Open Journal of Philosophy
2012. Vol.2, No.1, 10-16
Published Online February 2012 in SciRes (http://www.SciRP.org/journal/ojpp) http://dx.doi.org/10.4236/ojpp.2012.21002
Copyright © 2012 SciRes.
Justifying Tolerance in Liberal Societies:
The Need for Public Morality
Metropolitan College of New York, New York, USA
Received January 2nd, 2012; revised January 30th, 2012; accepted February 10th, 2012
One of the most important assumptions in liberal societies is that citizens should be tolerant of a diversity
of values. We are challenged by this assumption to justify restraint when we confront what we oppose,
disapprove of, or perceive to be immoral, even if we have the power to suppress perceived immoralities.
Based on the work of Elliot Turiel, Jonathan Haidt, and Gerald Gaus, the argument developed in this arti-
cle is that the best way to address the challenge is to distinguish between public morality and other nor-
mative categories such as convention and private morality. Public morality circumscribes what should not
be publicly tolerated. Conventional and private immoralities that are not prohibited by public morality
should be tolerated.
Keywords: Elliot Turiel; Gerald Gaus; Jonathan Haidt; Liberal Society; Public Morality; Tolerance
One of the most important assumptions in liberal societies is
that citizens should be tolerant of a diversity of values. We
celebrate value diversity and regularly admonish people to be
tolerant as if it is a self-evident truth. It may be surprising, then,
to hear from a philosopher, Catriona McKinnon, that there is a
dangerous complacency about toleration or tolerance in the
political zeitgeist and a mysterious inattention to the subject of
tolerance in the Academy. In a recent statement, McKinnon
(2006) says that her book aims to reassert the significance of
toleration by exploring the best current theoretical answers to
the following questions: How is toleration possible? Why is
toleration required? And, what are the limits of toleration? The
dangerous complacency about toleration in the political zeit-
geist is mirrored by a mysterious quiet in the Academy: with
few exceptions, the subject of toleration has been largely absent
from the academic literature for the last twenty-five years, and
the questions just listed have rarely been addressed directly (pp.
Philosophers may think that McKinnon’s claim about the
Academy is an exaggeration because there are some good
theoretical (philosophical) attempts to answer her questions if
not answers in the literature, although the bulk of the latest
work on tolerance is English, not American (see the introduce-
tion to her bibliography, p. 202). It is not an exaggeration,
however, to say that there is virtual silence, if not a mysterious
quiet, in the work of moral psychologists. Perhaps psycholo-
gists are silent because they understand that their work is de-
scriptive, not normative, although the boundary between the
empirical or descriptive and the normative has been dissolving.
Another possible explanation for the silence among moral
psychologists is that a defense of tolerance did not seem so
urgent in Elliot Turiel’s research program because of an as-
sumption that tolerance applied to the large conventional do-
main provided for in the theory. Most of the interesting theo-
retical action, especially concerning moral dilemmas, seemed to
be in the smaller moral domain involving harm, justice, rights,
and welfare. Extending the theory and research of Richard
Shweder (1997), Jonathan Haidt (2001) and others (Haidt &
Kesebir, 2010) have been pressing their fellow moral psycholo-
gists to expand the moral domain to include issues of in-group
loyalty, respect for authority, and spiritual purity. With a di-
verse moral domain nearly co-extensive with all of life, finding
the answers to McKinnon’s questions becomes more urgent. If
there is to be tolerance within this extensive domain, it seems
that we will inevitably be required to tolerate at least some
immoralities. This is the paradox of toleration articulated by
philosophers (McKinnon, 2006: p. 19).
In this article, my first task will be to clarify the meaning and
essential features of tolerance. I will then show that it is not
difficult to justify tolerance in Turiel’s moral theory. Jonathan
Haidt’s research represents a much more serious challenge to
the justification of tolerance. Based on the social philosophy of
Gerald Gaus (1999), I will argue for a normative theory that
justifies tolerance and resolves the paradox of tolerance pre-
sented by Haidt’s expanded moral domain.
Essential Features of Tolerance
McKinnon (2006, p. 14) provides a useful outline of the most
important features of tolerance on which most philosophers
1) Difference: what is tolerated differs from the tolerator’s
conception of what should be done, valued, or believed.
2) Importance: what is tolerated by the tolerator is not trivial
3) Opposition: the tolerator disapproves of and/or dislikes
what she tolerates, and is ipso facto disposed to act so as to
alter or suppress what she opposes.
4) Power: the tolerator believes herself to have the power to
alter or suppress what is tolerated.
5) Non-rejection: the tolerator does not exercise this power.
6) Requirement: toleration is right and/or expedient, and
the tolerator is virtuous, and/or just, and/or prudent.
As McKinnon notes, the first four features concern the cir-
cumstances of tolerance, that is, “the conditions in which it is
meaningful to describe one agent as tolerant of another” (p. 14).
Tolerance is a meaningful concept only if people differ in their
values, beliefs, or practices. Tolerance may be a meaningful but
not relevant concept if differences are not important or signify-
cant. Some people like to wear turtleneck sweaters in the winter,
but I do not. Because this difference is trivial to me, I simply
The third feature, opposition, is perhaps the most important.
This feature creates the most difficulty in justifying tolerance.
We can see this by considering a couple of examples. For ex-
ample, suppose that Mary Smith is a well-paid attorney who
works for a reputable law firm. I also work for the law firm and
live in Mary’s neighborhood. When in the local grocery store, I
notice that Mary selects several cans of dog food and puts them
in her cart. I walk by and ask about her dog. Mary says that she
does not have any dogs. She eats dog food because she likes it.
I do not say anything but walk away feeling disgusted. Mary
and I differ about food in a way that is not trivial to me; we
have an oppositional difference. I would strongly dislike eating
dog food, but actually I do not care if Mary eats it. I am op-
posed but indifferent. Our difference involves only a matter of
non-moral value. Tolerating Mary’s eating practice is not diffi-
Tolerance becomes more difficult, however, when opposition
involves a moral matter. For example, suppose that Mary has a
colleague, Lisa Jones, who Mary believes is a lesbian, although
Lisa ha s never di rect ly rev ealed her sexual orientation. One day
in a conversation at the water cooler, Lisa not only discloses
that her roommate is really her life-partner but she also explains
in vivid detail what they do together sexually in bed and their
plans to marry. Mary is horrified. She is a devout Catholic who
not only dislikes homosexuality but also believes that gay sex
and marriage are morally wrong. Because Lisa’s difference
with Mary involves a moral matter, not just a difference in val-
ues, Lisa finds it difficult to tolerate Mary’s sexual practice and
marriage plans. In Lisa’s mind she would be tolerating immoral
acts. This is the paradox of tolerance that must be resolved in
order for tolerance to be justified.
McKinnon (2006) says that “Features (4) and (5) relate to the
control the tolerator believes herself capable of exercising over
what she tolerates” (p. 15). The fourth feature (power) defines a
circumstance of tolerance and the fifth non-rejection describes
how a tolerator responds to the circumstance. A person may not
believe that she has the power to either alter or suppress what
she dislikes or disapproves of. In this circumstance, it would be
a misunderstanding of the term to say that this person is tolerant.
Rather, we would say that this person is resigned to the circum-
stance. “Prisoners do not tolerate their guards, or slaves their
masters” (McKinnon, 2006: p. 15). However, if a person be-
lieves that she has the power to alter or suppress what she op-
poses, then she is in a position to tolerate what is opposed, if
she does not exercise this power. In the example of Mary and
Lisa, if Mary believes that she has no power to affect Lisa’s
marriage plans, then she is not tolerant of but resigned to the
marriage. On the other hand, if Mary believes that she does
have the power to affect Lisa’s plans, perhaps by supporting a
powerful lobby against gay marriage, and she refrains from
using the power, then we can say that she tolerates Lisa’s mar-
The last feature entails two possibilities: tolerance is either a
prudential or moral requirement. One might hold that tolerance
is a prudent strategy to avoid serious conflicts, perpetual war
and the violence associated with war. In this case, both what is
tolerated and the prevalence of tolerance are variable, depend-
ing on the power relations among individuals and groups and
other cultural factors in a society that might contribute to or
ameliorate social cohesion and/or conflict. The other possibility
is that tolerance is a moral requirement and the tolerant person
is virtuous. In this case, peace is not inevitable, but the terms of
peace are clarified by moral requirements. A moral theory is
needed to establish the moral force of tolerance in this possibil-
ity. In the following, I will outline such a theory and suggest
how it is possible to say that a person who tolerates some im-
moralities is a virtuous person.
How challenging the justification of tolerance will be de-
pends on the definition of morality or the moral domain. For
example, if the moral domain is relatively small, perhaps with
one or two universally agreed upon moral principles, then jus-
tifying tolerance will not be difficult. Violations of the central
moral principles are intolerable, but actions, practices or value
differences outside the moral domain should be tolerated. In the
crucial fifth feature, non-rejection, we see that the major reason
a tolerator is not justified in exercising her power to eliminate
differences is that these differences belong to the non-moral
However, as the moral domain is expanded, the justification
of tolerance becomes more and more challenging. If the moral
domain is defined co-extensively with most life-activities, a
tolerator will be asked to refrain from exercising her power to
eliminate many differences that belong to the moral domain. Of
course, a principled justification of tolerance in this situation is
not necessary for tolerance to be meaningful. In an authoritarian
society, what should be tolerated and not are decided for the
citizens by one or more individuals who are powerful enough to
coerce conformity wi th t hei r preferences.
Justifying Tolerance in Turiel’s Moral Theory
It is not often recognized that the definition of morality de-
termines our understanding of the meaning and role of toler-
ance in society. Traditionally, the task of defining morality has
been the province of philosophers. As one might expect, phi-
losophers have not been able to agree on the definition (see the
review by Gert, 2008). Meanwhile, mostly outside the attention
of philosophers, developmental psychologists have been defin-
ing and empirically testing morality. One reason for this inat-
tention is that philosophers assume that a move from fact to
value is logically illicit. No matter what facts the psychologists
discover empirically, the development of a normative theory of
morality requires the kinds of logical argument that philoso-
phers are in the best position to provide. After twenty years of
reflection, analysis and study, philosophers and psychologists
have now begun a dialogue. Since the beginning of the twenty-
first century, philosophers have become increasingly interested
in and influenced by the empirical study of morality (Nado,
Kelly, & Stich, 2009: pp. 1-2).
Elliot Turiel is one of the central figures in this psychological
research tradition. Turiel began with a distinction between the
moral and conventional domains and a definition of the kinds of
rules that are appropriate to each domain. “Prototypical exam-
ples of moral rules include those prohibiting killing or injuring
Copyright © 2012 SciRes. 11
other people, stealing their property, or breaking promises. Proto-
typical examples of conventional rules include those prohibiting
wearing gender-inappropriate clothing (e.g., men wearing dresses),
licking one’s plate at the dinner table, and talking in a class-
room when one has not been called on by the teacher” (Kelly,
Stich, Haley, Eng, & Fessler, 2007: p. 117).
Moral rules are pan-cultural (universal), historically invari-
ant, objective and prescriptive; conventional rules are not.
Moral rules do not depend on the authority of any individual,
group, or institution; conventional rules do. Violations of moral
rules are more serious than violations of conventional rules.
Violations of moral rules involve harm to victims, violation of
rights, and unjust treatment; violations of conventional rules do
not. The purpose of conventional rules is to maintain social
order and coordinate social activities. A conventional rule may
be easily changed if another rule is determined to better achieve
this purpose (Turiel, 1983: pp. 2-4; for a succinct summary, see
Nado, Kelly, & Stich, 2009: pp. 2-3).
Over the course of thirty years, Turiel and his associates
tested these definitions using “an impressively diverse range of
participants differing in religion, nationality, culture and Age-
from 3.5 years to adulthood” (Nado, Kelly, & Stich, 2009: p. 4).
They found support for a robust distinction between the moral
and conventional domains and their characterization of the
difference between moral and conventional rules. They also
found evidence of the authority-independent, universal and
objective nature of moral rules and their early emergence in
Two observations about these findings are important. First,
transgressions of prototypical moral rules usually involve clear
harm to a victim, but subjects might justify their judgments
variously in terms of harm, justice, or rights. Kelly, Stick, Ha-
ley, Eng, & Fessler (2007) note this fact but not its significance
(p. 119). Harm, justice, and rights are different concepts, and
beliefs about all three may not be pan-cultural and universal. In
the work of philosophers who pay attention to research findings,
the assumption seems to be that harm-based violations are uni-
versal but not necessarily violations of rights or justice. For
example, Nichols (2004) concludes that “despite the cross-
cultural differences in moral judgment, the evidence indicates
that all cultures share an important basic capacity, what I will
call “core moral Judgment.” The capacity to recognize that
harm-based violations have a special status (as compared to
conventional violations) is an important indicator of the capac-
ity for core moral judgment. As a first approximation, the ca-
pacity for core moral judgment can be thought of as the capac-
ity to recognize that harm-based violations are very serious,
authority independent, generalizable and that the actions are
wrong because of welfare considerations (p. 7).
The other observation is that research on harm-based moral
violations and the philosophical use of this research assume an
open-ended, common-sense understanding of harm as “psy-
chological harms like pain and suffering” (Nichols, 2004: p.
What are the implications of Turiel’s definition of the moral
domain for the meaning and justification of tolerance? The
answer seems to be relatively straightforward. Violations of
moral rules that cause harm should not be tolerated. Violations
of conventional rules may be tolerated depending on the au-
thority. This simple solution does not mean that all disagree-
ment will be eliminated. In any application of the do-no-harm
principle, we still have to define precisely what counts as a
violation of the principle. People may also disagree about trade-
offs when violations conflict in moral dilemmas.
There will also be disagreement about rules within the con-
ventional domain. The distinguishing feature of a rule within
this domain is that it covers an activity or behavior that is less
serious than harm. It is also the case that some rules are autho-
rity-independent. No harm would be done if a conventional rule
were changed. For example, the authority-dependent rule pro-
hibiting driving on the left side of the street could be changed to
prohibit driving on the right side of the street. Two different
teachers could have opposite rules about chewing gum in the
classroom. These rules create necessary order in a society. On
the other hand, there are some conventional rules that are au-
thority-independent in that no particular person, group or insti-
tution established them. They are authority-dependent only in
the sense that they depend on social solidarity, influence, or
constraint. Some of the rules, for example, in regard to dress or
hairstyle, involve general social pressure to conform to what-
ever is customary at a particular time. Other authority-inde-
pendent conventional rules involve matters of etiquette. Some
people will disagree about what counts as a violation of a con-
ventional rule. They will disagree with the decisions of those in
authority and resist social pressure to conform to custom and
Another difficulty with my relatively straight-forward answer
is that I seem to have committed the naturalistic fallacy: I have
illicitly argued from facts to values-from pan-cultural facts
about the harm principle to the normative conclusion that harm-
based violations of rules should not be tolerated. Since Hume,
philosophers have worried about committing this fallacy; moral
psychologists have not (see Pojman’s chapter on the fact-value
problem for a basic account of the fallacy and various philoso-
phical responses to it, 2006: pp. 208-234). One reason may be
that intuitively we know that the fallacy can be logically cor-
rected by adding factual and normative premises to the argu-
1) All cultures share the basic “capacity to recognize that
harm-based violations are very serious, authority-independent,
generalizable and that the actions are wrong because of welfare
considerations” (Nichols, 2004: p. 7).
2) This basic capacity is the necessary and sufficient condi-
tion of a moral obligation not to engage in harmful activities
3) The violation of a moral obligation is intolerable.
4) Therefore harmful activities and behaviors should not be
The Challenge of Jonathan Haidt
The first serious challenge to Turiel’s conceptualization of
the moral domain came from Richard Shweder in the 1980s.
Based on research conducted in India, Shweder claimed that the
moral domain is culturally variable and extends beyond issues
of harm, rights and justice. He created a scheme of the moral
domain that includes three different ethics: an ethics of auton-
omy (judgments relating to issues of harm, rights and justice),
an ethics of community (issues of respect, duty, hierarchy, and
group obligation), and an ethics of divinity (issues of purity,
sanctity, and the recognition of divinity in each person) (Sh-
wedder, Much, Mahapatra, & Park, 1997; Haidt & Bjorklund,
2008a: pp. 196-197).
Turiel and his colleagues argued that Shweder was mistaken
Copyright © 2012 SciRes.
(Turiel, Killen, & Helwig, 1987). Properly interpreted, the In-
dians actually did understand that the violations in the research
scenarios were based on harm (according to an account by
Haidt & Bjorklund, 2008a: p. 196). In a study drawing subjects
who were from different social classes from Brazil and the
United States, Jonathan Haidt and his colleagues hoped to settle
the disagreement by including research scenarios that described
harmless taboo violations: “For example, a family eats its pet
dog after the dog was killed by a car; a woman cuts up an old
flag to create rags with which to clean her toilet; a man uses a
chicken carcass for masturbation, and afterwards he cooks and
eats the carcass” (Haidt & Bjorklund, 2008a: p. 196; see the
original design and findings in Haidt, Koller, & Dias, 1993).
Haidt and associates found that groups from the upper social
class (college students) did indeed support Turiel’s prediction
that harmless taboo violations would be classified as conven-
tional (though perhaps disgusting) and not as moral violations.
The other groups, however, supported Shweder’s claim that
some cultural groups operate with a broader moral domain.
These groups said that these harmless taboo violations represent
universal moral violations.
Haidt concluded from this study that Shweder was largely
correct: the moral domain is not universally confined to harm-
based violations (and violations of rights and justice). After a
review of some of the relevant literature on moral systems,
Haidt and his colleague, Craig Joseph, expanded the moral
scheme developed by Shweder (Haidt & Joseph, 2004). They
relabeled Shweder’s three ethics and added two additional sets
of moral concerns. The result is a moral scheme with five dif-
ferent categories (Haidt & Bjorklund, 2008a: p. 203):
1) Harm/care (a sensitivity to or dislike of signs of pain and
suffering in others, particularly in the young and vulnerable).
2) Fairness/reciprocity (a set of emotional responses related
to playing tit-for-tat, such as negative responses to those who
fail to repay favors) .
3) Authority/respect (a set of concerns about navigating
status hierarchies, e.g., anger toward those who fail to display
proper deference and respect).
4) Purity/sanctity (related to the emotion of disgust, neces-
sary for explaining who so many moral rules relate to food, sex,
menstruation, and the handling of corpses).
5) Concerns about boundaries between in-group and out-
These five categories represent the “psychological founda-
tions of human morality” (Haidt & Joseph, 2007: p. 16). The
foundations have been pre-wired by evolutionary forces, but
they are built upon by each culture. Cultures construct the vir-
tues and vices in variable ways on the basis of these founda-
tions. Haidt and Bjorklund (2008b) theorize that
Evolutionary forces have “prewired” human brains to read-
ily develop concerns about harm/care, fairness/reciprocity, in-
group/loyalty, authority/respect, and purity/sanctity. This pre-
wiring explains the otherwise uncanny similarity in cultural
practices such as initiation rites, or displays of deference, or
rules about purity and pollution that regulate food, sexuality,
and menstruation in so many cultures. Yet at the same time our
theory requires that the first draft be heavily edited by each
culture (p. 245).
All five foundations are candidates for moralization, which
can simply be defined “as the acquisition of moral qualities by
objects and activities that were previously morally neutral”
(Rozin, 1997: p. 380). Moralization occurs when people adopt
“virtue and vice words with which to praise and condemn peo-
ple, and to instruct their children” (Haidt & Joseph, 2007: p. 17;
for Haidt and Joseph’s chart listing the relevant virtues and
vices associated wi t h each foundation, see p. 31).
Societies do not moralize the five sets of concerns in the
same way. This accounts for moral variability across cultures.
But the variability is constrained by the foundations; it is not
indiscriminate. Moral variability can also be present within a
culture. One of the examples that Haidt and his associates fre-
quently use in their work is the moral variability that underlies
the so-called culture war between liberals and conservatives in
the United States (Haidt & Graham, 2007; Haidt & Graham,
2009; Haidt & Hersh, 2001). They believe that conflicts be-
tween liberals and conservatives can be explained by how each
side understands the moral status of the last three foundations:
in-group/loyalty, authority/respect, and purity/sanctity. Liberals
and conservatives agree on the moralization of concerns repre-
sented by the first two foundations because “all cultures have
virtues and concerns related to harm/care and fairness/recip-
rocity” (Haidt & Bjorklund, 2008a: p. 209). This is a universal
cultural factor, although Kelly, Stich, Haley, Eng, and Fessler
(2007) present evidence that harm norms may not be culturally
universal in all circumstances. The different ways in which
cultures moralize concerns represented by the other three foun-
dations is the variable factor: “cultures are quite variable in the
degree to which they construct virtues on top of the in-group/
loyalty, authority/respect, and purity/sanctity foundations” (Haidt
& Bjorklund, 2008a: p. 209).
Haidt and Bjorklund suggest that liberals are inclined to treat
the sets of concerns in the last three foundations as conven-
tional matters or as matters of prejudice and exclusion. Con-
servatives, like many or most people in other cultures, see these
sets of concerns as moral matters. Liberals want moral regula-
tion only of the rules related to the first two foundations (harm/
care and fairness/reciprocity) and advocate tolerance of behave-
iors classified under the last three (in-group/loyalty, author-
ity/respect and purity/sanctity). According to Haidt and Bjork-
Liberals value tolerance and diversity and generally want
moral regulation limited to rules that protect individuals, par-
ticularly the poor and vulnerable, and that safeguard justice,
fairness, and equal rights. Cultural conservatives, on the other
hand, want a thicker moral world in which many aspects of
behavior, including interpersonal relations, sexual relations, and
life-or-death decisions are subject to rules that go beyond direct
harm and legal rights (p. 209).
The liberal restriction of moral status to rules related to
harm/care and fairness/reciprocity is essentially the position
defended by Turiel. But, as conservatives see it, if the moral
domain includes all five foundations, the request to tolerate
many behaviors within three out of the five foundations is a
request to tolerate immoralities. The implication is that for
conservatives the conventional domain is much smaller than for
liberals. At the very least, conservatives must assume that lib-
erals have misclassified many kinds of immoral behavior as
conventional. Ever worse, perhaps liberals have created a state
of moral chaos. As Haidt and Bjorklund (2008a) say, “Conser-
vatives are horrified by what they see as the ‘anything goes’
moral chaos that liberals have created, which many see as a
violation of the will of God and as a threat to their efforts to
instill virtues in their children” (pp. 209-210).
We can see that my justification of Turiel’s liberal theory is
Copyright © 2012 SciRes. 13
seriously challenged by Haidt and associates. Despite their
claim that only two sets of concerns have been universally mor-
alized, most cultures, including our own, have constructed vir-
tues and vices based on all five foundations. Perhaps the dif-
ferences in the various culturally edited versions of the first
draft that is pre-wired in the brain by evolutionary forces have
more to do with the particular virtues and vices that are con-
structed and not with the fact that some cultures do not moral-
ize one or more sets of concerns. But if the concerns of all five
foundations are usually moralized, then we are confronted with
the paradox of tolerance: as conservatives imply, liberals are
asking them to tolerate immoralities. How can liberal tolerance
Justifying Tolerance after Haidt
We should recognize that the liberal request for tolerance in
our society is quite unusual. Tolerance is an unnatural and in-
secure value or virtue in the five-foundation theory because the
theory does not provide for a pre-wired foundation of tolerance
in the brain that might form the basis of culturally edited ver-
sions. Without a natural foundation, it is not difficult to under-
stand why most cultures are intolerant of almost any kind of
deviant behavior. One cannot commit the naturalistic fallacy of
moving from the natural or cultural fact of tolerance to the
moral value or virtue of tolerance if tolerance as a fact does not
exist. It should also be recognized that the five-foundation the-
ory results in a kind of moral relativism: moralization of the
foundational concerns in the edited versions is culturally vari-
able. And contrary to popular opinion, we cannot logically
move from the fact of moral relativism to the value of tolerance
without a normative argument: “relativism is a meta-ethical
doctrine; that is, it is a view about what morality is. However,
any principle of toleration can only be justified by a normative
argument; that is, by an argument about what ought to be done”
(McKinnon, 2006: p. 39).
In order to justify tolerance, my proposal is that we base our
normative argument on reason rather than nature or culture. I
suggest that we adopt the basic framework of Gerald Gaus’s
social contract theory. One reason for adopting Gaus’s frame-
work is that the harm principle is central to his theory. Turiel
and Haidt have compatible understandings of this principle.
Another reason is that Gaus’s theory is a social philosophy, that
is, a social or public morality. Gaus’s public morality is not a
complete moral theory. It is a theory about how people should
regulate their common life together as citizens, but strangers, in
Gaus explicitly explains how his theory is located within the
liberal tradition of social philosophy.
The liberal tradition in social philosophy maintains that each
person is free to do as he wishes until some justification is of-
fered for limiting his liberty. As liberals see it, we necessarily
claim liberty to act as we see fit unless reason can be provided
for interfering. I shall call this the Liberal Principle: 1) A per-
son is under no standing obligation to justify his actions. 2)
Interference with another’s actions requires justification, un-
just-tified interference is unjust (pp. 118-119).
According to Gaus’s formulation of the liberal principle,
what requires justification is the interference with a person’s
actions, not the action itself. A major assumption in the liberal
tradition is that liberty is the natural state of affairs or moral
status quo “in the sense that it requires no justification while
limitations of it d o ” (Gaus, 1999: p. 119).
How is the interference with another’s actions justified? In-
terference is justified if an action is harmful. For Gaus (1999),
harm to others is “the core principle of social morality” (p. 111).
Gaus (1999) claims that all goal-pursuing persons would agree
that interference can be justified when an action sets back the
welfare interests of others. He argues that they would agree
because everyone has reasonable grounds for accepting the
harm principle and no one has reasonable grounds for rejecting
it (Gaus, 1999: p. 26). Reasonable grounds are ones that are
clearly not defective, “that is, they are not based on clear mis-
takes in reasoning, or on clearly false information, or on mani-
fest ignorance” (Gaus, 1999: p. 26). Welfare interests are those
interests that are instrumental to the realization of our goals. It
would be impossible to produce a complete, uncontroversial list,
but some of the interests typically included are bodily security
and health, liberty itself, security of property, self-respect, pri-
vacy, and being told the truth (Gaus, 1999: p. 148).
Gaus’s definition of harm as a set back of welfare interests is
broader than the definition assumed by moral psychologists. As
I have already noted, moral psychologists understand harm in
terms of psychological pain and suffering. Following Feinberg
(1984), Gaus takes some kind of interest to be the object of
harm. Welfare interests are one kind. What Gaus calls “regula-
tive interests” is the other most important kind of interest.
Regulative interests are the goals of persons. Gaus excludes
harm to goals or regulative interests as an agreed-up, justified
reason for interference. Gaus (1999) defines them as neutral
interests: “Regulative interests are neutral, in the sense that
while we are free to pursue them, we are not harmed when they
are thwarted” (pp. 150-151). If regulative interests were in-
cluded, unjustifiable harms to others would be impossible to
limit. For example, I may have an interest, my goal or regula-
tive interest, in marrying a particular woman. She thwarts my
interest by marry ing someone else. This kind of harm in society
is so pervasive that if it we re counted as a legitimate re ason for
limiting action the harm principle would be meaningless and
social life would be impossible.
Gaus (1999) also argues that everyone has reasonable grounds
for excluding any conception of a harmful act that depends on a
belief that something is perverse, bad, wrong or immoral (p.
143). For example, the production of pornography would not be
classified as a harmful act just because someone is distressed by
knowing that somewhere it is produced and distributed. We
cannot make belief in the inherent immorality of something and
unpleasant mental states grounds for prohibittion. If these are
counted as harms, then individuals have a right to be protected
from such harms. This right must be excluded. Otherwise, a
public or social morality would be impossible because a dis-
tinction could not be made between personal ideals about the
good and social morality. As Gaus (1999) explains,
Whenever people fail to live up to my idea of the good-
whenever they do things that I think are perverse or evil-I can
get them to conform to my personal ideals by getting distressed
about their actions. That, though, would undermine the whole
project of generating a shared social morality that is distinct
from personal ideals (p. 144).
In Gaus’s public morality, we can justify the moralization of
only some concerns about harm. The core of Gaus’s theory is
built upon Haidt’s harm/care psychological foundation, only
one of five foundations of moral importance. This means that
Copyright © 2012 SciRes.
for Gaus the moral domain is quite narrow, even narrower than
Turiel’s moral domain, which also includes concerns about
rights, fairness, and justice, that is, Haidt’s fairness/reciprocity
foundation. (However, Gaus, 1999, incorporates “some sup-
plemental reasons as genuine public reasons to justify limiting
liberty” in his framework that I have excluded, p. 111). Gaus
assumes that all other concerns must be classified under a non-
moral category of personal ideals or values. Following Turiel,
we can say that Gaus’s theory requires a large conventional domain.
In addition to the concerns of fairness/reciprocity, conservative
concerns related to the foundations of in-group/loyalty, author-
ity/respect, and purity/sanctity must have a non-moral status
and be classified under the conventional domain.
Although Gaus’s theory creates a large zone of tolerance, it
entails the demotion to conventional status of a large number of
concerns that many people believe have moral status. I think
that we can preserve the status of these concerns by adding the
category of private morality to Gaus’s framework. This addi-
tion allows us to recognize that concerns about fairness/recip-
rocity and other conservative concerns are moral matters, but
they cannot be justified as part of public morality. Invoking
Gaus, we can say that some individuals have reasonable grounds
for rejecting these concerns as a basis for rules and laws that
everyone in society is obligated to obey.
I also recommend that we explicitly acknowledge rules of
etiquette as another moral category that is not a part of public
morality. Turiel classifies these rules under the conventional
domain, but they are actually based on Haidt’s five foundations.
These rules concern minor infractions, and we usually do not
criminalize violations. For example, bumping into a stranger on
the street, pushing in a crowded store, or going to work with a
bad cold represent minor harmful acts. Going to the head of the
line in a supermarket checkout is unfair. Eating the discharge
from one’s nose in close proximity to another passenger on a
crowded bus is offensive because it involves a concern related
to the purity foundation. We all probably believe that rules of
etiquette have some obligatory force, but following them is
optional in the sense that they are not included in public moral-
ity and enforced.
Using my framework, it is now possible to say how, in light
of Haidt’s five-foundation theory, the paradox of tolerance
might be resolved. A public morality based on the foundation
of harm/care will entail the toleration of a variety of immorali-
ties classified under the category of private morality. It is also
true that my framework does not avoid the demotion of private
moralities to a lower status. I have created a hierarchy in which
the harm principle is supreme. For these reasons, it may be
more accurate to say that I am proposing that we live with
rather than resolve the paradox. If so, tolerance should be un-
derstood as a prudential virtue.
However, invoking Gaus again, I would argue that everyone
has reasonable grounds for tolerating some kind of immoralities
and no reasonable grounds for rejecting such toleration because
everyone needs terms of cooperation and peace to survive and
flourish in a society of strangers and even to have a private
morality. For this reason, my proposal does entail a resolution
of the paradox, and tolerance should be understood as a moral
virtue because everyone in society has an obligation to tolerate
private immoralities. Thus, tolerance of some private immorali-
ties is not just optional and prudent; it is a moral requirement.
My argument for a resolution of the paradox of tolerance-that
the moral virtue of tolerance requires the toleration of at least
some immoralities—has progressed through several steps. I
began with a question about why tolerance has not received
more political and academic attention. I suggested that one
answer is that tolerance may not be seen as problematic in the
kind of liberal moral theory proposed by Elliot Turiel in which
the moral domain is restricted to justice, rights, and welfare. I
showed how easy it is to correct a standard criticism that Tu-
riel’s work is only descriptive by adding a normative premise to
a logical argument for tolerance of conventional co ncern s.
I concluded with a more significant challenge to the justifica-
tion of tolerance presented by Jonathan Haidt and associates.
Continuing the seminal work of Richard Shweder, Haidt’s re-
search suggests that the moral domain is much more extensive
than Turiel allowed. With broad moral and small conventional
domains, we are inevitably confronted with the prospect of
moral relativism and the paradox of tolerance. I proposed a way
for us to move logically from descriptive moral relativism to a
normative moral theory and resolve the paradox by adapting
Gerald Gaus’s moral framework to include several categories:
public morality, private morality, etiquette, and convention.
Feinberg, J. (1984). The moral limits of the criminal law. Harm to
others (Vol. 1). New York, NY: Oxford University Press.
Gaus. G. F. (1999). Social philosophy. Armonk, NY: M. E. Sharpe.
Gert, B. (2008). The definition of morality. In Stanford encyclopedia of
philosophy. URL (last checked 29 Dece mber 2011)
Haidt, J. (2001). The emotional dog and its rational tail: A social intu-
itionist approach to moral judgment. Psychological Review, 108,
Haidt, J., & Bjorklund, F. (2008a). Social intuitionists answer six ques-
tions about moral psychology. In W. Sinnott-Armstrong (Ed.), Moral
psychology. The cognitive science of morality: Intuition and diversity
(Vol. 2, pp. 181- 217). Cambridge, MA: MIT Press.
Haidt, J., & Bjorklund, F. (2008b). Social intuitionists reason, in con-
versation. In W. Sinnott-Armstrong (Ed.), Moral psychology. The
cognitive science of morality: Intuition and diversity (Vol. 2, pp.
241-254). Cambridge, MA: MIT Press.
Haidt, J., & Graham, J. (2007, March). When morality opposes justice:
Conservatives have moral intuitions that liberals may not recognize.
Social Justice Research, 20, 98-116.
Haidt, J., & Graham, J. (2009). Planet of the Durkheimians, where
community, authority, and sacredness are foundations of morality. In
J. T. Jost, A. C. Kay, & H. Thorisdottir (Eds.), Social and psycho-
logical bases of ideology and system justification (pp. 371-401). New
York, NY: Oxford University Press.
Haidt, J., & Hersh, M. A. (2001). Sexual morality: The cultures and
reasons of liberals and conservatives. Journal of Applied Social Psy-
chology, 31, 191-221.
Haidt, J., & Joseph, C. (2004). Intuitive ethics: How innately prepared
intuitions generate culturally variable virtues. Daedalus, 133, 55-66.
Haidt, J., & Joseph, C. (2007). The moral mind: How five sets of innate
intuitions guide the development of many culture-specific virtues,
and perhaps even modules. In P. Carruthers, S. Laurence, & S. Stich
(Eds.), The innate mind. Foundations and the future (Vol. 3). New
York, NY: Oxford University Press. URL (last checked 29 December
2011) http://people.virginia.edu/~jdh6n/publications.h tml
Copyright © 2012 SciRes. 15
Copyright © 2012 SciRes.
Haidt, J., & Kesebir, S. (2010). Mora lity. In S. T. Fiske, G. Lindzey, &
D. T. Gilbert (Eds.), Handbook of social psychology (pp. 797-832).
Hoboken, NJ: Wiley.
Haidt, J., Koller, S., & Dias, M. (1993). Affect, culture, and morality,
or is it wrong to eat your dog? Journal of Personality and Social
Psychology, 65, 613-628. doi:10.1037/0022-3522.214.171.1243
Kelly, D., Stich, S., Haley, K. J., Eng, S. J., & Fessler, D. M. T. (2007).
Harm, affect, and the moral/conventional distinction. Mind & Lan-
guage, 22, 117-131. doi:10.1111/j.1468-0017.2007.00302.x
McKinnon, C. (2006). Toleration: A critical introduction. New York,
NY: Routledge. doi:10.4324/9780203300640
Nado, J., Kelly, D., & Stich, S. (2009). Moral judgment. In J. Symons,
& P. Calvo (Eds.), Routledge companion to the philosophy of psy-
chology. New York, NY: Routledge. URL (last checked 29 Decem-
Nichols, S. (2004). Sentimental rules: On the natural foundations of
moral judgment. New York , NY: Oxford University Press.
Pojman, L. P. (2006). Ethics: Discovering right and wrong (5th ed.).
Belmont, CA: Thomson Wadsworth.
Rozin, P. (1997). Moralization. In A. M. Brandt, & P. Rozin (Eds.),
Morality and health (pp. 379-401) . New York, NY: Routledge.
Shweder, R. A., Much, N. C., Mahapatra, M., & Park, L. (1997). The
“big three” of morality (autonomy, community, divinity) and the “big
three” explanations of suffering. In A. M. Brandt, & P. Rozin (Eds.),
Morality and health (pp. 119-169) . New York, NY: Routledge.
Turiel, E. (1983). The development of social knowledge: Morality and
convention. New York, NY: Cambridge University Press.
Turiel, E., Killen, M., & Helwig, C. C. (1987). Morality: Its structure,
function, and vagaries. In J. Kagan, & S. Lamb (Eds.), The emer-
gence of morality in young children (pp. 155-243). Chicago, IL:
University of Chicago Press.