Journal of Personality and Social Psychology
1999, Vol. 76, No. 4, 574-586
Copyright 1999 by the American Psychological Association, Inc.
The CAD Triad Hypothesis: A Mapping Between Three Moral Emotions
(Contempt, Anger, Disgust) and Three Moral Codes
(Community, Autonomy, Divinity)
Paul Rozin and Laura Lowery
University of Pennsylvania
University of Virginia
It is proposed that 3 emotions—contempt, anger, and disgust—are typically elicited, across cultures, by
violations of 3 moral codes proposed by R. A. Shweder and his colleagues (R. A. Shweder, N. C. Much,
M. Mahapatra, & L. Park, 1997). The proposed alignment links anger to autonomy (individual rights
violations), contempt to community (violation of communal codes, including hierarchy), and disgust to
divinity (violations of purity-sanctity). This is the CAD triad hypothesis. Students in the United States
and Japan were presented with descriptions of situations that involve 1 of the types of moral violations
and asked to assign either an appropriate facial expression (from a set of 6) or an appropriate word
(contempt, anger, disgust, or their translations). Results generally supported the CAD triad hypothesis.
Results were further confirmed by analysis of facial expressions actually made by Americans to the
descriptions of these situations.
Mora) judgment and the condemnation of others, including
fictional others and others who have not harmed the self, is a
universal and essential feature of human social life. Many social
animals respond to violations, attacks, or defections against the
self in dyadic relationships (Trivers, 1971), but something seems
to have happened in the evolution of primate social cognition that
makes primates, particularly human beings, chimpanzees, and
bonobos, exquisitely sensitive to violations of the social order
committed by others against others (de Waal, 1996). In these few
species that exhibit what we might call "third-party" morality,
individuals react emotionally to violations, and these reactions
often have long-term effects on social relationships between vio-
lators and third parties. Could these emotional reactions be part of
the foundation of human morality?
Philosophers have long been divided as to whether human
morality is built on our rationality (e.g., Kant, 1789/1959) or our
emotionality (Hume, 1740/1969). Psychological work on morality
has generally focused on rationality and cognitive development
Paul Rozin and Laura Lowery, Department of Psychology, University of
Pennsylvania; Sumio Imada, Department of Psychology, Hiroshima-Shudo
University, Hiroshima, Japan; Jonathan Haidt, Department of Psychology,
University of Virginia.
Laura Lowery is now in independent practice in Berkeley, California.
This research was supported by grants from the Whitehall Foundation
and from the University of Pennsylvania.
Correspondence concerning this article should be addressed to Paul
Rozin, Department of Psychology, University of Pennsylvania, 3815 Wal-
nut Street, Philadelphia, Pennsylvania 19104-6196. Electronic mail may be
sent to email@example.com.
(e.g., Piaget, 1932/1965; Kohlberg, 1969; Turiel, 1983). Moral
development was thought to be driven by the cognitive process of
role taking as the child learns to respect a kind of moral logic (e.g.,
"If I were in her position I would not like this, therefore I should
not do this"). However, since the 1980s, increasing attention has
been paid to the emotional basis of morality. Authors in a variety
of fields have begun to argue that emotions are themselves a kind
of perception or rationality (de Sousa, 1991); that emotions are
embodied thoughts (Rosaldo, 1984); and that "beneath the extraor-
dinary variety of surface behavior and consciously articulated
ideals, there is a set of emotional states that form the bases for a
limited number of universal moral categories that transcend time
and locality" (Kagan, 1984, p. 118; see also Shweder & Haidt,
1993). Cross-cultural work has begun to demonstrate that
cognitive-developmental theories work less well outside of West-
ern middle-class populations and that emotional reactions are often
the best predictors of moral judgments (Haidt, Roller, & Dias,
1993; Shweder, Mahapatra, & Miller, 1987).
We believe that work on the moral emotions has progressed to
the point where we can begin to systematize and taxonomize some
of the moral emotions and relate them in an orderly way to the
structure of the social world. We focus on two principal clusters of
moral emotions that should be of interest to social psychologists,
for they make people care about the social order. The first cluster
of moral emotions is shame, embarrassment, and guilt (SEG), all
of which involve ongoing assessments of the moral worth and fit
of the individual self within a community. These emotions moti-
vate the individual to want to fit in, to behave in a culturally
acceptable fashion, and to avoid harming people. They are self-
focused and are sometimes referred to as the self-conscious emo-
CAD TRIAD HYPOTHESIS
tions (Lewis, 1993). They can be distinguished from each other,
yet they are interrelated (Keltner, 1995; Tangney, Miller, Flicker,
& Barlow, 1996). These emotions may be crucial for human
civilization, for they reflect (or implement) the internalization of
the social order in the individual (Freud, 1900/1976).
The second cluster of moral emotions reflects a similar concern
for the integrity of the social order, but now turned outward to
others. Contempt, anger, and disgust, we argue here, are the three
main "other-critical" moral emotions, a cluster of related but
distinguishable emotional reactions to the moral violations of
others. The present study tests the hypothesis that each of these
three emotions is triggered by a violation of a specific part of the
There are good reasons to focus on contempt, anger, and disgust
as a coherent cluster of moral emotions. Izard (1971, 1977)
grouped these three emotions together as the hostility triad and
found that they were often experienced together in day-to-day
interactions. Furthermore, he noted that they all involve disap-
proval of others. However, there are also differences among the
three emotions. Anger has often been studied in animals and in
humans as a nonmoral emotion, a reaction to frustration or goal
blockage, linked to an action tendency that marshals the resources
required to mount an aggressive response to the blockage. Yet this
"primordial" form of anger, visible throughout the animal kingdom
(Plutchik, 1980), seems to have been elaborated among human
beings into a largely moral emotion. Commentators from Aristotle
(trans. 1941) through Lazarus (1991) have linked anger to insults,
transgressions, and rights violations against the self or those close
to the self.
Similarly, disgust has an animal precursor, called distaste, and it
has a nonmoral primordial form, called core disgust (Rozin &
Fallon, 1987; Rozin, Haidt, & McCauley, 1993). Core disgust is
best described as a guardian of the mouth against potential con-
taminants. However core disgust appears to have been elaborated
into a more complex moral emotion that we call animal nature
disgust in which actions and events that remind us that we are
animals are repressed, hidden, or condemned. Such regulation of
bodily functions, including sex, eating, defecation, and hygiene,
are often incorporated into the moral codes of cultures and reli-
gions (e.g., the Book of Leviticus in the Old Testament), where
they appear to function as guardians of the soul against pollution
and degradation (Rozin, Haidt, & McCauley, in press-a).
Disgust is often further elaborated beyond bodily concerns into
what is called sociomoral disgust (Rozin et al., 1993, in press-b).
For Americans, sociomoral disgust is triggered by a variety of
situations in which people behave without dignity or in which
people strip others of their dignity. Miller (1997) suggested that
disgust is the principal emotion that responds to the vices of
hypocrisy, cruelty, fawning, and betrayal. Sociomoral disgust is
often triggered by third-party violations that may not directly
affect the self. For example, when we asked American adults to
describe three events in which they felt disgust, many responses
involved hearing about sociomoral violations such as racism and
child abuse (Haidt, Rozin, McCauley, & Imada, 1997). The same
was true for a sample of Japanese adults, although the nature of the
sociomoral events differed from the American events. And the
heavily moralized usage of the English word disgust does not
appear to be a quirk of the English language; it occurs as well in
French, German, Hebrew, Russian, Japanese, Chinese, Oriya (an
Indian language related to Hindi), and many other languages
(Haidt et al., 1997).
Contempt differs from disgust and anger in that it does not have
a clear animal origin. However, like the moral forms of anger and
disgust, contempt is usually said to involve a negative evaluation
of others and their actions. Contempt is often linked to hierarchy
and a vertical dimension of social evaluation. Izard (1977) noted
that contempt is often felt by members of one group for members
of other groups regarded as inferior and that it is therefore impor-
tant in prejudice and racism. Izard added that contempt is the most
subtle and coldest of the three emotions in the hostility triad.
Ekman (1994) expressed a similar view of contempt as disapprov-
ing of and feeling morally superior to someone. Miller (1997)
called contempt a close cousin of disgust, which works with
disgust to maintain social hierarchy and political order. Like Izard,
Miller characterized contempt as cooler than disgust, because it
involves an element of indifference toward the object of contempt.
We are not suggesting that these two clusters of moral emotions
(CAD and SEG) exhaust the list of moral emotions. There is a third
important cluster, which might be called the other-suffering emo-
tions, for it includes emotions triggered by the suffering of others,
such as pity and sympathy. These emotions were at the heart of the
moral theories of David Hume (1739/1969) and Adam Smith
(1759/1966), and they have been well studied in modern times
(e.g., Eisenberg, 1989; Hoffman, 1987). Furthermore, many other
emotions can play a role in moral behavior and cognition—for
example, fear (of punishment) and love or admiration (toward
moral exemplars). However, we think that the two clusters of CAD
and SEG are particularly rich and important because they are so
closely tied to the internalized respect for an external social order.
Mapping the Moral Domain
We believe the time is right to map out the moral domains of
contempt, anger, and disgust because of the appearance of an
important new theory of morality (Shweder, Much, Mahapatra, &
Park, 1997). Shweder and his colleagues proposed that there are
three distinct ethics that cultures use to approach and resolve moral
issues: the ethics of community, autonomy, and divinity. Each
ethic is based on a different conceptualization of the person: as an
office-holder within a larger interdependent group-family-
community (community), as an individual preference structure
(autonomy), or as a divine creature bearing a bit of God within
(divinity). Depending on which of these three views one holds of
the person, a different set of moral goods and obligations becomes
paramount. To summarize the three ethics, we present here the
exact descriptions we provided to participants in Study 2, de-
1. [The ethics of Autonomy] Individual freedom/rights violations. In
these cases an action is wrong because it directly hurts another person,
or infringes upon his/her rights or freedoms as an individual. To
decide if an action is wrong, you think about things like harm, rights,
justice, freedom, fairness, individualism, and the importance of indi-
vidual choice and liberty.
2. [The ethics of Community] Community/hierarchy violations. In
these cases an action is wrong because a person fails to carry out his
or her duties within a community, or to the social hierarchy within the
community. To decide if an action is wrong, you think about things
CAD TRIAD HYPOTHESIS
some AUs occurred for the most effective situation in less than
20% of participants.
At the bottom of Table 8, we list five additional AUs that showed
a substantial relation to one of the moral codes. These are AU 2 (outer
brow raise) for community violations, AU 24 (lip press, somewhat
similar to AU 23, lip tighten), and AU 38 (nostril dilate) for auton-
omy, and AU 6 (cheek raise) and AU 20 (lip stretch) for divinity.
We began this article with a proposal that organized the emo-
tions often linked to the moral domain. We used this organization
to offer a hypothesis about the relation between what are desig-
nated "other-critical" moral emotions and a taxonomy of morality
that may have universal validity as proposed by Shweder et al.
(1997). The CAD triad hypothesis holds that three other-critical
moral emotions map rather cleanly onto three different moral
codes. In this article we presented evidence that supports this
hypothesis, using emotion words, recognition of facial expressions
of emotion, and actual facial expressions of emotion. The evi-
dence, from students in the United States and Japan, supports the
predictions of the hypothesis; that is, it suggests that one aspect of
the organization-appraisal-meaning of the other-critical moral
emotions has to do with something like the Shweder moral codes.
We consider this a first attempt at organizing the moral emotions
and matching some of them in a systematic way to moral principles.
The results are promising, not in the sense that this is a total account
of the moral aspects of the other-critical moral emotions but that this
may account for some of the relation between these emotions and
moral issues. We have considered one possible confound, the degree
of negativity of a violation, which separates most community-
contempt violations, at least among Americans, from anger and dis-
gust violations. Our data indicate an overall weaker negativity for
community as opposed to divinity or autonomy violations. Insofar as
this can be considered a confound, it is insufficient to account for the
triad pattern we report. Furthermore, there is no reason to think that
violations of one or another moral code, in general or in any particular
culture, would be judged to be of equal seriousness.
There is quite a bit of consensus in our results, for both face and
emotion words as applied to situations, and this consensus extends
across two different cultures. Our biggest problem in the interpretation
of results has to do with application of the three moral codes to
specific moral situations. We encountered more difficulty than we
anticipated in this area, both within and between cultures. Looking
back on this study, we would be inclined to do a more careful job of
situation selection and of shaping of situation to specific cultural
frames. We are chastened by the difficulty of selecting appropriate
situations; the conceptual clarity of the Shweder codes does not easily
translate into unambiguous interpretations Of real moral situations.
This is not a criticism of the validity of the Shweder analysis, which
we believe to contain something that is deeply true. Rather, it is a
testimony to the multiple construals of which the human mind is
capable. Thus, within culture, there was far from consensus on the
moral categorization of many of the violations, and we have ourselves
had different moral construals of a few of the original 46 items. We
consider some of the problems in the presentation of the results of
Study 2: One is simply that the concept of one's rights or freedom
varies across individuals and cultures. Do they extend to one's family,
one's ancestors, one's body after death, one's soul? The same can be
said for the construal of harm, which can be projected in the same way
as rights (to family, soul, etc.).
We used participants from two cultures primarily as a way to
test the generality of the CAD hypothesis. However, there are
differences between Japan and the United States that might be
expected to be manifested in our data. The greater focus on a
communal self in Japan (Markus & Kitayama, 1991) and the
higher respect for social hierarchy should lead to a greater salience
of the community ethic in Japanese participants. In fact, the
communal ethic was endorsed, across all 27 situations, slightly and
nonsignificantly less by Japanese participants (30.0% of total
endorsements) than by American participants (33.7%).
We acknowledge three shortcomings in the design that limit the
conclusiveness of these studies. The first is the oft-mentioned
problem of the unstable or at least highly variable mapping of
situations into Shweder moral codes. The second is the particular
problem of disgust as a moral emotion, that is to say, in terms of
the CAD triad hypothesis, the low salience of divinity as a moral
matter in the two cultures we examined. As we expand our
understanding of disgust as a moral emotion (Miller, 1997; Rozin
et al., in press-b), it may become easier to craft more appropriate
divinity violations. The third is the constraints of a four-choice
response mode in the first two studies (three emotion or moral
CAD choices and "no appropriate response"), which might be
relaxed to provide more definitive results.
Further research can be expected to clarify and test the CAD triad
hypothesis. This would include getting some control over the vari-
ability in classification of the moral codes in two manners. First, a
more careful development of situations and a more thorough moral
code rating procedure are needed. Some progress has already been
made along this line in recent research relating to emotional expres-
sion and the Shweder codes comparing Filipinos and Americans
(Vasquez, Keltner, Ebenbach, & Banaszynski, 1998). Second, use of
a within-subject design in which the same participant would classify
situations by moral codes and assign words or faces to them would
provide valuable data. In addition, empirical results should be gath-
ered from individuals other than college students and from individuals
from other cultures. Rural Hindu India would be of particular interest,
because the salience of the moral codes is very different from that of
the United States (Shweder et al., 1987; Shweder et al., 1997).
It would also be of interest to attempt a similar type of analysis
for the self-conscious moral emotions of shame, embarrassment,
and guilt (or SEG), leading perhaps to a more complete CAD-SEG
hypothesis. So far we have not been able to find a way to satis-
factorily assign these emotions in a systematic way to the three
moral codes. The task is also more difficult because these emotions
do not have facial or bodily expressions that are as distinctive as
those for contempt, anger, and disgust. However, recent work has
advanced our understanding of the expression of these self-
conscious moral emotions (Keltner, 1995), and recent work in the
United States and Hindu India has begun to clarify the moral status
of these emotions (Haidt & Keltner, in press).
In conclusion, we think the CAD triad hypothesis offers a new
and useful way of thinking about the moral emotions and the
emotional basis of the social order. The hypothesis makes testable
predictions, which have found support in the present study and in
other recent research (Vasquez et al., 1998). More generally, we
hope that the CAD triad hypothesis and the results we present
reinforce the current resurgence of interest in affect. We also hope
ROZIN, LOWERY, IMADA, AND HAIDT
that this work helps to emphasize that the human moral world
involves strong feelings as well as reasoning and that there are
universal and culture-specific linkages between the affective and
cognitive aspects of morality.
Aristotle (1941). The basic works of Aristotle (R. McKeon, Ed., & J. I.
Beare, Trans.). New York: Random House.
de Sousa, R. (1991). The rationality of emotion. Cambridge, MA: MIT
de Waal, F. B. M. (1996). Good natured. Cambridge, MA: Harvard
Eisenberg, N. (Ed.). (1989). Empathy and related emotional responses. San
Ekman, P. (1994). Antecedent events and emotion metaphors. In P. Ekman
& R. J. Davidson (Eds.), The nature of emotion: Fundamental questions
(pp. 146-149). New York: Oxford University Press.
Ekman, P., & Friesen, W. V. (1971). Constants across cultures in the face
and emotion. Journal of Personality and Social Psychology, 17, 124—
Ekman, P., & Friesen, W. V. (1978). Facial action coding system: A
technique for the measurement of facial movement. Palo Alto, CA:
Consulting Psychologists Press.
Ekman, P., & Friesen, W. V. (1986). A new pan-cultural facial expression
of emotion. Motivation and Emotion, 10, 159—168.
Freud, S. (1976). The interpretation of dreams. In J. Strachey (Ed. and
Trans.), The complete psychological works. New York: Norton. (Orig-
inal work published 1900)
Haidt, J., & Keltner, D. (in press). Culture and facial expression: Open-
ended methods find more faces, and a gradient of recognition. Cognition
Haidt, J., Roller, S. H., & Dias, M. G. (1993). Affect, culture and morality,
or Is it wrong to eat your dog? Journal of Personality and Social
Psychology, 65, 613-628.
Haidt, J., Rozin, P., McCauley, C. R., & Imada, S. (1997). Body, psyche,
and culture: The relationship between disgust and morality. Psychology
and Developing Societies, 9, 107-131.
Hoffman, M. L. (1987). The contribution of empathy to justice and moral
judgment. In N. Eisenberg & J. Strayer (Eds.), Empathy and its devel-
opment: Cambridge studies in social and emotional development (pp.
47—80). New York: Cambridge University Press.
Hume, D. (1969). A treatise of human nature. London: Penguin. (Original
work published 1739 & 1740).
Hunter, J. D. (1991). Culture wars: The struggle to define America. New
York: Basic Books.
Izard, C. E. (1971). The face of emotion. New York: Appleton-Century-
Izard, C. E. (1977). Human emotions. New York: Plenum.
Jensen, L. A. (1997). Culture wars: American moral development across
the lifespan. Journal of Adult Development, 4, 107-121.
Kagan, J. (1984). The nature of the child. New York: Basic Books.
Kant, I. (1959). Foundations of the metaphysics of morals (L. W. Beck,
Trans.). Indianapolis, IN: Bobbs-Merrill. (Original work published
Keltner, D. (1995). Signs of appeasement: Evidence for the distinct dis-
plays of embarrassment, amusement, and shame. Journal of Personality
and Social Psychology, 68, 441-454.
Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental
approach to socialization. In D. A. Goslin (Ed.), Handbook of social-
ization theory and research (pp. 347-480). Chicago: Rand McNally.
Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford Uni-
Lewis, M. (1993). Self-conscious emotions: Embarrassment, pride, shame,
and guilt. In M. Lewis & J. Haviland (Eds.), Handbook of emotions (pp.
563-573). New York: Guilford Press.
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications
for cognition, emotion and motivation. Psychological Review, 98, 224-
Miller, W. I. (1997). The anatomy of disgust. Cambridge, MA: Harvard
Piaget, J. (1965). The moral judgement of the child (M. Gabain, Trans.).
New York: Free Press. (Original work published 1932)
Plutchik, R. (1980). Emotion: A psychoevolutionary synthesis. New York:
Harper & Row.
Rosaldo, M. (1984). Toward an anthropology of self and feeling. In R.
Shweder & R. LeVine (Eds.), Culture theory (pp. 137-157). Cambridge,
England: Cambridge University Press.
Rosenberg, E., & Ekman, P. (1995). Conceptual and methodological issues
in the judgment of facial expressions of emotion. Motivation and Emo-
tion, 19, 111-138.
Rozin, P., & Fallon, A. E. (1987). A perspective on disgust. Psychological
Review, 94, 23-41.
Rozin, P., Haidt, J., & McCauley, C. R. (1993). Disgust. In M. Lewis & J.
Haviland (Eds.), Handbook of emotions (pp. 575-594). New York:
Rozin, P., Haidt, J., & McCauley, C. R. (in press-a). Disgust. In M. Lewis
& J. Haviland (Eds.), Handbook of emotions (2nd ed.). New York:
Rozin, P., Haidt, J., & McCauley, C. R. (in press-b). Disgust: The body and
soul emotion. In T. Dalgleish (Ed.), Handbook of cognition and emotion.
Chichester, England: John Wiley.
Rozin, P., Haidt, J., McCauley, C. R., & Imada, S. (1997). The cultural
evolution of disgust. In H. M. Macbeth (Ed.), Food preferences and
taste: Continuity and change (pp. 65—82). Oxford, England: Berghahn.
Rozin, P., Lowery, L., & Ebert, R. (1994). Varieties of disgust faces and
the structure of disgust. Journal of Personality and Social Psychol-
ogy, 66, 870-881.
Russell, J. A. (1991). Negative results on a reported facial expression of
contempt. Motivation and Emotion, 15, 281-291.
Russell, J. A. (1994). Is there universal recognition of emotion from facial
expression? A review of the cross-cultural studies. Psychological Bul-
letin, 115, 102-141.
Shweder, R., & Haidt, J. (1993). The future of moral psychology. Truth,
intuition, and the pluralist way. Psychological Science, 4, 360-365.
Shweder, R. A., Mahapatra, M., & Miller, J. (1987). Culture and moral
development. In J. Kagan & S. Lamb (Eds.), The emergence of morality
in young children (1-83). Chicago: University of Chicago Press.
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. Brandt & P. Rozin (Eds.),
Morality and health (pp. 119-169). New York: Routledge.
Smith, A. (1966). The theory of moral sentiments. New York: Kelly.
(Original work published 1759)
Tangney, J. P., Miller, R. S., Flicker, L., & Barlow, D. H. (1996). Are
shame, guilt and embarrassment distinct emotions? Journal of Person-
ality and Social Psychology, 70, 1256-1269.
Trivers, R. L. (1971). The evolution of reciprocal altruism. Quarterly
Review of Biology, 46, 35—57.
Turiel, E. (1983). The development of social knowledge: Morality and
convention. Cambridge, England: Cambridge University Press.
Vasquez, K., Keltner, D., Ebenbach, D. H., & Banaszynski, T. L. (1998).
Morality and culture: Cultural variation and similarity in moral rheto-
rics. Unpublished manuscript, University of California at Berkeley.
Received September 15, 1997
Revision received October 15, 1998
Accepted October 26, 1998 •