PART II
CHAPTER FOUR
Over the last 200 years in the history of modern societies, shame has virtually disappeared. The denial of shame has been institutionalized in Western societies.
—THOMAS SCHEFF, “SHAME IN SOCIAL THEORY”
IN THIS PART of the book, I ask the same set of questions I asked about anger in the first part. What work does shame do in morality? What work ought shame to do in morality? Is shame an outdated emotion, one we would be better off without? What are the costs of using shame to enforce norms? How does shame differ from guilt? Is shame always outer, the emotion that results from the real or imagined scrutiny of others, the disapproving gaze, whereas guilt is always inner, the voice of a stern parent metabolized eventually into one’s own conscience? Is there something about the nature of shame that causes antisocial responses—withdrawal or “humiliated fury” (Lewis 1971)? Is there something about the nature of guilt that yields dispositions to apologize and make reparations? Are there really shame cultures and guilt cultures? What are the arguments that favor the use of one or the other in inculcating and enforcing a morality? Is there some third emotion that is neither shame nor guilt, perhaps some more domesticated hybrid that could be substituted for either or both if we judge the costs of standard-issue shame and guilt to be too high? What do psychology and anthropology teach us about cultural variation in the nature and function of guilt and shame?
As before, I claim that examining a range of different conceptions of what shame and guilt are, what they can do, their functional roles in different cultures and subcultures, and their costs and benefits can be enormously helpful in considering how we—where the “we” is pegged to some regional, political, socioeconomic status, religious, educational, ethnic, gendered, age, or other cohort—do shame and guilt, and what we might, upon reflection, want to do differently with these emotions. My aim is to mount an argument in favor of an appropriately targeted, mature sense of shame. The defense matters because we live in shameless times. There is the shameless hypocrisy of simultaneous commitment to the Golden Rule and to disrespecting out-group members; there is the shameless disregard for truth-telling, and shamelessness about selfishness and greed.
One set of obstacles to making the argument that these things are shameful involves distinctively American views about the priority of liberty even over truth, love, and solidarity, so that, in effect, it is an individual right to be a liar, a misanthrope, or an egomaniac. Another obstacle to making the argument in defense of shame is the widespread view that shame is a primitive, easily distorted emotion, generally useless for advancing personal, interpersonal, and moral good.
In certain respects, the situation is hard to read. Some psychologists find that guilt and shame are used almost as synonyms by American college students (Tangney et al. 1996). Others find, especially when subjects are asked to theorize emotions, that WEIRD people prefer not to normatively certify any emotion that authorizes others to judge them or their behavior. The cultural psychologist Batja Mesquita writes:
Shame is “wrong” in WEIRD cultures. European Americans have all but banned shame out of their everyday lives. When do you hear someone talk about being ashamed? Feeling “funny,” maybe; but ashamed? Not much chance.… In WEIRD cultures, we do not like to imagine being dependent on others, least on their judgment: We should be feeling good about ourselves, independent from others. And because of this, shame itself becomes uncomfortable. Shame is a taboo emotion. Instead of acknowledging their shame, European American individuals will feel “uneasy,” maybe “a bit distracted,” they will stammer, play with their hair, but they won’t say, or admit to themselves, that they feel shame. In European American communities, shame itself is shameful. When I did the interviews [for her research], I found that shame was the hardest emotion to talk about for North American respondents. Quite a few, especially in the non-student European American sample, told us that this type of event had never happened to them. (2022)
Shame has an undeserved bad rap in WEIRD countries. This bad rap has nothing, exactly zero, to do with any natural features of shame as a complex social emotion, and entirely to do with bad values. To put it bluntly, shame’s bad rap has to do with appalling values that result in people being taught or encouraged to be ashamed of things that no one should ever be ashamed about—the color of their skin, their sexual orientation, their gender, their whole self. The badness of these kinds of shame has to do with their terrible content, not with shame itself. The problem in these cases is with the values that shame is recruited to endorse and protect. Shame about breaking promises, or about being rude to service workers, or for being a malicious gossip, or shame at being a racist or a sexist are all fine, because one ought to feel ashamed of these things.
The next four chapters proceed this way: In this chapter, I explain what shame is and begin the argument against the bad rap. In chapter 5, I argue specifically against what the science of shame says about shame, which, if true—but it isn’t—might support the bad rap. Then in chapter 6, I examine some of the ways that shame works in other cultures to advance the project of appreciating shame, and, finally, in chapter 7, I lay out the defense of shame as an ideal protector of deep value commitments.
Shame is an emotional instrument that can be used to teach and protect values. Good shame is the kind of shame that starts out feeling bad but is eventually autonomously endorsed as a positive self-monitoring emotion. The emotion or, more likely, the complex affective disposition is the mature sense of shame.
What Is Shame?
What is shame? It depends on how a culture defines it. How does shame differ from guilt? It depends on whether and how a culture marks the difference. But, so as not to violate ordinary usage, I’ll stipulate that, for my purposes, shame in its basic form is the emotion that normally has the following properties:
· Shame is one kind of bad feeling (first pass, “bad” simply means it has negative valence) that one has when one is judged by actual others or by oneself to have violated an accepted, and eventually first personally endorsed, social and/or moral norm.
· The distinctive feeling of shame, insofar as it has a distinctive feeling, is commonly reported to involve some combination of feeling embarrassed, fearful, anxious, and sad about being judged to have done something, or to possess a trait or characteristic, that is disvalued or disapproved of. Note: the distinctive feeling, the qualitative character of shame, involves the metaphorically described inner part plus the fact that the feeling is about having done something that is disvalued or disapproved of, or about possessing a trait that is disvalued or disapproved of. Shame marks the possibility of social exclusion. It is, as Heidi Maibom says, “a profoundly social emotion” (2010, 567). Cheshire Calhoun puts the basic idea this way: “Shaming moral failures are paradigmatically ones that might, if exposed, reduce one’s social standing in some actual group or might degrade the quality of one’s social interactions” (2004, 130).
· Genealogically, shame is initially a group-based, as opposed to an agent-based, emotion (Maibom 2010). The norms we are called upon to feel shame for violating are never, at least at first, of our own making.
· Whereas guilt typically focuses on acts, deeds, or doings that are disapproved of, shame typically focuses on aspects of the person that are personal or characterological weaknesses that, in many cases, are judged to cause bad acts, deeds, or doings.
· But there is no necessity here. Often, in Anglophone countries, “shame” and “guilt” are used synonymously (Tangney et al. 1996; Boiger, De Deyne, Mesquita 2013; Boiger et al. 2013; Henrich 2020; Mesquita 2021). It is commonplace for children to be told that they should be ashamed of doings and deeds. It is actually rare for a young child to be told that he ought to feel guilty for taking more than his fair share of M&Ms. The more natural way of speaking is to say that he ought to be ashamed.
· The initiator of shame (caretakers at first) follows cultural norms and scripts that specify what kinds of acts, traits, and dispositions are shameful, the kind and amount of shame warranted, and how the shame is expressed and enacted. The object of shame (children at first) learns how to avoid shame at the same time they learn what shame is: don’t do that again, avoid detection, extinguish the bad disposition, self-cultivate, self-modify.
· Shame can be initiated or enacted by others, as when a mother says a child ought to be ashamed for not sharing his M&Ms with his sister. This is being shamed. Its aim is to produce feeling ashamed and to change future behavior. “Being shamed” and “feeling ashamed” both come in degrees and are not connected to each other by conceptual or empirical necessity. Disappointment, disapproval, and anger are not typically forms of shaming, but they are common invitations for the child to be ashamed. The case of the mother saying to the child who is not sharing his M&Ms with his sister that he “ought to be ashamed” is simple, benign shaming. There is also malicious shaming and humiliation, which is cruel, an overreaction rarely proportional to the norm violation.
· The initiator of shame is not normally ashamed. Although they might be if, for example, their child’s behavior reflects negatively on them by revealing deficiencies in how they raised the child, in the kind of parent they are.
· Shame can be initiated by and enacted upon oneself, as when one fails at one’s resolve to keep a pledge to oneself, not to drink, for example. Agent-based, for one’s eyes only, shame increases over time as values are internalized.
· Across all cases of shaming and feeling ashamed, actions are entailed and enjoined. The child who shrinks away after being told she has done something shameful is doing what comes naturally when one feels shame, perhaps in this limiting case, by nature’s design. The adult who judges herself to be shameful for failing to keep her resolution is moved to try something different, to step up her game, to find a work-around in order to align her behavior with her resolve.
· The mature sense of shame involves endorsing as one’s own norms, values, and virtues that were originally not one’s own. It is agent-based and autonomous.
Shame: Heteronomous or Autonomous?
This is what basic shame is and what, in the last step, it can be transformed into—a mature sense of shame. Two wise theorists of shame, Cheshire Calhoun (2004) and Heidi Maibom (2010), point out that we are born into communities in which the conventions, norms, values, and virtues are already in place. We did not choose them. This is true. They conclude that this makes shame about normative violations heteronomous, not autonomous. This is false.1
All normative learning starts with heteronomous value. No one enters the world knowing much about its ways. But values that are at first accepted purely on the authority of others can come to be accepted autonomously. The situation is identical for shame and guilt. Guilt is encouraged for bad acts but not for acts that a child initially sees as bad herself. She did not choose to be born into a world in which she must share M&Ms with her sister. But eventually she comes to endorse the values that she did not invent or choose. The point is simply that an emotion that begins as a protector of heteronomous values needn’t remain heteronomous.
Maibom concludes her excellent analysis of the evolutionary descent of shame in the animal world by writing: “Primarily concerned with the opinion of others, shame is heteronomous. It cannot be the case both that morality is an essentially autonomous practice and that shame plays an important role in it” (2010, 590). But this does not follow if what starts as heteronomous values, and even what starts as a heteronomous emotion, can be reflectively self-authorized, internalized, and cognitively and affectively self-sustained. I love to play tennis, but everything about the game, and the norms and scripts that govern playing it, are heteronomous. The game was invented six centuries prior to my conception, and pretty much all I bring to it is that I am a featherless biped.
Calhoun pays special attention to the fact that sexism and racism can make one feel ashamed to be a woman or a person of color, even when one knows reflectively that there is nothing whatsoever to be ashamed about. She explains, what is almost certainly true, that this is possible because, as gregarious social animals, we are extremely vulnerable to positive and negative social messages about perceived statuses. She then concludes that shame “is not the emotion of a critical, normatively reflective, autonomous agent” (2004, 145).
This, too, is one step too far and too fast. We don’t work for shame. Shame is something that works for us, an emotional syndrome whose shape and function(s) is determined by us, collectively. There is nothing about shame as such that keeps us from enacting it in a “critical, normatively reflective, autonomous” way. That said, so long as powerful people hold terrible values, many of us will feel entirely unwarranted shame. The long-term solution is to turn the tables by advancing pride movements while simultaneously teaching that racism and sexism are something to be ashamed about.
Shame and the Whole Self?
In addition to its heteronomy, it is important to mark another feature that shame is thought to have but needn’t have, and that in fact it does not have nor is theorized to have except in WEIRD countries, possibly only in America or North America or, even more parochially, only in Anglophone precincts of North America. This is the alleged feature of taking the whole self as its evaluative object and judging the entire self to be a failure (Sedgwick and Frank 1995; Wong and Tsai 2007). That is, it is the received view that shame asks an individual to feel bad about their global self.
But, in fact, shame doesn’t normally do this. The mistake comes in part from misunderstanding the grammar of sentences like “You ought to be ashamed of yourself.” There are extreme cases where the whole self is really being globally disdained; normally, however, context shows that a person is not being asked to be ashamed of their global self but, at most, about some global trait they possess—dishonesty, sloppiness, or lack of conscientiousness at work.
I just said shame doesn’t normally do this. There are exceptions. First, there are some abusers who wish to make others believe that everything about them is wrong. This is loathsome, and we describe it best in words of the pathology of sadism. Second, there are varieties of racism, sexism, queerphobia, and ableism that really do take an essential, identity-constituting feature of a fellow human being and deem that essential feature as worthy of disdain and shame. There are about a million things wrong with racism, sexism, queerphobia, and ableism, one of which is that shame is weaponized to serve these evil masters. But the abuse of shame is not the main problem we need to aim to resolve. The psychological, historical, sociological, political, and institutional causes of racism, sexism, queerphobia, and ableism must be addressed directly and insistently, every day in every way, by people of goodwill. When the causes of these forms of disdain, fear, and hatred are rendered impotent, shame will have no place to take root.
Generic Shame
The picture of shame abides in the view of emotions as functional syndromes, which I proposed in the introduction. Generic shame has this structure:
1. The child is invited to feel bad in shamelike way(s), which involves feeling scared, anxious, and sad, or these plus embarrassment about being judged to have done something wrong, and/or to possess a trait or characteristic that is disvalued or disapproved of that is thought to be connected to doing the wrong thing.2 Shame marks the possibility of social exclusion if the child does Φ in situation S, where Φ names a violation of a social or moral norm and protects some value v. Feeling bad in a shamelike way is designed to inhibit future acts of Φ type, as well as traits, habits, or dispositions that might dispose the child to Φ.
2. The mature person accepts that Φ-ing in S, or being disposed to Φ in S, violates the social and/or moral order and embraces the value system V (v1, v2, … vn) that makes it so. S (situation) matters because Φ or dispositions to Φ can be fine in some situations but not in others. Lying in a play is fine; lying in real life is shameful. Talking is fine, good, necessary; talking during a wedding or a funeral is shameful. Once a person accepts, avows, and internalizes a set of values that set standards for things that they would feel ashamed to do in their own eyes, as well as things they would feel ashamed not to do in their own eyes, they have a mature sense of shame in the formal sense.
The first step focuses on generic shame from the side of the caretaker, who invites the child to be ashamed. The object of shame—in the generic version, the child—learns how she is supposed to respond to shame judgments. There are complex socially constructed norms for how to do shame on both sides, that of the giver and that of the receiver. Usually, the giver is not experiencing or expressing shame but anger, disappointment, or disapproval. Although sometimes they might also feel shame—for example, in situations where their child misbehaves in a restaurant or other public space. In the second step, the mature person knows when to make shame judgments of others or themselves to protect value.
In both steps 1 and 2, the person experiencing shame is acting. They are perhaps, at first, more scared than ashamed (because the caretaker is angry; normally, not herself ashamed), and disposed to stop what they are doing, either in shock at the caretaker’s anger, or—if they get that the caretaker is judging what they are doing or disposed to do as shameful—disposed to hide, shrink away, or otherwise indicate appeasement by way of conformity to the will of the more powerful. At the mature stage, the community’s intention is that the person has inspected and approved of the value system V that makes certain actions in certain situations shameful, and comes to naturally abide the normative order, works to sand its rough edges, and works on himself as need be.
My view is that judgments about the badness of shame are almost never judgments about the intrinsic badness of expressing and receiving judgments of social disapproval or about informing others (the objects of shame) of the possibility of punishment or social exclusion. This is true even if cultured shame maintains some features of evolutionary, ancient protoshame, such as deference and appeasement. Negative judgments about shame are about how we fill in the variables Φ, about the judgments that deem Φ-ing in S, or being disposed to Φ in S as bad, or about the quality of some value v or the overall value system V that shame is called upon to protect. Shame can go wrong in four spots: Φ, S, v, V.
If the values that shame enacts or protects are bad, the shame is bad. Mature generic shame, in the formal sense, can protect good values or bad ones, because it is only formal. If the values that shame protects are good, accurately assesses an individual’s disposition to Φ, calls attention to why not Φ-ing in S is good by indicating the way not Φ-ing protects v and V, is proportional to the importance of the value (v) being protected, then the main consideration that remains is whether some other disciplinary emotion that doesn’t feel bad, or feels less bad, or feels bad in a different way—such as guilt, sadness, or fear and trembling—might substitute for shame in step 1 or in the internalized sense of shame achieved in step 2.
The difference between shame that is careless with the truth or frivolous with promises—or that treats other people as a mere means to your ends—and shame about being a minority is that the first serves good values and the second serves terrible values. Shame isn’t the problem; the value it is deployed to defend is the problem. If this is right, then it opens the conversation to reconsidering shame as a useful disciplinary emotion, thus saving it from its undeserved bad rap.
Shame: The Bad Rap
We are social animals. We have numerous ways of influencing and shaping each other’s hearts and minds, and making social relations reliable, orderly, and harmonious. Uptake by others is of maximal importance to a good human life, whereas disapproval, downgrading, and social exclusion are extremely painful. In an important sense, we do not really even know what we are doing until others read our actions for what they are. Simone de Beauvoir explains the necessity of parental uptake to make a child’s act—in her example a drawing—the act that it is:
As soon as a child has finished a drawing or a page of writing, he runs to show his parents. He needs their approval as much as candy or toys; the drawing requires the eyes that look at it.… These disorganized lines must become a boat or a horse for someone. So the miracle is accomplished, and [the child] proudly contemplates the multicolored paper. From then on there is a boat, a real horse there. By himself, he would not have dared to put confidence in those hesitant lines. (2005, 116)
A child or adult who doesn’t receive uptake eventually feels alienated and unwelcome. Loving, attentive parents, siblings, grandparents, aunts and uncles, neighbors, and parents of playmates uplift us, teach us what we are doing and how to be a person. We also fear their disappointment and disapproval as they convey dos and don’ts, norms and permissions, rules and values, schedules, and the rest. In WEIRD cultures, the received view is that all the carrots—positive reinforcement, building self-esteem, self-respect, pride—can and should be used first to convey the normative order. If the negative disciplinary emotions need to be used, then guilt that focuses on the wrongness of particular actions is to be favored over shame, which allegedly undermines the global self.
It is widely believed in WEIRD countries that shame is amoral or, at best, morally superficial, that it takes the whole person as its evaluative object and leads the shamed person to flailing about aggressively, which is met with more shame. The moral superficiality of shame comes from the fact that it is allegedly heteronomous rather than autonomous. One is shamed by the judgment of others, not by failure in one’s own eyes. It is a common, self-congratulatory view in certain philosophical and theological quarters that shame is both a more primitive and a more superficial moral emotion than guilt, and that shame cultures are typically ancient and/or non-Western. Guilt, we are told, aligns nicely with the judgmental pose of an omniscient God, whereas shame aligns with the less epistemically and morally attuned judgment of ordinary people and whatever social consensus they reach. Guilt is internalized; shame awaits nosy third parties and finger waggers. Shame is an ugly and worthless emotion.
This is shame’s bad rap. Good shame, if there is such a thing, has none of these properties. Let’s see if I can make the case for there being such a thing.
Shame and Fitness
First, I should say something about what psychobiological function, if any, shame is thought to serve. There are two plausible hypotheses about the evolutionary basis of shame: appeasement ritual hypothesis and social devaluation defense hypothesis. They are different but not incompatible.
The appeasement hypothesis: Protoshame involves behaviors such as eye aversion and making one’s body smaller and more diminutive (the opposite, as it were, of puffing up one’s chest) in the face of a dominant conspecific. One sees protoshame in humans and in nonhuman primates, as well as in seals, rabbits, crayfish, and many birds. The thought is that, in such cases, protoshame indicates some fear and anxiety and is a display of subservience and/or appeasement, a signal indicating that I aim to cooperate, to do what you say (Keltner and Buswell 1996, 1997; Keltner and Harker 1998; Gilbert 2003; Maibom 2010). The appeasement hypothesis can, in principle, go with the view that protoshame was assembled phylogenetically as something like the unified basic emotion of shame, the genuine item, or with the view that shame is easily and quickly assembled ontogenically, as soon as, say, the caretakers assert their power to endorse and enforce norms and cause the child to blend feelings of anxiety, sadness, and fear into what we call “shame.”
The thought behind the ontogenetic assembly hypothesis is that shame is not a basic emotion but rather is quickly assembled ontogenically from basic emotions or core affective systems for sadness, anxiety, and fear (whereas guilt incorporates sadness, anxiety, fear, and anger—eventually turned inward). Shame, in this view, is a complex emotion assembled from the alchemical binding of sadness, anxiety, and fear.
Bernard Williams (1993) thinks of shame and guilt as pretty basic, but not “basic” in the basic emotions sense, where basic emotions are adaptations that involve innate ballistic situation-feeling-action syndromes, such as the manner in which fear of snakes or ledges causes movement away from the feared object. Like Gabrielle Taylor (1985), Williams thinks shame is rooted in certain natural and self-protective dispositions, including fear of losing power, dislike of being blamed and punished, and the desire for social acceptance and love.
Whether protoshame is atomic, unitary, and basic, or molecular and assembled, it is an adaptation. Heidi Maibom writes: “Ultimately, shame in animals is adaptive because living in groups is (Bekoff 2007)” (2010, 578). Maibom follows the primatologist Frans de Waal (1996) in conceiving the shame system in primates as coevolving with norms that regulate mutuality (picking fleas, disciplining youth, sexual permissions), and that work to maintain a sense of social regularity (2010, 581).3
The disposition to cooperate with caretakers is fitness enhancing. We are gregarious social animals who both need and want convivial social relations, the trust of others, and so on. In the human case—perhaps in some of the other cases as well—protoshame can then be extended by socializers to yield cultured forms of shame that safeguard social and moral norms.
But because shame has its roots in appeasing the dominant and more powerful, we can also see how shame can be weaponized by oppressors in nonegalitarian systems—for example, in patriarchies, in the caste system of India, and in racist, sexist, and queerphobic precincts the world over. Here, the solution is to turn the tables. Transvalue the bad values. Don’t give up on shame. Make these terrible practices the shameful ones.
The second evolutionary hypothesis, the social devaluation defense hypothesis, starts from the general assumption that the fragility of human life and the extraordinary dependence of humans on caretakers make social uptake a matter of survival. Thus, “preventing social devaluation—and minimizing its costs if it occurs—is a major adaptive problem” (Sznycer et al. 2018, 9702). A reasonable hypothesis is called “the information threat theory of shame,” which says that “the emotion of shame is the expression of a neuro-cognitive system that evolved to defend against social devaluation” (ibid.). Just as pain evolved to protect the body from tissue damage, shame evolved to protect the person from social ostracism.4
The appeasement hypothesis and the social devaluation defense hypothesis are not inconsistent with each other. Perhaps they can be combined into a hybrid theory. Either or both are compatible with my defense of shame. The appeasement hypothesis emphasizes our natural species sensitivity to the judgment of others, which accounts for the primitive feelings of protoshame. The social devaluation defense hypothesis provides a plausible evolutionary explanation for how and why a well-honed predictive and proleptic sense of shame would have a foothold in our nature as socially dependent animals.
Cultured Shame
Here are some safe generalizations about how, on such a basis, we learn about the nature of cultured shame and guilt, and acquire the norms and scripts that govern doing them:
1. Cultured shame and guilt are learnable either because there is an initial shame setting (protoshame) or because there are initial settings to experience sadness, anxiety, fear, frustration, and anger that can be quickly assembled (in early childhood) into shame. If caretakers (perceived either as threats or as loved ones or, more likely, as both) disapprove of something a child is doing, this can activate the shame system directly, or it can do so by arousing fear, anxiety, or sadness (shame), or these plus frustration/anger (guilt). In the first hypothesis, the basic emotion of shame is engaged. In the second hypothesis, associative mechanisms are called upon to bind some subset of the possibly more basic affective responses of fear, anxiety, sadness, or frustration/anger into composite emotions, shame or guilt. Cultured forms of shame reflect and depend on the lower-level phenotype in the ways dances reflect and depend on the ability to walk.
2. The intensity of shame and the power of disapproval to motivate normative compliance are proportional to the desire for love, solidarity, friendship, and union with caretakers and other conspecifics. Shame and guilt work because social connection is overwhelmingly important to us as gregarious social animals. Caretakers encourage children to associate good actions with love and affection, and as warranting pride, and bad actions and dispositions with shame and guilt.
There are several more things we can say about how we acquire the scripts for enacting shame and guilt:
1. “Shame” and “guilt” are theoretical terms, because they name emotions, and all emotion terms are theoretical for reasons that I discussed earlier. The words “shame” and “guilt” cannot be taught by ostension. Caretakers cannot see or point to instances of feeling ashamed or feeling guilty in others. They don’t even try to do this. Individuals learn about these emotions, what they are, what they are for, and how to enact them, not by way of some inner eye that captures phenomenal feelings but rather, especially in the first instance, by detecting patterns that connect what they do with external situations and the responses of others, which cause changes in the emotional feelings and expressions of all involved. Learning about the emotions of shame and guilt requires learning a complex theory about how one is “supposed to be in the world.”
2. If a culture differentiates between two emotion concepts—one that is to cover feeling bad about some aspect of oneself (trait, characteristic, habit, disposition, common act type) and another to cover feeling bad about one’s actions—and if, in addition, the culture teaches different scripts for reconciling, reforming, or making restitution after the two different kinds of violations or transgressions, then such a culture might be said to make a distinction between shame and guilt. Such a culture has differentiated between shame and guilt by what they are about, typically aspects of the person or acts of that person, respectively, and by the scripts governing what the violator and audience are entitled to or ought to do next—for example, change some aspect of oneself or pledge not to do the offending action again.
There is a further way in which shame is theoretical (I now mostly drop guilt from the discussion, but the reader is invited to keep thinking about it5): What exactly shame is, what it does, and what it is supposed to do is always embedded in a complex form of life that incorporates a theology or metaphysics, a psychology and philosophical anthropology, and ethics. Shame is entirely, but diversely, cultured.6 That said, it is an interesting and important question whether and how social shame, especially in its mature form, retains the features it has in its primitive form, either inborn or quickly assembled. The caterpillar is two steps away (with the pupa or chrysalis in between) from being a butterfly. There is considerable metamorphosis since the caterpillar digests itself and becomes soup inside the pupa before the butterfly—its closet continuer—emerges. Can shame metabolize itself and become something very different from its early forms? I think so. This is exactly what happens in the transformation of other-imposed shame to an autonomously endorsed and self-orchestrated mature sense of shame.
Theoretical Views of Shame
Shame and other emotions—what they are, how they feel, what actions they enjoin in the “to and fro” of everyday life—are nowadays understood in terms of the roles they play in forms of life with complex philosophical foundations. Often, the philosophical view is detectable and revealed, although not usually in a precise and orthodox way, in the ethnotheories of different folk. Understanding shame in multicultures presents special difficulties, because there are multiple, mutually inconsistent theories of what shame is, its functional role, its warrant, and so on, in the blood and bones of different folk in multicultures.
One theoretical view of the nature and origins of shame is the Genesis story, which places shame and guilt at the origin of the Abrahamic traditions. Adam and Eve’s nakedness in the garden of Eden becomes shameful after they are revealed—by committing the original sin of disobeying God’s order not to eat from the tree of knowledge—as filled with unruly carnal desires.
Aristotle discusses shame in three texts: Nicomachean Ethics, Eudemian Ethics, and Rhetoric. He thought that shame (aidos; aischune) is an emotion important for socializing the youth. Shame, like every other emotion—anger, pity, fear, love—can be used for good or ill; it can be overdone or underdone, and it can protect bad values. A mature and fully virtuous person would never feel shame, because she would not do wrong, nor even be tempted to do wrong. Virtue aims at what is good. The aim of shame is to avoid disrepute.
Aristotle judges shamelessness to be vicious. What is a shameless person? The best answer is that he fails this test: were he to want to do such and such, where such and such is shameful, he would not feel shame for wanting to do such and such, nor upon doing such and such. The shameless person’s not feeling shame is due to the fact that he is indifferent to the normative order (Raymond 2017). He is a wanton. He goes whichever way the wind blows.7
Aristotle doesn’t discuss guilt or distinguish it from shame. We do, and we make a good deal of the distinction. Gabrielle Taylor offers a sensitive analysis of shame and guilt as they are conceived in some Anglophone precincts. Shame is linked with being exposed in spheres of life where privacy is normally expected, and with experiencing disapproval by a real or imagined audience for failure to comply with shared social norms. She gives two examples, one from Sartre, where a man makes a vulgar gesture and then realizes he is seen and feels shame (1985, 57–60); the other from Scheler, where an artist’s model feels shame when she realizes that the artist no longer sees her as a model but as a woman (60–61). The second case is one in which the ashamed person did nothing wrong. It is a different and interesting question, worth pondering for extra credit, whether the model has any reason to feel ashamed despite not having done anything wrong, and if so, what the reason could possibly be. Taylor writes: “Guilt, unlike shame, is a legal concept. A person is guilty if he breaks a law, which may be of human or divine origin” (85). “Guilt and liability are conceptually connected” (89); “repayment and punishment are appropriate to guilt but not to shame” (90).
For Confucius, appropriate shame is an essential component of virtue. Analects 13.20 reads: “Zigong asked, ‘How must one be in order to deserve being called a gentleman?’ The Master said, ‘One who conducts himself with a sense of shame and who may be dispatched to the four quarters without disgracing his lord’s commission, such a one may be termed a gentleman.’ ” Mencius, Confucius’s philosophical heir, anticipates both the appeasement hypothesis and the social devaluation defense hypothesis and sees shame as an inborn psychic disposition that both detects and resists being downgraded by another human being.8 The sprout of shame is then cultivated into a mature virtue or sense of righteousness (yi), the disposition to perceive what is right and wrong, to do what is right, and to experience shame when one doesn’t.
Bongrae Seok (2017b) distinguishes among subordination shame, social shame, and moral shame in the Confucian tradition. Subordination shame is shame about status, such as the shame of a son who does not defer to his father’s judgment about how to run the family business. Social shame is experienced because one has been sloppy or indifferent to ceremonial duties. But it is only moral shame, specifically, the desire not to experience it, because one always does what is right that is associated with excellence. An excellent person lives shamelessly. The Confucian sense of moral shame is a virtue, not a price, cost, or punishment. It involves an inner set of boundaries that a good person would not think of crossing or touching. Tactile-kinesthetic metaphors pertaining to inner control—so as not to touch and be tainted and not to cross certain boundaries—are the embodied metaphors that describe the virtue. Anglophone shame typically involves visual threats from the outside, not tactile-kinesthetic threats that are produced by one’s own impulses to overreach. However, there is another Chinese virtue that involves “saving face,” and that does involve being seen and judged by one’s actions or because of how one’s actions look. But the Confucian tradition is not univocal in its admiration for shame. Although Confucius and Mencius praise a refined sense of shame, Xunzi, who thought human nature was undisciplined, compared us to “crooked wood” long before Kant thought shame was not good, only necessary, like jail, say, or the threat of jail.
For Siddhartha Gautama, the historical Buddha, moral shame (hiri) and moral dread (hiri-otappa) serve proleptically. When one is tempted to do wrong, or positively imagines doing wrong, the prospect of experiencing shame announces itself, and ideally one catches oneself up short: “(1) When there is no sense of moral shame and moral dread, (2) restraint of the sense faculties lacks its proximate cause.… (1) When there is a sense of moral shame and moral dread, (2) restraint of the sense faculties possesses its proximate cause.… When there is restraint of the sense faculties … the knowledge and vision of liberation possesses its proximate cause” (Bodhi 2012, 1070). Thomas Aquinas analyzes shame and guilt in terms of the desire to be unified with the divine and failing—humans’ desire to be beloved by God while on the earth and to eventually live with him for all eternity. For Aquinas, guilt is a psychological sign of the legitimacy of God’s disapproval, possibly his anger, for things one has done. Shame, meanwhile, calls into question one’s worthiness to be in communion with others; it marks the possibility of repudiation by judgmental others, possibly by God. “A person who is and feels shamed … may suppose that what he has done warrants in real or imagined others (or in himself) a repudiation of him” (Stump 2018, 340–41).
Eleonore Stump explains how in Aquinas the earthy dialectic of guilt and shame reflects in mundane ways one’s ultimate predicament vis-à-vis one’s relationship with God: “With regard to a person who is guilty, others would be warranted in being angry at him or wanting to punish him or extract satisfaction from him—in other words to visit on him things that he himself would find not good in some sense or other. With regard to a person who is shamed, others would be warranted (on one scale of value or another) in rejecting not his good, but him, that is, putting distance between themselves and him” (2008, 45).9 Many philosophers and psychologists claim that shame is an extremely painful emotion and guilt less so. I think this is not obvious and depends entirely on how these theoretical terms operate in a sociomoral ecology. I was raised as a Roman Catholic, and I learned that shame was something to deal with, but only sins could get you a ticket to hell. Sins are things that God can be justifiably angry about, that one should feel guilty for, and that one must atone for. God sees you naked and doing socially graceless things all the time, but he is not going to bring those things up on Judgment Day. If this is true about how the terms functioned in my youthful moral ecology, then one lesson is that we—1950s and ’60s New York Irish Catholics—used these terms with different senses and in different ways than Stump’s highbrow, Aquinas-inspired Catholics do. Bernard Williams (1993) supports my view by arguing that guilt in the North Atlantic is associated with morality in the narrow sense, the system theorized by Kant in his pietistic Lutheran mode, which is severe, stern, punitive, and guilt-inducing (see also Kristjánsson 2010, 81). Shame in its origins is associated with abiding social norms, but not with eternal damnation.10
Freudians and neo-Freudians read the roots of guilt and shame in the psychic economy of a nuclear family, which produces the stern voice of the father (guilt) and the set of aspirations or ego ideals that one can fall short of and be seen doing so (shame). Whereas Aquinas grounds shame and guilt in desires for union with God, orthodox Freudians ground them in desires of the narcissistic child to always get its way.11
Meanwhile, as we have seen, philosophers and psychologists impressed by Darwin’s theory of evolution see shame as an extension of an ancient mammalian submission ritual designed to appease disapproving or dominant others (Keltner and Buswell 1997; Maibom 2010), or as a new emotion, easily assembled from the basic affects for sadness, fear, and anxiety, which is also an appeasement response to disapproval, downgrading, and disrespect.
Scholars in countries dominated by the Abrahamic religions favor guilt over shame as the more mature moral emotion, and as the proper tool for maintaining and enforcing moral rectitude. A single, all-loving, and all-knowing God—Yahweh, God, Allah—is the only judge you need to worry about. Listen for the voice of the Father to authoritatively mark what is right and wrong. This voice will warn you off acts you want to commit that are sinful. If the voice fails to stop you from sinning, it will continue to threaten you, possibly with eternal damnation, until you atone for the sins you are guilty of committing.
Many North American psychologists claim that shame is a very destructive emotion because it requires downgrading one’s entire self (Tangney and Dear 2002). But by now, and partly in response to irresponsible scholarship on shame, there is pushback. There is slowly increasing agreement among scholars that shame does not in any way require a negative evaluation (by self or others) of a whole person. I can be ashamed that I did something shameful, or ashamed that I have certain recurring shameful desires, without thinking I am a bad or unworthy human being.
One consequence of the fact that shame and guilt are theory-laden is that normative critiques cannot be about the goodness or badness of the relevant emotion simpliciter, especially if this is taken only to involve analysis of the narrow phenomenal properties of shame. Shame feels bad, that is a given. It is supposed to feel bad. In this respect, shame is like fear, anxiety, and sadness (which some say make it up), and unlike happiness, joy, bliss, and pride. How bad shame feels depends largely on intensity settings, which depend on how the community sets the “shame-o-stat” (think of a thermostat but one that governs when shame is supposed to kick in, and at what intensity), what values shame is designed to protect (conventional and moral), what kind of metaphysical or theological enforcement mechanisms back up shame judgments, and the extent to which the individuals themselves endorse the values shame is designed to protect. What shame is, even how shame seems, depends on all these features of a social ecology that uses shame as a disciplinary emotion.
The upshot is that normative critique requires wide social, cultural, and theoretical appreciation and analysis, and this requires charitable interpretation of the functional role the emotion plays in the sociomoral life of a people, culture, social group, or subculture. It also requires sensitive assessment of background or implicit metaphysical and epistemological views about the nature and fate of persons, the order of the universe, what is knowable, and so on (Nisbett 2004; Flanagan 2017). People prefer what they are used to, and by way of the confirmation bias will often judge that the way they do things, including how they do emotions, is the natural or the right way. Patient attention to other ways of doing emotions, and being a person, can open up possibility space for being better or doing things better than one—individually or as a people, lineage, or subculture—now does. So it is with shame. We, my people, would be better off with more of it.
Three Kinds of Holism for Emotions
In thinking about what emotions are, the norms and scripts for doing them, and how to critically assess them, it is useful to think in terms of three kinds of holism. First, there is conceptual holism, which has two components: (1) the concept of an emotion like anger or shame—what the emotion is—is best understood in wide functional terms, in terms of a schema comprising typical causes + inner phenomenal features/feelings + characteristic content + typical dispositions to act + typical action (Roseman, Wiest, and Swartz 1994); and (2) emotion concepts are connected inside a linguistic community, culture, or subculture, in community-specific ways, to other emotion concepts and evaluative concepts.
To illustrate: The Ifaluk word fago, which we might translate as romantic love, is conceptually linked to a poignant recognition that the lovers might, indeed they will, one day be lost to each other. There is no similar conceptual connection reliably in play when Americans experience love. Both sociolinguistics and psychology offer overwhelming evidence that emotion concepts across cultures differ in meanings, in frequency of use, and in proximity or distance relations to other emotion concepts. There are places where grief and regret are interdefined and thus conceptually linked and places where they are not, and places where envy is conceptualized as a type of anger and places where it is not. Because conceptual holism abounds, so does variation of meaning (Lutz 1988; Lakoff 1990; Wierzbicka 1999; Beatty 2014, 2019; Jackson et al. 2019). One implication is that the word for an emotion will not always capture the complexity of the concept, which often requires an archaeology of concepts, an excavation of a concept’s history and semantic connections. In principle, conceptual archaeology could reveal that people in different cultures parse the world so differently that they can’t think the same thoughts or have the same emotional experiences (Dorter 2018).
The second kind of holism is normative holism, which says that the value of an emotion depends, and thus is legitimately assessed, on the basis of its role across personal, commercial, and political life. Because conceptual holism is true, the evaluation of whether or not an emotion is justified or functioning as it is supposed to cannot simply be based on whether the narrow phenomenal qualities of the emotion are pleasant or unpleasant. Narrow feel doesn’t reliably distinguish among emotions; think how confusing it can be sometimes to decide whether one is sad or scared or angry without thinking hard about causes, effects, and behavioral dispositions. The normative rationale for an emotion requires analysis of the socially endorsed norms and scripts for that emotion. Whether a norm for experiencing and enacting an emotion, and a script for what to do next, are good depends on what causes one to feel the emotion + how it feels + what the emotion is about + what one is disposed to do. Some causes warrant shame, and there are positive, constructive things that shame can lead one to do.12
Third, there is network holism, which is modeled on a thesis from the philosophy of science about confirmation and theory adjustment when there is feedback that something in a sociomoral ecology is off, causing trouble. If I test that water is H2O, I am testing not only my belief that water is H2O but also my belief that hydrolysis is the right method to determine whether water is H2O, plus my beliefs that my instruments are in working order, that I am awake and not dreaming, and so on. If I come upon what seems like a sample of water that is not H2O, I can give up any of these beliefs. I can conclude that water is not H2O, or that hydrolysis is a bad method to reveal the molecular structure of liquids, or that the catalyst used to start the separation of water molecules was contaminated, or that I am asleep and dreaming, and so on.
Network holism says that the same situation applies to ethics. The evaluation of moral values, principles, and moral emotions takes place inside a complex form of life, and implicates the entire way of worldmaking. If I believe that killing is wrong but also find myself thinking that it was acceptable for some individual to kill in self-defense, I can reject one belief or the other by thinking that one was ill-advised, a semi- or quasi-belief, but not deserving of all-things-considered assent, or I can modify my belief that killing is wrong to the belief that killing innocents is wrong.
If I value filial love but also value the law, and find that a loved one is a killer, I find myself intrapersonally in the situation that Confucius and the Duke of Sheh found themselves in interpersonally (Analects 13.18):
1. The Duke of Sheh informed Confucius, saying, “Among us here there are those who may be styled upright in their conduct. If their father has stolen a sheep, they will bear witness to the fact.”
2. Confucius said, “Among us, in our part of the country, those who are upright are different from this. The father conceals the misconduct of the son, and the son conceals the misconduct of the father. Uprightness is to be found in this.”
What is the right answer here? Which is the shameful act, protecting one’s father from legal punishment or testifying to one’s father’s crimes? Is the Duke of Sheh right that justice is impartial, or are there special relations that allow one to protect, without shame, a guilty loved one from justice?
If I highly value filial loyalty but find out that my father steals sheep or is a killer or, what is different, that he is a homicidal psychopath, there is pressure for something in my value system to be changed or adjusted. There are a million ways to perform the adjustment from outright rejection of my former commitment to unconditional paternal loyalty to reconceiving what loyalty to him means. For example, I could rethink things and decide that loyalty means saving my father from his worst self or that loyalty requires standing by the values my father himself espoused in his best fatherly moments, or would have espoused in such moments had there been any, and so on. Moral ecologies are real-world dynamic systems in which a sociomoral version of maintaining homeostasis, or re-equilibrating, is constantly called for. This requires continuous tinkering and adjustments in a system of moral beliefs, values, and emotions.
Most psychologists and philosophers from WEIRD cultures think that shame is exposed as an ugly emotion because ashamed people are depressed, self-hating, and antisocial—either withdrawing from social interactions or engaging with others with “humiliated fury” (Lewis 1971). This is thought to count against shame as a suitable socializing emotion. I recommend utilizing the insights offered by these three kinds of holism, especially normative holism, and reading the situation in another way: Read the problem as due to the fact that the only kind of shame studied in WEIRD precincts is an unhealthy and unnecessary variety that takes whole persons as its evaluative object, and that has been weaponized to enforce terrible values. Interpreted my way, the sensible response is to give up the bad values. Shame, the general purpose emotion, is not what is causing the problem. My proposal: Turn the tables. Attack the bad values as shameful. Ask the people who hold them to change themselves, and then teach the children well—really well (Flanagan 2017, chap. 12).
The opponent of shame who accepts the received view that shame is bad might suggest that we use guilt to turn the tables. But I don’t see how that can work if we assume the orthodox view, that guilt takes wrong deeds as its object. The problem with racism, sexism, queerphobia, and ableism is not solved only by demanding that the racist or sexist or queerphobic or ableist persons stop acting in these ways. The problem lies inside; it is some state of their souls that needs fixing. The fixing might involve purging certain well-ingrained beliefs, values, and dispositions; removing themselves from poisonous social networks; aligning what they know to be right with how they act; and working for generational change. But it involves more than just not doing certain deeds. Shame is better suited than guilt to focus on the deed, the doer of the deed, and the aspects of the self that need work.
Shame’s strength is related to a feature of shame that many judge to be shame’s main weakness. It is said, but is not true, that shame takes the global self to be deficient and judges it to be a failure. There is a complex emotion that does something like this, but it is self-loathing not shame. However, shame does typically focus on dispositions, traits, and habits as much as it does doings and deeds. And thus it asks for more than simply refraining from doing some wrong thing. It calls upon the person to do work on themselves, to do something inner, to self-modify, to realign their inner dispositions with their outer behavior. Guilt asks for behavioral conformity; shame asks for aligning the self with what is valuable. If a culture cares about protection of deeply held values, shame has its merits.