that the balance tip in favor of protecting the rights of the child or children in
cases where parents object to one or another curricular requirement necessary or highly
conducive to meeting the minimum institutional demands of justice.  These include curricular
requirements necessary or highly conducive to bringing successive generations to the virtue or
virtues of reciprocity.

 We can, then, affirm the direct, non-derivative and non-instrumental right of parents to
determine much with respect to the education of their children and still insist rightly that every
just state, even a merely minimally just state, is a tutelary state.