The Moral Emotions

Oxford Handbooks Online

The Moral Emotions  
Jesse Prinz
The Oxford Handbook of Philosophy of Emotion
Edited by Peter Goldie

Print Publication Date: Dec 2009 Subject: Philosophy, Moral Philosophy

Online Publication Date: Jan 2010 DOI: 10.1093/oxfordhb/9780199235018.003.0024

Abstract and Keywords

Emotions play many roles in human psychology. Fear alerts us to dangers, surprise
registers novelty, and disgust helps us avoid potential sources of contamination. Many of
these functions unite us with simpler creatures, and are, in that sense, among our more
primitive or ancient psychological capacities. But emotions also play a role in the most
sophisticated aspects of human mental life: they play a role in forming enduring social
bonds to individuals and large groups, they give us pleasure in the arts, and they make
fundamental contributions to human morality. It is this latter vocation that this article
explores. Three questions are addressed: what roles do emotions play in morality? Which
emotions play those roles? And are the emotions that play these roles distinctively moral?

Keywords: moral emotions, human psychology, contamination, psychological capacities, mental life, social bonds

EMOTIONS play many roles in human psychology. Fear alerts us to dangers, surprise
registers novelty, and disgust helps us avoid potential sources of contamination. Many of
these functions unite us with simpler creatures, and are, in that sense, among our more
primitive or ancient psychological capacities. But emotions also play a role in the most
sophisticated aspects of human mental life: they play a role in forming enduring social
bonds to individuals and large groups, they give us pleasure in the arts, and they make
fundamental contributions to human morality. It is this latter vocation that we will explore
here. Three questions will be addressed: what roles do emotions play in morality? Which
emotions play those roles? And are the emotions that play these roles distinctively moral?
To forecast my response to this last question, I will argue that the emotions involved in
morality actually serve non‐moral functions as well or derive from non‐moral emotions; if
taken to designate a proprietary class, the term “moral emotions” is something of a
misnomer.

Page 1 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

(p. 520) 23.1 The Roles of Emotion in Morality

There is an old philosophical tradition according to which emotions are an impediment to
rational and noble behavior. On this view, emotions reflect our most primitive, animalistic
nature, and should be repressed wherever possible. This perspective is sometimes
expressed in moral philosophy, where emotions are seen as an impediment to good
conduct rather than a facilitator. This kind of approach is sometimes attributed to Kant,
who insists that morality cannot be based on mere sentiments or inclinations. This may be
a cartoon version of Kant's actual view, since he places emphasis on respect, and he says
we have a duty to be sympathetic. Kant also says that human beings have a “susceptibility
to feel pleasure or displeasure merely from being aware that our actions are consistent
with or contrary to the law of duty” (Kant, 1885/1996: 160). Kant even goes so far as to
say that “any consciousness of obligation” depends on these feelings. Thus, the textbook
claim that Kant wants to exorcize the emotions is inaccurate, even if he does not think
emotions are the source of moral laws. In any case, Kant's project is normative in nature
—it is a characterization of how we should make moral decisions, not how we actually do.
It is uncontroversial to claim that, as a matter of fact, emotions play a significant role in
moral psychology. Or rather, several significant roles. Before offering a taxonomy of which
emotions contribute to morality, it is important to gain some clarity on how emotions
contribute. Two major contributions can be distinguished.

First of all, emotions play an important role in motivating moral behavior. We want to be
good, and we find good behavior rewarding. Bad behavior is, in contrast, emotionally
costly. And we are motivated to help others, in part, because we feel affection for them,
affinity, or compassion. The nature of prosocial motivation is hotly debated: do we help
out of self‐interest or out of genuine and fundamental concern for others' welfare? But,
whether motivation is egoistic or genuinely altruistic, emotions seem to matter. Egoists
credit self‐love and altruists credit love of others. Emotions also motivate us to punish
and seek revenge, which are both moralistic behaviors. One might be able to ascertain
that a person violated a norm without emotion, and one might be able to ascertain that
punishment would be appropriate (e.g., that it would serve as a deterrent), but the
motivation to punish is often retributive in nature, and when it is, emotions are actively
involved.

Second, emotions play a role in moral epistemology. Moral evaluation is sometimes

compared to perception: we see things as good or as bad. This capacity is widely believed
to have an emotional basis. Something strikes us as good or bad in virtue of the emotional
response it elicits. By comparison, a food might strike us as delicious because of the
gustatory pleasure it produces, and a joke might strike us as funny because of its
tendency to amuse. As we will see, different emotions may contribute to moral evaluation
in different ways.

Page 2 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

The issue of moral evaluation raises an important question that is beyond the
(p. 521)

scope of this chapter: can one judge that something is morally good or bad in the absence
of emotion? Or is evaluation essentially affective? The answer depends on the most
fundamental question in metaethics: what is the ultimate basis of morality? If moral
judgments refer to a class of facts whose essence does not involve the emotions, then the
answer would seem to be yes. Suppose, for example, that a morally good action is one
which maximizes happiness (Mill). Or suppose a bad action is that which we cannot, on
pain of practical irrationality, will as a universal law (Kant). One could ascertain that a
given action is good or bad in either of these senses without having any emotional
response. One can ascertain that action produces happiness without being happy. So
these leading normative theories open the possibility for a dispassionate moral
epistemology. Still, one might believe that these ethical theories are right, while allowing
that emotions, as a matter of fact, play a role in moral epistemology. Properly trained,
emotions might come to register cases where utility is increased or moral duties are
violated. Emotions could serve as rough‐and‐ready tools for perceiving moral facts, in
much the way that fear alerts us to dangers. Danger is not itself an emotional thing, but
fear serves as a danger detector. Likewise, emotions could alert us to the presence of
morally significant events. A sympathetic response to a victim of cruelty might help us
perceive that the act is not conducive to happiness. One could discover that without
emotions, but emotions surely help.

Other ethical theories entail that emotions are essential. Consider emotivism, which says
that moral statements of the form “ɸ‐ing is morally bad” or “ɸ‐ing is morally good” are
not statements of fact but expressions of feeling. On this view, the analogy between moral
epistemology and perception breaks down. There are no moral facts out there to be
perceived. Morality is more like a projection of feelings onto the world (see Chapter 26,
this volume). But emotivism is not the only option for those who think emotions are
components of moral judgments. An alternative would be to adopt a form of subjectivism,
according to which the statement “ɸ‐ing is morally bad” is true just in case “ɸ‐ing is
disposed to cause a certain negative emotional response in the person making that
statement, under certain conditions. Subjectivist theories come in many variants (see
McDowell, 1985; Wiggins, 1987; Prinz, 2007), but the basic idea is that moral statements
express facts whose existence depends on the response of evaluators. Defenders of these
views sometimes compare moral facts to facts about properties such as funniness,
scariness, or deliciousness, all of which are arguably subjective as well. The tradition also
draws inspiration from John Locke's account of “secondary qualities”. For Locke, certain
perceivable properties, including colors, can be characterized in terms of the responses
that certain objects cause in observers: for example, red can be characterized as the
power that red objects have to cause a red experience in observers under adequate
lighting conditions. If morality is like this, the analogy to perception is fairly
straightforward. Red is perceivable and out there on the (p. 522) surface of objects, even
though it depends on us. Likewise moral badness is out there, even though it depends on
our responses, and we pick up badness through our emotions.

Page 3 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

Both of these approaches face serious challenges. Emotivists must explain why moral
statements appear to be assertoric if they are merely expressive, and subjectivists must
overcome charges of vicious circularity and must characterize ideal observation
conditions (or some alternative). Both approaches also face difficulties relating to moral
disagreement and relativism: moral debates look like genuine disagreements, but, if
emotivists and subjectivists are right, there may be no single set of moral facts that could
adjudicate a debate between two people with opposing moral values.

This is not the place to address such concerns. I mention emotivism and subjectivism to
illustrate the role that emotions play in some accounts of moral epistemology. Whether or
not some version of these views can be defended, there is wide agreement that moral
judgments often occur with emotions. Some will say those emotions cause moral
judgments, some will say they are components, and some will say they are effects. These
debates need not concern us here. What matters is that, when we come to regard
something as morally good or bad, we have characteristic emotional reactions.
Surprisingly, this widespread observation is often made without careful discussion of
what these characteristic reactions are. That will be one of the main tasks in what
follows. We will see that different emotions arise in the context of different kinds of
evaluations.

In summary, emotions are said to play different roles in morality, and chief among these
are roles in moral motivation and moral evaluation. In discussing the moral emotions, it
will be useful to bear these two roles in mind. I will begin by surveying emotions that play
a role in evaluation, and then move on to motivation. Some emotions, we will see, play
both roles, and others may not. Each moral emotion makes a distinctive contribution to
moral psychology.

One final note is necessary before we begin our survey. Throughout this discussion, I will
try to be neutral about the nature of emotions. On some theories, emotions are cognitive
in nature: they are or contain judgments. On other theories, emotions are non‐cognitive;
for example, they are feelings. But this classic debate in emotion theory can be bypassed
by finding some common ground. Both cognitivists and non‐cognitivists typically assume
that emotions have characteristic eliciting conditions; for example, fear occurs in
response to (real or anticipated) danger, and sadness occurs in context of loss. Both sides
also agree that emotions typically promote different patterns of behavior, and that
behaviors require bodily changes that can be experienced as conscious feelings. So, in
what follows, I will characterize emotions by their eliciting conditions and action
tendencies. In so doing, I will try to avoid begging any questions against cognitivists or
non‐cognitivists.

(p. 523) 23.2 Evaluative Emotions

Page 4 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

23.2.1 Emotions of Blame

There are many different kinds or moral evaluations: we might describe some behavior as
morally permissible, impermissible, or compulsory. We might describe an action as cruel,
or kind, or forgivable. We might say a choice is the right thing to do or the wrong thing to
do. We might describe a person as vicious or virtuous. But we also have a kind of
evaluation that is simpler and, arguably, more fundamental. We judge that certain things
are morally good or bad (moral or immoral). We apply these assessments to actions,
events, persons, political regimes, and so on. Williams (1985) says that judgments of
goodness and badness are thin, as opposed to thick, because, unlike “cruel” or “kind”,
they have no overt descriptive content, demarcating the kind of thing being evaluated. I
will focus on these thin concepts here, because the fact that they lack overt descriptive
content means they are useful for discussing the psychology of evaluation as opposed to
description. Judgments of moral goodness and badness can be characterized as purely
evaluative. They also seem to be characteristically associated with emotions. As we will
see, however, the emotions used in pure evaluation may smuggle in constraints on
content akin to Williams' thick concepts. If so, the idea that we ever make an evaluation
that is truly thin might be called into question. I return to this issue at the end of this
section.

Pure evaluations come in two flavors: good and bad. Let us begin with the bad. The
emotions that play a role in negative evaluations are, presumably, negative, but which
ones? We have some negative emotions, such as fear and loneliness, that are not
characteristically deployed in moral evaluation. But others, such as anger and guilt, are
frequent players in morality. To demarcate the difference between negative emotions that
are important for evaluations of moral badness and those that are not, we might say that
the former are emotions of blame. Morality is a domain of blame. If someone does
something bad or has bad character, we blame that person. We don't necessarily conclude
that the person acted intentionally or with freewill, but we regard them with blame, in
some broad sense: we regard them as morally faulty. It is difficult to characterize the
notion of blame in a non‐circular way, but one salient feature is that when we blame
someone, we typically think they should be punished. Emotions of blame are typically
associated with a desire to punish, and, as we will see, they may even exert an influence
on the kind of punishment we come to desire.

Within emotions of blame, there is another broad distinction: some are self‐directed and
some are directed at others. Let's begin with other‐directed moral emotions. One of these
was already mentioned: anger. When we evaluate something as bad, we sometimes feel
angry. Anger can be characterized by its eliciting conditions (p. 524) and action
tendencies. Anger is typically elicited when one person does something harmful to
another, especially when the harm is intentional or negligent. The harm can be symbolic,
as in the case of an insult; indirect, as in the case of property damage or harm to a loved
one; or direct, as in the case of physical assault. Anger is associated with physical

Page 5 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

aggression. When we get angry, blood flows to our extremities, heart rate increases, we
make facial expressions that communicate potential attack (such as glaring or baring
teeth), we raise our voices, and fists may clench.

As a moral emotion, then, anger can be associated with a specific kind of transgression:
crimes against persons. A crime against a person is a behavior that harms or threatens to
harm an individual. In a study by Rozin et al. (1999), subjects were presented with a list
of moral transgressions, and they were asked to select which emotion they would feel for
each. Some of these were crimes against persons (what Rozin et al. call violations of
Autonomy norms). The cases included scolding, beating, embezzling, line‐cutting,
stealing, and so on. Anger was overwhelmingly and systematically chosen for these.
Anger can also be associated with a specific kind of moral behavior: harmful punishment.
When a harm occurs, and we get angry, we often desire that perpetrators be harmed by
taking their property, depriving them of liberty, or making them experience physical pain
or even death. Lerner et al. (1998) showed that desire for punishment is increased by
anger induction, even when the anger is introduced independently of the behavior being
punished. Fehr and Gächter (2002) showed that anger correlates with the desire to
punish defectors in economic games. Parrott and Zeichner (2002) showed that men who
are more angry in temperament administer more intense and longer electric shocks in a
competitive task with another player.

The term “anger” can be regarded as an umbrella term for a range of more specific
emotions that share the same broad category of elicitors and action tendencies. Anger
varies in degrees and includes mild emotions such as irritation and intense emotions such
as rage. Presumably, intensity will vary, in part, with the severity of the harm. If a man
leaves the toilet seat up, his female roommate might be irritated, and, if someone
commits and act of willful murder, the emotion might be rage. Intensity will also vary with
proximity to the victim: if I am harmed I will normally feel more anger than if a stranger
is harmed. We also have terms for specific species of anger that vary as a function of the
kind of harm that has been perpetrated. The term “indignation” is often reserved for
cases where the harm is a case of unfairness, such as an inequitable distribution, and the
term “exasperated” usually refers to repeated mild offenses.

Contemporary moral philosophers tend to focus on harm, and, as a result, they might
tend to think that anger is the primary (or perhaps even only) other‐directed emotion of
moral blame. But there may be other kinds of moral norms that don't primarily involve
harming an individual. In some societies, victimless crimes are central to morality. For
example, consider societies that condemn various victimless sexual behaviors, such
masturbation or homosexual intercourse. Liberal (p. 525) Western societies continue to
enforce many such norms; we have rules against bestiality (assuming the animals don't
mind), necrophilia, plural marriage, and consensual incest between adults. In some of
these cases, a party may be harmed, but it is easy to imagine cases where no one is
harmed and where there is nevertheless a tendency to find the behavior morally suspect
or reprehensible. Sometimes, the person who judges such behavior to be wrong may
experience anger. This will happen when the act is perceived as a potential threat (for

Page 6 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

example, homosexuality is sometimes portrayed as a threat to the fabric of society). But,

in cases where there is no perceived threat (such as bestiality), the emotion experienced
will be different: it will usually be disgust (Rozin et al., 1999). Gutierrez and Giner‐Sorolla
(2007) showed that in cases of taboo violation, anger occurs when there is a presumption
of harm, and disgust occurs otherwise.

Outside the moral domain, disgust is a response to contamination (decay, rot, sickness,
deformity). In this respect, it has much to do with the body. It protects the body of the
disgusted by causing it to avoid contact with the object of disgust. Predictably, therefore,
moral disgust arises most typically in the case of norms that involve bodily contact and
bodily fluids—potentially sources of contamination. Sexual norms are the most obvious
cases. We call these unnatural acts or crimes against nature, and we see violators as
impure. Moral disgust also arises in cases of norms that have to do with animals or
animality, since animals are a potential source of contamination. This includes bestiality,
but also dietary norms pertaining to animals, which can be found in some cultures, such
as kosher rules or rules against eating human flesh (even if the people died of natural
causes). If disgust deals with the body, then moral disgust should also arise in cases
where a transgression causes salient bodily damage, as in the case of mutilation or
genocide (because it draws attention to masses of dead bodies). In these cases, anger
may also be experienced because harm is involved. We may also extend disgust to cases
of “moral deformity” or “moral sickness”: a racist or someone who shows egregious moral
indifference might be regarded as aberrant and repulsive.

Disgust has action tendencies. Contamination leads to expulsion and avoidance. We want
to distance ourselves from the things that disgust us. We pull away and shield ourselves
from disgusting things. We close our eyes and mouths, and we wrinkle our noses (to
reduce inhalation). This action pattern also carries with it a tendency towards magical
thinking. Since disgust protects us from contamination, and contaminants are often
invisible, we avoid things that have come into contact with something that is visibly
disgusting. This is true even if the object is known to be harmless. For example, we won't
drink from a cup that has been used to transport a urine sample, even if we know it has
been sterilized (Rozin et al., 1986). The fact that knowledge does not eliminate disgust
suggests that disgust is not elicited by explicit beliefs about contamination but is rather
an evolved response that uses simple heuristics, such as contact with bodily fluids and
rot, to keep track of potential sources of contamination. In the moral case, the same
(p. 526) principles operate. People refuse to put on a sweater if they believe it was worn
by Hitler (Rozin et al., 1994).

The contrast between disgust and anger is striking, because disgust promotes withdrawal
and anger promotes approach behavior (aggression). This may have an impact on such
things as how we treat criminals: you might prefer to live next to a convicted thief than a
convicted pedophile, even if you have no children. The emotions also seem to govern
significantly different domains. Nevertheless, both might be used in the context of
evaluating something as morally bad. For example, someone might feel disgust when they

Page 7 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

judge that it's “morally bad” to derive sexual pleasure from a pet cat and “morally bad” to
steal money from the Oxfam donation jar, even though the underlying emotions differ.
Distinguishing moral emotions can help us distinguish kinds of moral badness.

Disgust and anger are certainly among the most prominent emotions of other‐directed
blame, but they are not the only ones. Another important moral emotion is contempt.
Rozin et al. (1999) show that disgust is the dominant response for “crimes against
community”. These include cases of disrespect to parents and authorities, cases where a
wealthy person shows bias against a worker, and cases of flag burning. In some of these
cases, individuals are harmed (e.g., if a teen is rude to her parents, they may become
upset), but the salient focus of these violations is that the perpetrator has failed to
appreciate a person's rightful place in society. In this sense, society itself is a kind of
victim. Contempt also arises in cases where public trust or public property is destroyed.

Contempt may be a blend of other emotions, especially anger and disgust (Prinz, 2004).
This makes sense on the proposal that contempt occurs in response to crimes against
community, because the community can be construed as the natural order of persons, so
a crime against community is both a crime against persons and an unnatural act.
Contempt also exhibits a kind of directionality. We look down on those for whom we have
contempt. This suggests that contempt may also be related to emotions that regulate
social status hierarchies in mammals. Perhaps there are feelings associated with
dominance that come out in contempt. Those feelings may also be components of related
emotions such as derision and smugness, which are not fundamentally moral emotions
(they do not involve blame).

It is not obvious what the action tendencies are when we experience contempt. If
contempt blends anger and disgust, the action tendencies should be practically
inconsistent: contempt should promote both approach and withdrawal. This may in fact
be the case. Consider a politician who violates public trust. We may be angry and crave
punishment, but we will also say that the public office has been defiled, and we may be
literally repelled when we see the politician's image in the newspaper.

Another emotion of blame is disappointment. Disappointment occurs when we had high

expectations for someone and the expectations were not met. In the moral case, it arises
when we expect someone to be morally good and they are not. This is most likely to be
true for people we like and care about, such as loved ones. If your (p. 527) child is caught
cheating in school, you may be disappointed rather than angry. The action tendency may
not be aggressive or punishment‐focused. You may sulk or feel a general reduction in
motivation—a feeling of hopelessness or defeat.

These emotions (anger, disgust, contempt, and disappointment) may not be the only
emotions of other‐directed blame, but they are certainly the most prevalent. Each one
arises under somewhat different conditions, and each results in somewhat different
behavioral dispositions. In addition to these other‐directed emotions of blame, there are

Page 8 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

self‐directed emotions of blame. Two such emotions are especially common: guilt and
shame.

Guilt and shame are sometimes treated as if they are the same, but they are actually
quite different (see, e.g., Tangney, 1996). Guilt arises when a person does something that
harms another person, most typically someone they regard as part of their in‐group. It is
most intense when the victim is a loved one (or when the harm is egregious). Because it is
harm‐related, guilt is an analogue of anger. The harm dimension also means that guilt
arises when a person has performed an action that can threaten an attachment
relationship with another individual (Baumeister et al., 1994). If I harm my friend, I may
lose my friend's affection. So guilt carries with it a threat of loss. It is not surprising then
that the primary action tendencies associated with guilt are reparative in nature. We try
to either apologize, confess, or make amends. When people feel guilty, they are more
likely to be helpful, as if being helpful can compensate for the harm they have done.
Carlsmith and Gross (1969) showed that people are more likely to volunteer for a charity
after they have been asked to administer electric shocks to a stranger. Such acts can even
the moral score card and reduce feelings of guilt. Guilt can also be reduced when an
apology is accepted. That informs the perpetrator that the threatened attachment
relationship has been restored.

Shame differs from guilt in its elicitation conditions and action tendencies. Shame arises
when one performs an action that is perceived as being likely to bring about unwelcome
attention from others. Unwelcome attention isn't just anger or, say, amusement. Shame is
particularly likely when one fears that one will be viewed as a defective person. It is said
that guilt is act‐focused (I did something bad), and shame is person‐focused (I am a bad
person). Consequently shame is likely to arise when an action is seen as rendering one's
self impure, corrupt, or contaminated. It is, thus, an analogue of disgust. We feel ashamed
when we violate sexual norms, or if we cause harms that might be viewed as monstrous.
Like disgust, shame is associated with magical thinking. My shame can contaminate
others. My shameful acts can bring shame to my family.

Simple reparative behaviors, such as confession or apology, cannot alleviate shame very
effectively because these behaviors cannot eliminate feelings of contamination. Instead,
shame is associated with concealment. People lower their heads in shame (Fessler, 2004)
or try to hide. Tangney et al. (1996) shows that guilt increases social health, whereas
shame tends not to because it does not (p. 528) promote reparation. Shame may be
related to embarrassment, but it is more aversive and harder to “save face”. Confession
can help with embarrassment (a blush reveals that I know I did something inappropriate),
but a blush will not suffice for shame. In extreme cases, shame may promote suicidal
behavior, because the ashamed person feels so corrupted that death in the only option.
This has even been culturally ritualized, as with seppuku in feudal Japan. A person who
brings shame may also be killed, as in the case on honor‐killing.

Page 9 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

Though incomplete, this survey of blame emotions is sufficient for illustrating how
important it is to treat moral emotions differentially. They play different functional roles.
This is true even when the emotions belong to the same broad category, such as emotions
of blame. It is plausible that people experience such emotions when they use so‐called
thin moral concepts, immoral or morally bad (Williams, 1985). But on each use of the
concept morally bad, a different emotion might occur depending on what kind of
transgression has been committed and the identity of the perpetrator. Thus “morally bad”
can express different emotions on different occasions.

This discovery suggests that we should be somewhat cautious about embracing Williams'
idea that there are thin moral concepts. Admittedly, the word “immoral” conveys very
little about what kind of norm violation has occurred, in comparison to words such as
“cruel” or “cowardly”. But when someone sincerely describes an action as immoral, there
may be an emotion of blame that arises, and the emotion may serve to identify or delimit
the kind of norm violation that has occurred (Was someone harmed? Am I the culprit?).
There is a thorny issue here about whether concepts should be identified with linguistic
items and their referents or inferential roles in language, or with the underlying
psychology states that occur when linguistic items are used. If the concept immoral is
identified by its linguistic role, then it may qualify as thin, because it is used in so many
contexts that there is little by way of a descriptive common denominator uniting the cases
to which is applies (in sharp contrast to “cruel”). But if the concept immoral is a
psychological state, then individual tokens of that concept may in fact be thick: they may
have content that covers specific kinds of cases and not others. Put differently, there may
be a family of different concepts that we express with the word “immoral” and these may
have contents that could best be expressed using some of the kinds of descriptive
information that Williams associates with thick concepts. A token of the concept immoral
that is realized by feelings of disgust will represent crimes against the body, and a token
that is realized by anger will represent crimes against the autonomy of persons. Now I am
not suggesting that these differences are reflected in mental descriptions (it's rare to
explicitly think, “there has been a crime against the body!”). Rather, each emotion has the
function of occurring in response to different kinds of crimes, and, consequently, they
represent different things. Thus, tokens of these moral concepts may be structurally
simple but semantically rich, and each token has a distinctive motivating feeling and a
(p. 529) circumscribed class of transgressions to which it refers. If this picture is right,
then, when we use words that seem thin, such as “immoral”, we may actually be
expressing psychological states that are both affective and referentially restrictive. In this
respect, the psychological states function more like Williams' thick concepts, even if the
vocabulary used conceals this fact.

Page 10 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

23.2.2 Emotions of Praise

I have been focusing on emotions of blame. These are emotions that arise when we judge
something to be bad or immoral. But there are also emotions of praise, which arise when
we judge and action to be good or moral. Sometimes, a good action is just an action that
it would have been bad not to do. In that case, moral praise involves withholding negative
emotions. But sometimes good actions are supererogatory: they go beyond the call of
duty. In these cases, we may experience emotions that reflect positive attitudes towards
the do‐gooder. Comparatively little research has been done on these emotions of praise,
but I will offer some brief speculations (see also Prinz, 2007).

It does not seem that we experience different emotions of praise for different
praiseworthy actions. For example, it is far from obvious that acts of charity elicit
different emotions than acts of praiseworthy sexual conduct (imagine a society where
pre‐marital abstinence is regarded as a rare and morally praiseworthy achievement). On
the other hand, it does seem to be the case that emotions of praise vary as a function of
who performs the good deed or who receives the benefit. For example, if someone gives
generously to a worthy cause, we might experience admiration. If someone gives
generously to me when I am in need, I will feel gratitude. If I give generously to a worthy
cause, I might feel pride.

These emotions may also be associated with different action tendencies. When we admire
someone, we bestow praise, when we are grateful we express thanks or reciprocate, and
when we feel pride, we may take the opportunity to do ourselves a good turn or we might
even boast or advertise our good conduct. These behaviors have not been well studied,
but certainly warrant investigation.

One interesting fact about moral praise is that we tend not to praise people for simply
conforming to moral norms. Decent behavior is usually expected and goes
uncongratulated. Thus, emotions of praise are typically reserved for cases of extreme
goodness. This suggests that they may be less important than emotions of blame. Social
stability depends on people conforming to norms, not on performing remarkable acts of
charity. A stable moral society can exist without heroes or saints. But we are grateful for
our saints, and we like to promote good deeds, thus we do express positive emotions
when people act nobly. This is even true when the noble act is small or symbolic. For
example, if someone allows me to walk through a doorway first, I will express thanks. We
praise common courtesy. A (p. 530) society without common courtesy might be stable but
it would be less pleasant. Perhaps the main function of positive moral emotions is to
safeguard against adversity. If things get bad, we depend on the charity of others. By
praising acts of supererogatory goodness, we make it rewarding to help those in need. If
conditions worsen suddenly (as when natural disasters strike), we benefit from the
kindness of those who have been less badly affected, and moral praise may play a role in
recognizing and promoting such kindness.

Page 11 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

I say “recognizing and promoting” because emotions of praise may serve a dual function.
They are first and foremost epistemic: they occur in response to some real or imagined
events and they serve to acknowledge that something good has been done. But they also
serve to promote prosocial behavior. The same is true of emotions of blame. I turn to
prosocial emotions now.

23.3 Prosocial Emotions

23.3.1 Blame and Praise Again

Why do we behave morally? One answer is that we are epistemically sensitive to norms.
In many contexts there is a gulf between knowledge and action. One can know that there
is a rule in place (such as a rule against jay‐walking) and not be remotely tempted to
follow it. But the moral emotions of praise and blame close the gap between knowledge
and action.

Consider blame first. If you have an emotion of moral blame, it serves as a way of
recognizing that something was bad. But it can also serve as a cause of good behavior.
This can happen in at least three ways. First, emotions of blame can motivate
punishment, and punishment can promote good behavior by deterrence. Second,
emotions of blame can contribute to moral education. We shape children's behavior when
we get angry or disgusted with them because they desire our affection. Children are also
great imitators and they can catch our negative attitudes and thereby internalize our
norms. Finally, emotions of blame can help us avoid the temptation to be bad. If
contemplating an action makes you anticipate guilt or shame, you may resist that action,
even if carries some obvious rewards.

Emotions of praise are similarly conducive to moral behavior. Receiving praise or thanks
feels good, and positive emotions are behavioral reinforcers. Actions that elicit social
praise are likely to be repeated. Pride also feels good, and actions that make us proud are
rewarding. It feels good to eat a candy bar, but it may feel even better and last longer to
give to charity. Anticipating and experiencing moral pride can increase the probability of
moral conduct.

(p. 531)It follows from this that an emotionally grounded recognition of moral goodness
or moral badness can, in and of itself, promote good behavior. Emotions of praise and
blame serve both epistemic and motivational functions. From a philosophical perspective
that is interesting because of a long‐standing debate about the relationship between
moral judgment and moral motivation. Motivational internalists argue that when we make
moral judgments, we are thereby motivated to act in accordance with moral rules.
Motivational externalists argue that we can make moral judgments without being the
least bit motivated. If emotions of praise and blame are morally motivating, then the

Page 12 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

debate between externalists and internalists may hinge on whether moral judgments have
an emotional component. Emotions of praise and blame could serve the role that
internalist theories require, so internalists would do well to argue that emotions are
intrinsic to moral judgment.

23.3.2 Sympathy, Empathy, and Concern

Emotions of praise and blame are moral emotions in two senses: they play epistemic and
motivational roles. But sometimes we have emotional states that are motivationally moral
but not epistemically moral. That is, they motivate us to do good things even if they do
not constitute recognition of something as morally significant. The dissociation is
important because it leaves room for the possibility that one might be motivated to do the
good without doing it because it's good. Indeed, sometimes morality demands that we act
this way. If all one's acts of charity were motivated by a desire to be moral, they might
seem less noble than acts of charity performed out of a desire to help someone in need,
regardless of the moral significance. Sometimes it is best to help someone because they
need help rather than to help someone because of the fact that helping behavior is
morally praiseworthy. This is particularly true in light of my discussion of emotions of
praise. Emotions of praise feel good when they are received. So if one helps someone
simply because doing do is praiseworthy, then there is a sense in which one's behavior
may be motivated by reward. Such behavior may be self‐serving. But sometimes we may
help because we are moved by the anguish of others, and helping in these cases seems
more genuinely altruistic. In this section, I want to consider altruistic emotional
responses, which can be collectively described as fellow‐feelings.

Consider, first, empathy. “Empathy” can be defined in a number of different ways, but I
will use it narrowly to refer to cases where a person experiences an emotion that another
person experiences (or is thought to experience) as a result of recognizing that emotion.
If I see that you are afraid and feel fear as a result, that is empathy. Empathy is a
vicarious emotional response. Empathy is not itself an emotion but is rather a way that
some emotions come about. Empathetic fear is (p. 532) fear brought on by recognizing or
imagining the fear of another person. In some cases, empathetic emotions might be
brought about automatically, in a bottom‐up way by emotional contagion (Hatfield et al.,
1994). Simply seeing fear can cause fear. This contagion processes is believed to be
underwritten by mirror‐neurons: neurons that are activated both when something is
observed in another and when it is experienced in the observer (Adolphs, 2002). In other
cases, empathy might be driven top‐down by cognitive processes. Reflecting on the
situation of rescue workers, I might realize that they are in danger, and I might come to
experience fear as a result, even if I haven't seen their expressions of fear.

Empathetic emotions are moral emotions insofar as they motivate morally praiseworthy
behavior. It is easy to see why they might contribute in this way. Suppose you are in
danger, and I feel fear as a result. In effect, I feel your fear. This may alert me to your
situation in a very visceral way, and I might form the desire to help you overcome your

Page 13 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

fear. Or, if you are outraged about the way someone has treated you, I may become
outraged too, and be motivated to help you respond to the person who has done you
harm. If emotions help a victim cope with bad circumstances, vicarious emotions may
help observers cope as well, by alerting them to the hardship, making the hardship of
another aversive to the self, and motivating intervention. In some cases, empathy may
lead us to catch others' emotions of moral praise and blame, but empathy can also lead us
to catch emotions of hardship, and, in these cases, we may become motivated to help
even if we have not judged that helping would be a morally praiseworthy thing to do.

Empathy, as defined here, was once called “sympathy” (compare Hume, 1739/1978;
Smith, 1759/2000), but it is now customary to distinguish these two (see, e.g., Darwall,
1998). One way to draw the distinction is to say that empathy refers to any vicarious
emotion and sympathy refers to negative emotions felt on someone else's behalf. Three
differences should be emphasized here. First, one can feel empathetic joy, but not
sympathetic joy, because sympathy is always negative. Second, sympathy is kind of
feeling‐for someone, whereas empathy is feeling‐because (it is causally linked to the
emotion of the person for whom one is empathetic, not semantically linked). If I am sad
that you are sad, then that's sympathy. Third, sympathetic emotions need not be the same
as the emotions experienced by the people for whom we sympathize; I may feel sad that
you are angry or afraid, for example. But like empathy, sympathy involves having a feeling
as a result of someone else's feeling. And like empathy, sympathy is not itself an emotion
but a process by which emotions are brought about and directed. Sympathetic emotions
are characteristically elicited by real or imagined emotions in others and they are
directed at those emotions.

Sympathy may be even more intimately tied to prosocial behavior than empathy.
Empathetic emotions can be alleviated by taking actions directed towards the self. If I
catch your fear by emotion contagion, I can rid myself of fear by ignoring you or leaving
the scene. But if my fear is directed towards your fear, as with sympathy, then leaving the
scene won't necessarily help. As long as I know you are afraid, and as long (p. 533) as my
fear is about that fact, then my fear will last. Helping you escape from the frightening
situation may be the best cure. Sympathy is more overtly other‐related than empathy,
and, consequently, it may lead more reliably to other‐directed behaviors. Moreover,
empathy can lead to a breakdown in helping behavior because the emotions we catch
empathetically can interrupt motivation. If I am scared as a result of your fear, I may flee,
since fear motivates flight. But suppose I have a sympathetic anger response because I
know someone has frightened you. I may elect to intervene rather than flee. When we
acquire emotions sympathetically, the resulting emotions may be more helpful than the
emotions towards which they are directed.

Empathy and sympathy can both be contrasted with a third kind of fellow‐feeling,
concern. Some authors treat “empathy” and “concern” as synonyms. Peter Goldie had
drawn my attention to one clear and egregious example of this; Dan Batson, who has
done some of the most important empirical work in this area, tells us that, “empathic
concern includes feelings of sympathy, compassion, tenderness, and the like” (Batson et

Page 14 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

al., 2007: 65). I think it is helpful to treat concern and empathy separately. One can define
concern as a negative emotion that arises when one thinks or perceives that another
person is in a bad situation. There are three points of contrast to emphasize. First,
concern in an emotion—it may be thought of as a kind of worry or anxiety about another's
well‐being. Second, the party for whom one is concerned need not be experiencing any
negative emotions. One can be concerned for someone who doesn't realize that danger in
immanent, for example. So concern isn't necessarily a matter of feeling‐with someone. Its
focus need not be the emotion of another person but rather the predicament one takes
another to be in.

Concern may be especially conducive to prosocial behavior. If you are worried about
another person, you will be thereby motivated to act on their behalf. This is true even if
the person does not realize there is any threat to her well‐being. Concern motivates
helpfulness even when the victims don't realize they are victims. Also, sympathy is a just
matter of feeling badly for someone, and bad feelings don't always help. If I am sad about
your pain, I may withdraw or sulk. If I say, “You have my sympathy”, I express that I am
sorry that you are suffering, not that I intend to do anything about it. But if I am
concerned I am more likely to intervene. If I say, “I am concerned about you”, that implies
that I want to actively find a way to help you.

In the empirical literature, these different responses are not always distinguished, and
that makes empirical research on fellow‐feelings difficult to interpret. For example,
Neuberg et al., (1997) use a meta‐analysis to show that empathy is not especially
motivating. It leads to moderate prosocial behavior only when there is little cost. But this
conclusion may apply to some forms of fellow‐feeling and not others. Empathy may be
relatively weak as a moral motivator (Prinz, forthcoming), and concern may be relatively
powerful. In the developmental literature, there is evidence that concerned attention, but
not shared emotion, predicts prosocial behavior (Eisenberg et al., 1989). This suggests
that empathy is not a major motivator in children. Indeed, (p. 534) empathy may normally
depend on concern if it is to promote prosocial behavior only when coupled with concern
(or at least sympathy). If I catch your misery, I need to be concerned about you and not
preoccupied with my own well‐being if my empathy is going to motivate me to help you
rather than myself.

All forms of fellow‐feeling are subject to bias. We feel more fellow‐feeling for those who
are perceived to be similar to ourselves (Batson et al., 2005), and fellow‐feeling can lead
to preferential treatment (Batson et al., 1995). But fellow‐feeling is important from a
moral perspective because its existence suggests that we can be motivated to help
someone even if we have not made the moral judgment that helping is good. In the case
of concern and sympathy, the helpful behavior is not driven by self‐interest. Batson (1987)
has shown that, when people are given an opportunity to alleviate their own negative
feelings of sympathy and concern by simply escaping an uncomfortable situation, they
often won't take that easy self‐serving way out. They will instead help the person whose

Page 15 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

well‐being is threatened. On this basis, Batson concludes that human beings have
genuinely altruistic motivations and are not driven solely by self‐interest.

Fellow‐feelings may even be part of our biological make‐up. There is evidence that other
creatures find suffering of conspecifics aversive. In some cases, this may merely be
empathetic distress. For example, rats show distress when they experience the suffering
of other rats (Rice and Gainer, 1962), but they may not be concerned for other rats (Rice,
1964). Evidence for concern is easier to demonstrate in primates. Great apes can be
extremely selfish and ungenerous (Silk et al., 2005), but they also engage in conciliation
behavior (de Waal, 1989) and helping (Warneken et al., 2007). And monkeys will starve
themselves to relieve the agony of another (Masserman et al., 1964).

Such findings do not entail that non‐human animals have a moral sense. Recall that
fellow‐feeling can be independent of moral evaluation. Animals may help each other out
of concern, not out of a desire to do what morality requires. Thus animals may do things
that are morally praiseworthy even if they never engage in moral praise or moral
evaluation. In this respect, moral goodness may precede the capacity to recognize the
morally good. Moral motivation may precede moral epistemology.

23.4 Are There Distinctively Moral Emotions?

I have been arguing that the term “moral emotion” is ambiguous: it can refer to emotions
that are involved in making moral judgments or to emotions that are involved in
motivating morally praiseworthy behavior. I have surveyed some of the (p. 535) main
emotions that play these two roles. But this survey leaves us with one final question: Are
any of these emotions distinctively moral? That is, are there moral emotions that are
dedicated to morality (either moral conduct or moral judgment) and do not serve
important functions outside the moral domain? And if there are some distinctively moral
emotions, are these biologically based, or do they derive from emotions that are not
distinctively moral?

This question can be addressed by reviewing the emotions and emotional processes that
we have been considering. Let's begin with anger. It is sometimes suggested that anger
depends on the judgment that there has been a moral offense (cf. Solomon, 1976). But I
think it is plausible that anger has non‐moral analogues. Some things irritate us without
being regarded as morally wrong. A sound can be irritating as can a difficult and
recalcitrant task. Sometimes we experience an aggressive response without judging that
we've been wronged. This can happen spontaneously if someone glares at you, or it can
be built up intentionally as when athletes compete against an opposing team. I think
these are cases of non‐moral anger, which may be an evolutionary ancient emotional state
that is the basis of moral anger. Moral anger can be described as a kind of aggressive

Page 16 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

irritation that occurs when a moral norm has been violated, especially a norm pertaining
to harm.

Disgust is an even more obvious case. As mentioned, disgust can occur as a response to
contamination and, in this context, it isn't moral. Arguably moral disgust is just physical
disgust as it arises in response to moral norm violations, especially norms conceptualized
as pertaining to the purity of the body.

Above I suggested that contempt might be a blend of anger and disgust, and perhaps
feelings of dominance. All these component emotions are not distinctively or originally
moral. Moreover, contempt itself has non‐moral applications. The rich feel contempt for
the poor, and the poor feel contempt for the rich, but these attitudes do not always
connote any kind of moral assessment.

Disappointment, of course, is just a form of sadness. We can be disappointed about any

event that does not live up to our expectations. These cases are only moral when the
expectation involves conformity to a moral norm.

Guilt and shame appear to be distinctively moral emotions, because we tend not to talk
about them outside of the moral domain. But here too an argument can be made that
these moral emotions derive from emotions that are not moral in nature. I said earlier
that shame might be an aversive form of embarrassment. Guilt may be linked to sadness.
We feel sad about harming people in our in‐group because doing so threatens attachment
relationships. Guilt may also occasionally encompass anxiety because loss of attachment
and punishment often follow when we cause harm. So guilt may be an anxious sadness
elicited by the harms we have caused (Prinz, 2004). If guilt and shame derive from non‐
moral sources, we might expect them to crop up in non‐moral contexts from time to time,
and indeed they do. Survivor guilt occurs when a person is among the few to survive a
calamity that killed members of the in‐group. In some societies shame arises in the
presence of (p. 536) people who have higher social station (Fessler, 2004). In both these
cases, the emotions may be genuine even if no moral judgment has been made.

Fellow‐feelings, I have argued, can occur in the absence of moral judgments, and are thus
not moral emotions in one meaning of the word. They can arise in creatures that make no
moral judgment. From the perspective of such creatures, the role of fellow‐feelings is not
to motivate morally good behavior. These emotions only qualify as moral because we
praise the behaviors they bring about. But they may also promote behavior that is not
especially praiseworthy, such as preferential treatment of the in‐group. Indeed, in non‐
human animals fellow‐feelings are best demonstrated where there are kinship relations or
direct reciprocity. Animals are concerned about their family members and members of
their cooperative groups. Caring for your family is not exactly a morally praiseworthy act;
it's just something we are expected to do. If I do something good for a relative, it's
nepotism, not charity. So fellow‐feelings may originate in contexts that we wouldn't even
characterize in moralistic terms. In human beings, the umbrella of fellow‐feelings

Page 17 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

broadens, and they become unambiguous as moral motivators, but it would be misleading
to call such feelings distinctively moral.

This discussion has been very brief, but the considerations brought to bear should at least
raise the possibility that there are no distinctively moral emotions. Moral emotions seem
either to serve additional functions that are not moral or derive from emotions that are
not moral. If this is right, there are two important implications. First, from an
evolutionary and developmental perspective, it may suggest that morality emerges out of
more general psychological resources (Prinz, 2008). This could mean that morality is not
innate or not modular or not vulnerable to selective deficits.

Second, if no emotions are distinctively moral, this raises a philosophical question about
the very category under consideration. What makes an emotion moral if it is not
intrinsically linked to morality? One possibility is that moral emotions are either ones that
occur as a result of moral judgments or ones towards which our judgments of moral
praise are directed. But this kind of answer implies that we have some purchase on what
moral judgments are independent of moral emotions.

Many philosophers would be happy with that assessment, but some think that moral
judgments must be defined in terms of moral emotions rather than the other way around.
This introduces a threat of circularity. There may be ways to get out of the circle, or at
least to show that it isn't vicious (see Prinz, 2007; Wiggins, 1987). For example, one can
define a moral rule as one that is underwritten by a disposition to experience emotions of
both self‐directed and other‐directed blame (depending on the protagonist). One can
identify these emotions by appeal to the non‐moral emotions from which they derive,
while also saying that they become moral when they converge together to form such
dispositions. The conspiracy of self‐directed and other‐directed emotions of blame is what
makes the (p. 537) resulting attitude qualify as a moral norm, and its status as a moral
norm is what makes these emotions of blame qualify as moral. Whether such a proposal
can work is beyond the scope of this chapter. I hope we have seen, however, that we can
do a reasonably good job of identifying moral emotions, even if there remain
philosophical puzzles and controversies about how to circumscribe the category.

References
ADOLPHS, R. (2002). Recognizing emotion from facial expressions: Psychological and
neurological mechanisms. Behavioral and Cognitive Neuroscience Reviews, 1, 21–62.

BATSON, C. D. (1987). Pro‐social motivation: Is it ever truly altruistic? Advances in

Experimental Social Psychology, 20, 65–122.

—— HÅKANSSON, H. J., CHERMOK, V. L., HOYT, J. L., and ORTIZ, B. G. (2007). An additional
antecedent of empathic concern: Valuing the welfare of the person in need. Journal of
Personality and Social Psychology, 93, 65–74.

Page 18 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

—— KLEIN, T. R., HIGHBERGER, L., SHAW, L. L. (1995). Immorality from empathy‐induced

altruism: When compassion and justice conflict. Journal of Personality and Social
Psychology, 68, 1042–54.

—— LISHNER, D. A., COOK, J., and SAWYER, S. (2005). Similarity and nurturance: Two possible
sources of empathy for strangers. Basic and Applied Social Psychology, 27, 15–25.

BAUMEISTER, R. F., STILLWELL, A. M., HEATHERTON, T. F. (1994). Guilt: An interpersonal

approach. Psychological Bulletin, 115, 243–67.

CARLSMITH, J. M., and GROSS, A. E. (1969). Some effects of guilt on compliance. Journal of
Personality and Social Psychology, 1969, 11, 232–9.

DARWALL, S. (1998). Empathy, sympathy, care. Philosophical Studies, 89, 261–82.

DE WAAL, F. B. M. (1989) Peacemaking among primates. Cambridge, MA: Harvard

University Press.

EISENBERG, N., FABES, R. A., MILLER, P. A., FULTZ, J., SHELL, R., MATHY, R. M., and RENO, R. R.
(1989). Relation of sympathy and personal distress to prosocial behavior: A multimethod
study. Journal of Personality and Social Psychology, 57, 55–66.

FEHR, E. and GÄCHTER, S. (2002). Altruistic punishment in humans. Nature, 415, 137–40.

FESSLER, D. M. T. (2004). Shame in two cultures: Implications for evolutionary approaches.

Journal of Cognition and Culture, 4, 207–62.

GUTIERREZ, R. and GINER‐SOROLLA, R. (2007). Anger, disgust, and presumption of harm as

reactions to taboo‐breaking behaviors. Emotion, 7, 853–68.

HATELD, E., CACIOPPO, J. T., and RAPSON, R. L. (1994). Emotional contagion. New York:
Cambridge University Press.

HUME, D. (1739/1978). A treatise of human nature. P. H. Nidditch (ed.). Oxford: Oxford

University Press.

KANT, I. (1885/1996). The metaphysics of morals. M. J. Gregor (Trans.). Cambridge:

Cambridge University Press

LERNER, J., GOLDBERG, J., and TETLOCK, P. E. (1998). Sober second thought: The effects of
accountability, anger, and authoritarianism on attributions of responsibility. Personality
and Social Psychology Bulletin, 24, 563–74.

(p. 538) MC DOWELL, J. (1985). Values and secondary qualities. In T. Honderich (ed.),
Morality and objectivity (110–29). London: Routledge & Kegan Paul.

MASSERMAN, J. H., WECHKIN, S., and TERRIS, W. (1964) “Altruistic” behavior in rhesus
monkeys. American Journal of Psychiatry, 121, 584–5.

Page 19 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

NEMEROFF, C. and ROZIN, (1992). Sympathetic magical beliefs and kosher dietary practice:
The interaction of rules and feelings. Ethos, 20, 96–115.

NEUBERG, S. L., CIALDINI, R. B., BROWN, S. L., LUCE, C., SAGARIN, B. J., and LEWIS, B. P. (1997).
Does empathy lead to anything more than superficial helping? Comment on Batson et al.
(1997). Journal of Personality and Social Psychology, 73, 510–16.

PARROTT, D. J. and ZEICHNER, A. (2002). Effects of alcohol and trait anger on physical
aggression in men. Journal of Studies on Alcohol, 63, 196–204.

PRINZ, J. J. (2004). Gut reactions: A perceptual theory of emotion. New York: Oxford
University Press.

—— (2007). The emotional construction of morals. Oxford: Oxford University Press.

—— (2008). Is morality innate? In W. Sinnott‐Armstrong (ed.), Moral Psychology, Vol. 1:

The Evolution of Morality: Adaptations and Innateness (367–406). Cambridge, MA: MIT
Press.

—— (forthcoming). Is empathy necessary for morality? In P. Goldie and A. Coplan (eds.),

Empathy: Philosophical and Psychological Perspectives. Oxford University Press.

RICE, G. E. J. (1964). Aiding behavior vs. fear in the albino rat. Psychological Record, 14,
165–70.

RICE, G. E. J. and GAINER, P. (1962). “Altruism” in the albino rat. Journal of Comparative and
& Physiological Psychology, 55, 123–5.

ROZIN, P., LOWERY, L., IMADA, S., and HAIDT, J. (1999). The CAD triad hypothesis: A mapping
between three moral emotions (contempt, anger, disgust) and three moral codes
(community, autonomy, divinity). Journal of Personality and Social Psychology, 76, 574–86.

——, MARKWITH, M., and MC CAULEY, C. R. (1994). The nature of aversion to indirect
contacts with other persons: AIDS aversion as a composite of aversion to strangers,
infection, moral taint, and misfortune. Journal of Abnormal Psychology, 103, 495–504.

——, MILLMAN, L., and NEMEROFF, C. (1986). Operation of the laws of sympathetic magic in
disgust and other domains. Journal of Personality and Social Psychology, 50, 703–12.

SILK, J. B., BROSNAN, S. F., VONK, J., HENRICH, J., POVINELLI, D. J., RICHARDSON, A. S., LAMBETH, S.
P., MASCARO, J., and SHAPIRO, S. J. (2005). Chimpanzees are indifferent to the welfare of
unrelated group members. Nature, 437, 1357–9.

SMITH, A. (1759/2000). The theory of moral sentiments. Amherst, NY: Prometheus Books.

SOLOMON, R. C. (1976). The passions. New York, NY: Doubleday.

Page 20 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018

The Moral Emotions

TANGNEY, J. P. (1996). Conceptual and methodological issues in the assessment of shame

and guilt. Behaviour Research and Therapy, 34, 741–54.

——, WAGNER, P. E, HILL‐BARLOW, D., MARSCHALL, D. E., and GRAMZOW, R. (1996). Relation of
shame and guilt to constructive versus destructive responses to anger across the lifespan.
Journal of Personality and Social Psychology, 70, 797–809.

WARNEKEN, F., HARE, B., MELIS, A. P., HANUS, D., and TOMASELLO, M. (2007). Spontaneous
altruism by chimpanzees and young children. PLoS Biology, 5, 1414–20.

WIGGINS D. (1987). A sensible subjectivism. In Needs, values, truth: Essays in the

philosophy of value (185–214). Oxford: Blackwell.

WILLIAMS, B. (1985). Ethics and the limits of philosophy. Cambridge, MA: Harvard
University Press.

Notes:

(*) I am grateful to Peter Goldie for comments that helped me think more deeply about
issues in this chapter.

Jesse Prinz

Jesse J. Prinz is Distinguished Professor of Philosophy at the City University of New

York, Graduate Center. His research focuses on the perceptual, emotional, and
cultural foundations of human psychology. He is author of Furnishing the Mind:
Concepts and Their Perception Basis (MIT Press 2002), Gut Reactions: A Perceptual
Theory of Emotion (Oxford University Press 2004), and The Emotional Construction
of Morals (Oxford University Press 2009). He also has two forthcoming books: The
Conscious Brain (Oxford University Press) and Beyond Human Nature (Penguin/
Norton). All of these books bring research in the cognitive sciences to bear on
traditional philosophical questions, and, in particular, all defend and extend core
tenets of classical empiricism.

Page 21 of 21

PRINTED FROM OXFORD HANDBOOKS ONLINE (www.oxfordhandbooks.com). © Oxford University Press, 2018. All Rights
Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in
Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: King's College London; date: 16 May 2018