The Predicament
This book snuck up on me. In some respects it grew organically out of long-standing interests in philosophy of the mind, ethics, moral psychology, and cross-cultural philosophy on the nature and function of emotions. But it also responds to a persistent practical worry I’ve had throughout the first two decades of the twenty-first century, especially the last decade, and that I found myself talking about to family, students, and friends. I have never lived in angrier times. I’ve lived in fraught and bloody times before. I was thirteen in September 1963 when four innocent black girls were killed by a bomb at the Sixteenth Street Baptist Church in Birmingham, Alabama. Two months later, President John F. Kennedy was assassinated in Dallas. I was fifteen in 1965 when Malcolm X was killed; eighteen in 1968 when Dr. Martin Luther King Jr. and Robert Kennedy were assassinated. The year 1967 was the “Summer of Love” in Haight-Ashbury in San Francisco, and also the summer of 159 race riots from Watts in Los Angeles to Detroit to Newark. On May 4, 1970, one month before I graduated from college, twenty-eight members of the Ohio National Guard fired sixty-seven rounds in thirteen seconds at antiwar protestors at Kent State University, killing four students, injuring nine others, and paralyzing one for life.
I was a young man through the 1970s, which many say, and I agree, were transformative times. The 1960s and ’70s were a time of passionate causes: civil rights, women’s rights, gay rights (the Stonewall Uprising was in 1969), and the unjust war in Southeast Asia, which we discovered had extended into Laos and Cambodia, and was no longer just the Vietnam War. There was anger and there was blood, but one sensed at the time that both were in the service of hope.
Our times seem angrier than that time, but also mostly absent of high ideals and hopes.1 Our anger is fierce and frantic but not ameliorative. Politics especially is a zone where the communal spirit, patient listening, and public reasoning of the New England Town Meeting is a quaint memory, replaced by politics as the expression and performance of resentment and disgust. The conception of politics as the vocation of working for the common good, for justice and equality for all, is paid lip service but is recessive in practice, replaced by a model of politics as an expression of ego and the will to power, which is paid for by special, not common, interests to crush other not special interests as necessary.
I started to wonder how we could turn down the temperature on anger on both the left and the right, as a way of making room for hope, idealism, and solidarity.2 But again and again I was met with people on all sides who explained that the anger I was seeing was rational and normal. I found myself explaining that it might be statistically normal here at this moment, but it wasn’t statistically normal over the earth and over time. And it wasn’t normatively normal. It wasn’t good. I found myself going to sources outside my own tradition for examples of philosophers or saints or exemplars or whole traditions that offered arguments against being as angry as we were, even in the service of noble ends. It suits my view, although it makes me sad, that Bob Woodward entitled his latest book Rage (2020). “Rage” names both former president Trump’s character and modus operandi, and the state of current American social psychology.
At the same time that I was becoming convinced that we should turn down the temperature on anger, I worried that we (mainly my fellow Americans) were emotionally and morally off-kilter in another way that I can only describe as a kind of shamelessness. There were social roles that in previous generations would have been filled by people of good character that were not filled by people of good character, Trump being exhibit number 1. He represents, but did not remotely create, a new type, a type that shamelessly rejects the commitment to the true and the good, a type that makes fun of people who care about facts, a type that uses words like “good” and “bad,” “fair” and “unfair,” but no longer in any recognizable moral sense.3 Politicians on both the left and the right are condescending but not compassionate, indignant but not righteous, and moralistic but not moral. Truth is fungible for advantage.
It is not uncommon for people in good faith to disagree about what, exactly, is true and what, exactly, is good, but we are suddenly in an age where people who function as role models don’t care about the true and the good at all. So my thought was that it would be good to turn up the dial on shame and not mistake moral and epistemic recklessness for a kind of refreshing unconventionality or as the victory of some kind of healthy antielitism. It is neither. It is a kind of shameless nihilism that serves the interests of rapacious egomaniacs. Minimally, it would be good to reinstate norms of civic life that require commitment to truth-seeking and respectful interaction rather than allowing summary dismissals of fellow human beings as deplorable, stupid, or unworthy, or people of the wrong race, sex, gender, sexual orientation, or country of origin. There is some reason to hope. In the United States, President Joe Biden and Vice President Kamala Harris have pledged to work to overcome the cacophonous, bitter Babel that is American social and political life, and to work to reinstate norms of truthful speech, patient listening, and commitment to the common good. Former president Trump’s lies about the 2020 election, his coup attempts, and his incitement of his supporters to invade the Capitol during the certification of President-elect Biden on January 6, 2020, has been followed by calls to cease and desist with shameful behavior. Use of the words “shame” and “shameful,” expressions such as “he/they ought to be ashamed,” and fitting questions such as “Where is his/their sense of shame?” shot up in frequency in the first weeks of January (they have been rising since 1980). A dear friend joked that this was “shit for the country but good for Owen’s book.” But as you will see, I think it is very good for the country and for individuals to rediscover the great but recessive good of a mature sense of shame.
If my diagnosis of our predicament is on the right track, then it calls for recalibrating our emotions, specifically doing something different with the emotions of anger and shame. Emotions express values, abide by norms, and figure essentially in virtues and vice. In trying to argue (often only to myself) about the best form for anger and shame to take, I found myself thinking once again of using resources from cultural psychology, anthropology, and cross-cultural philosophy. Sometimes when one is in a rut, one needs to be creative in finding one’s way out. Imaginative exercises where one is encouraged to “think outside the box,” as we say, can give us permission, information, and the tools we need to explore previously unexplored, unfamiliar, often unknown possibility spaces. Parochial assumptions about one’s realistic options can be challenged.
I will introduce the reader to some of the incredible cultural diversity that is actual in our world with regard to how people do anger and shame. The aim is to reveal the different ways different people do these emotions, and to see if there are any ways of doing them that might be better for us, even by our own lights.
There is an advantage and a disadvantage to this method. The advantage is that it opens one’s eyes to alternative ways of doing emotions and thus of being a person. The disadvantage comes from this positive feature: seeing the possibility space can be daunting, possibly destabilizing, because it can seem to require us to entertain possibilities for self and social transformation that might force us to think of undoing ourselves in certain ways. Examining the possibilities for changing how we do emotions requires courage.
One reason that I advocate the method of critique for how we do emotions by way of philosophical anthropology is because it is, in a certain sense, a realistic response to our current situation. Those of us who live in increasingly multicultural cities and regions—essentially all of us—are continually exposed to people who speak different languages and who have different values and spiritual beliefs; they also do emotions differently from us. Perhaps we can start to pay more respectful attention to the possibilities in our midst and not simply demand emotional assimilation of the other. We have some things to learn.
Doing Emotions
Emotions are things we do, or better: emotions are feeling-action circuits, affective enactments. The capacity to have emotions is not gifted by Mother Nature because she thought we would enjoy the bright lights and thrill of an inner life. Emotions were selected to quickly and efficiently motivate smart, socially adept action. The way an emotion feels is motivationally powerful, designed to have us do things: head for the hills, confront an obstacle, express grief or solidarity, or mate. Experiencing emotions is closely linked to expressing emotions, which is closely linked to the regulation and coordination of social life.
Thinking of emotions as episodes, specifically as enactments, is ecologically valid. The natural arc of negative emotional episodes includes the perception of a situation as scary or sad or infuriating and culminates in the feeling of fear or sadness or anger dissipating or evaporating after an escape or tears or an angry expression. Emotion permeates the entire episode. It starts in a flash when one perceives the situation as scary or sad or infuriating and ends when some act or other leads to a release from the snares of that very emotion. In the case of positive emotions, the situation is somewhat different. The feeling of love or joy is sometimes released, as when words of love express what one feels or when laughing at a joke releases the “that’s funny” feeling. Other times, enacting a positive emotion deepens the positive emotions between lovers or shared by an audience at a comedy club, and entrains multiple emotional enactments until that set of enactments ends.
Even the most basic emotions involve “doings” and are not simply reflexes, tropisms, or ballistic reactions. A snake triggers fear and one runs. Which way one runs depends on voluntary action in an environment that affords a limited number of escape routes (uphill, downhill, or sideways). Likewise, facial expression of the emotions is under voluntary control. First, facial expressions can be suppressed. Second, there are cultural display rules that govern expression and make emotions legible to compatriots, but often not as reliably to outsiders. Third, children as young as eighteen months can make pretend faces for emotions (that is, without being in the actual emotional state that they can nonetheless mimic) and, by age three, children are becoming adept at controlling their facial and vocal expressions. This is useful for lying convincingly, but it also reveals that the child is learning what adults expect when it comes to emotional expression—fewer tears, using your inside voice, and so forth.4
We enact emotions, display emotions, and actively and emotionally engage the emotions of others. This is especially true of emotions like anger, shame, and guilt. These emotions are used to inform others that they are out of normative conformity or, at minimum, that they are doing something we don’t like or approve of. When these emotions are self-directed, as when one is angry with oneself, or ashamed of what one has done, or is suffering from a guilty conscience, the self is both agent and recipient of negative emotional judgment. Anger, shame, and guilt are disciplinary emotions. When we are objects of disciplinary emotions, we have various scripted options at our disposal: to resist, hide, dissemble, implicate others, attack, apologize, confess, atone, change our behavior, or work to modify our desires, inner tendencies, and dispositions.
The idea that emotions are doings, or are for doing, has a long pedigree. In American philosophy and psychology, William James, James Mark Baldwin, and John Dewey emphasized that humans actively explore the world by way of motivated attention and schemas of expectations that are attuned to, and prepared for, affective affordances the world provides. When we see a rattlesnake or a cliff’s edge, we see them as scary and as something to move away from quickly. The whole episode, from the sighting to the escape, is suffused with emotions. Emotions theorist Robert Solomon emphasized in his final work that emotions are psychophysical “engagements with the world” (2004, 83).5
The word “emotion” is, in the first instance, a superordinate term in folk psychology that names a motley class. When I use the term “emotion” in this book, I’m employing a model or schema that characterizes emotions in wide functional terms, as syndromes or episode types defined in terms of characteristic causes and effects. An emotion type or kind is defined by a schema comprising typical causes + inner phenomenal features/feelings + characteristic content + typical dispositions to act + typical action.6
Philosophers call the inner phenomenal features/feelings quale (singular) or qualia (plural). The inner phenomenal features/feelings include the “feeling scared” or “feeling angry” aspects of fear or anger, if there are any such aspects. I say “if there are any such aspects” because I am skeptical that “feeling scared” or “feeling angry” has a shared and unique phenomenal or qualitative feeling across all types of fear and anger, and so, too, for all other emotions. There are two reasons for my skepticism that emotions can be defined or characterized in terms of narrow phenomenal feelings. First, I am impressed by the vagueness and imprecision of reports about the narrow “what-it-is-like-ness” of emotions. Ask someone to explain what inner phenomenal feeling is shared by being scared of snakes or heights or of losing one’s job. The second reason to resist defining emotions exclusively in terms of inner phenomenal properties is that if emotions are functional syndromes, then how they seem is not simply, or even mainly, a matter of some inner phenomenal feeling that instances share independently of their causes, effects, contents, and the actions enjoined. That is, the emotional state one is in is a matter of what the causes are; how the causes are understood, interpreted, and affectively evaluated; the content of the emotion (what it is about); the action tendencies that are activated; and so on. An intense, negatively valenced response to a cliff’s edge makes one want to move away from the edge. The bad feeling one has about a possible or impending job loss, its consequences, and what might be done to avoid that outcome is a different kind of case. Even if affective scientists discover profiles for each emotion that capture shared, narrow, phenomenal, and somatic features, the wide approach has the advantage of being ecologically realistic, since no emotional episode has ever occurred—not since the beginning of time—in the narrow form without causes, effects, and action dispositions.7
To describe emotions such as anger, shame, and guilt as “disciplinary” is not to imply that they are merely or only punitive. Insofar as these emotions are more sticks than carrots, the goal of using them must be to reap the rewards of a shared, harmonious, mutually beneficial common life. It can’t be because enacting these emotions is good in itself. Doing anger, shame, and guilt correctly involves perceptual-cognitive know-how, acuity in assessing accurately what is going on, and knowing how to respond. Learning the whys, wherefores, and skill set required to experience and properly express disciplinary emotions is to acquire norms and scripts that previous generations have passed down to us because they judged emotional maturity as a necessary condition for living a good human life. The “to and fro” of these emotions is normative and scripted; it has conditions of legibility that are normally determined inside a culture. Americans use harsh words to express anger. Ifaluk people stop eating. Japanese people leave the room.
A smooth operator, a person with emotional intelligence, knows the norms and the scripts, and is a reliable detector of how others are doing regarding the norms and the scripts. The philosopher and actor Ronald de Sousa (1981) and the psychologist James Russell (1991) developed this enormously helpful idea of emotions as scripts. The idea is to get away from thinking that emotions are only or primarily “inner things,” and that emotion words are intended to refer to these inner things. Instead, it is better to think of an emotion as an event or enactment comprising a “sequence of subevents.… According to the script hypothesis categories of emotion are defined by features. The features describe not hidden essences but knowable subevents: the causes, feelings, physiological changes, desires, overt actions, and vocal and facial expressions. These features are ordered in a casual sequence, in much the same way that actions are ordered in a playwright’s script” (J. Russell 1991, 442).8
There are cultural rules about the sweet spot for enacting emotions, for expressing an emotion in a way that suits the situation—for example, not getting too angry or feeling too ashamed, or making another feel more ashamed or guilty than they deserve. Finger wagging, moral grandstanding, calling out, canceling, and deplatforming are styles of performing disciplinary emotions that are intentionally punitive and attention grabbing. They are judged by many to violate rules for decorous emotional display, as well as social consensus about where the mean (as determined by the doctrine of the mean)—the sweet spot—for emotional expression lies. Some advocates of these techniques for enacting emotions think our norms have been too polite and permissive, too patient and insufficiently attuned to what social justice requires. Others say that mercy and forgiveness are always warranted and that no human being should ever be permanently written off. How do we resolve such disagreements, especially given that they themselves are heated? The debates about whether and how angry we should be enact the very angry emotional displays that we are trying to simultaneously, rationally assess and critique. These are hard problems.
Cultures and subcultures differ in both norms and preferred scripts. The norms of friendliness, politeness, anger, and annoyance vary greatly across different ecologies, both between cultures and inside cultures, where the conditions of legibility can be extraordinarily intricate because of embedded subroutines based on status, age, gender, ethnicity, and wealth. This is especially so in cities and regions that contain people of multiple lineages. How do we figure out the right emotional scripts when there is variation among the options?
The question can be divided into two questions, one causal and one normative. The causal question is: What happens when a large immigrant group with, let us imagine, its own practices for expressing anger, pride, gratitude, grief, forgiveness, shame, and guilt arrives in a city or country that already has its own dominant practices for these things? Emotional norms and scripts usually designate certain states of affairs as ones for which, say, anger is warranted. These norms and scripts determine the degree of anger warranted by these causes; how intense the feeling of anger ought to be, given the causes; and what sort of action the anger warrants.
One answer to the causal question, studied in northern Europe, is that the dominant norms and scripts typically crush or swallow up the nondominant norms and scripts in a matter of three generations (De Leersnyder, Mesquita, and Kim 2011; Jasini, De Leersnyder, and Mesquita 2018; Mesquita, Boiger, and De Leersnyder 2016). The reason: the simple power of majority practices. Presumably, there is some size that an immigrant group might reach, such that it could affect the practices of the dominant group; or, on the other hand, what certainly does happen: emotional practices of a minority group that are not endorsed in public life are preserved at home, in religious places, and in immigrant neighborhoods, and are thus not extinguished in the private zones of life.9
The normative question is: What should happen when different norms and scripts for emotional expression meet? Can such clashes serve as opportunities for reflection—on both sides—about whether each group’s ways of enacting emotions are good, whether they are suited to the new ecology (new to them and new again after people with different norms and scripts for emotions come together and interact)? How should individuals on all sides assess or reassess their practices? Are there rational ways to assess, evaluate, and appreciate alien ways of doing emotions for the sake of improving one’s own way of doing them?
This book is devoted to this question: How should we live, or, more specifically, how should we think about how to live in a world in which there are so many live possibilities for being a person? These live options involve matters of which projects to pursue, what to value, and which emotional norms and scripts to abide by for high-quality personal and interpersonal relations. These norms and scripts always involve standards for apt or fitting emotions. Many of us live in multicultural cities, or visit such cities often enough to experience alien ways of being human among people who are nonetheless living well by our own lights and standards. Often such people enact different emotional norms and scripts. How are we to understand these differences? We know that differences can often result in thinking of the other as odd or weird. What are the chances that the way the odd duck or weirdo conducts itself might be good for us if only we could see its strengths and adopt its practices?
Emotional Variation, Animal Natures, and Ecologies
Modern Homo sapiens is 250,000 years old and currently the only living species of the genus whose members extend back between two and three million years. Homo sapiens plus the four still-living higher primates—chimps, bonobos, orangutans, and great apes—are descendants of a common ancestor who lived fourteen million years ago. Anthropologists have reliably identified about a dozen extinct species in the genus Homo. Neanderthal (Homo neanderthalensis) is a closely related species that roamed Europe as recently as 24,000 years ago and interacted with us before we invented agriculture and domesticated animals 12,000 years ago. New evidence suggests that the diminutive Homo floresiensis, which some think may have been an offshoot of Homo erectus, lived in the South Pacific as recently as 17,000 years ago. In the autumn of 2015, a new extinct species, Homo naledi, was confirmed based on skeletal remains in what looks to be a burial site, at least a body disposal chamber, in a cave in South Africa. Think about it: we once, over tens of thousands of years, commingled with other species of humans.
Over the course of the descent, Mother Nature equipped us with certain emotional dispositions that are psychobiological adaptations and come in the form of “initial settings” for doing emotions, such as a fear of falling off high ledges and the anger or protoanger that an infant shows when its desires are obstructed. These initial settings mark certain situations as scary or angry-making; they are seen or experienced as such, and they motivate certain characteristic actions, such as crawling away from the edge, crying, and so on. These are what some call “basic emotions” or “core affect programs.” Basic emotions are always for the sake of action and involve settings that have an active, perceptual side. For example, the crawling child is set up to quickly perceive that it is at an edge and moves back.10
How does the basic emotional profile of Homo sapiens compare to that of the other Homos? We don’t know, since the others are extinct. But different bodies and different types of brains usually bespeak different phenotypic traits to some degree, so we can safely assume that there were likely some differences in initial emotional settings among us and the extinct Homos. Evidence from the genus Pan with two still living species—chimps and bonobos—reveals that chimps, who show greater sexual dimorphism that bonobos, for whom males and females are similarly sized, are patriarchal, aggressive, and conniving, while bonobos are matriarchal and like to resolve disputes with sexual healing.
Voles, not to be confused with moles, are another mammal popular among philosophers interested in emotional regulation. Prairie voles and meadow voles look alike but differ greatly in “attachment style,” and the differences can be traced to variations in species-specific genes that regulate oxytocin, vasopressin, and dopamine. Prairie voles are monogamous and attentive to their offspring, whereas meadow vole males are promiscuous and comparatively indifferent to their young (Carter, DeVries, and Getz 1995; Wang et al. 1999; Churchland 2012).
Thinking in evolutionary terms requires thinking historically, in terms of lineages with long, natural histories. It requires thinking of lineages as coevolving with ecologies, often microecologies, high mountain ranges, equatorial forests, different postal codes, a particular street in a particular neighborhood, a family. Sometimes microecologies are discontinuous veins in a shared space coalescing around such features as age, sex, gender, sexual orientation, ethnicity, education, or religious affiliation. Strictly speaking, there are no shared ecologies. A family that raised clones would invariably offer a novel ecology to each child. The parents have changed in age and experience, they are more patient or less patient, there are age differences among the sibling clones, there are novel effects of sibling interactions, and, of course, the world outside is ever changing.
The unshared features of what might seem to be a single ecology, but isn’t, cannot be emphasized enough. We often speak in general terms about, for example, China, how the Chinese are and how they are raised, as if, at a certain time slice, there is a coherent, unified way of life and a single ecology named “China” or “the Chinese.” This can be useful shorthand, a way of orienting ourselves to a certain geography and demography, a kind of political and economic life, and a historical landscape. But it is only that: a handy typology that reduces the noise that would attend thinking realistically about China, which is, strictly speaking, cognitively impossible. There is no such thing as “China” or “the Chinese” in the intended sense, despite the fact that we can truthfully say such things as that China—now marking features of a certain nation-state plus distinctions drawn by the Chinese government—is one of the least ethnically diverse countries in the world, along with Australia and Argentina.
But there isn’t a unified or homogeneous ecology in China or almost anywhere else, at least not at any fine-grained level. Han Chinese make up 90 percent of the Chinese population, but there are more than fifty other, mostly indigenous, Chinese ethnic groups with their own histories and traditions that are recognized by the Chinese government. The most ethnically diverse countries in the world include New Guinea, India, Mexico, the northern parts of Central America, the western parts of South America, and all of West and Central Africa. In India, there are more than two thousand ethnic groups. The US Census Bureau identifies only six ethnicities, marked by racial characteristics, but America, in fact, contains a huge number of mostly exogenous ethnic groups, since the indigenous peoples suffered multifarious degrees and kinds of extermination during colonization. Still, America is middling among the countries of the earth in terms of comparative human ethnic diversity. Nonetheless, the point about differences at every level of ecological grain obtains in countries like America with a common language and centralized government. As an Irish Catholic in New York in the 1950s and ’60s, I grew up in a different ecology than my Jewish acquaintances and even than my Italian Catholic friends who attended the same school as I did. I was both puzzled and delighted by interactions with the Garguilo and Mancuso families and the Steinbock and Sternberg families. They ate differently than we did, celebrated different holidays (Saint Anthony’s Day rather than Saint Patrick’s Day; Hanukkah not Christmas), and they really did emotionally engage each other differently than we Flanagans did. The Garguilo and Mancuso men, for example, hugged and kissed each other. We Flanagans did not do that, although my brother and I do now. Mrs. Sternberg yelled a lot but never seemed angry. When my mother raised her voice, she was angry.
Many countries in the world that are not among the most diverse overall contain cities that are incredibly diverse—for example, Houston, Jersey City, and Stockton, California, in the United States. Half of the people in Miami and Toronto are foreign born. Amsterdam, Sydney, Melbourne, London, and New York all have foreign-born populations in the vicinity of 40 percent. Close to two hundred languages are spoken in both New York and Los Angeles—but not the same two hundred languages. Languages parse the emotional universe in different ways. And cultures prize different kinds of emotional expression. To the WEIRD (Western, Educated, Industrialized, Rich, and Democratic [Henrich, Heine, and Norenzayan 2010; Henrich 2020]) eye, Mexicans might laugh too much at work, and Chinese people might be judged for hiding their emotions.
There is another point: ethnic, linguistic, and country of origin diversity are not the only ways to measure diversity and to think about interaction between diverse groups. There is, for example, economic diversity within and between countries, and socioeconomic status is known to also mark different habits of the heart and mind inside a nation-state independently of race, ethnicity, or gender. The Gini coefficient measures economic inequality. Namibia and South Africa have the highest Gini coefficients. The United States is about fortieth of 180 countries (top 20 percent) in terms of economic inequality. Northern European countries and former Soviet Union countries are the most equal economically (but for different reasons). China, even though it is 90 percent Han Chinese and thus on the ethnically homogeneous side, has a high Gini coefficient—about the same as the United States, which is much more ethnically diverse than China. With money comes the power to employ, to sue, and to buy out, and sets the stage for paternalism, condescension, and the power to enforce rules of emotional deference and submissive expressions of courtesy and gratitude.
FIGURE 1. Waddington epigenetic landscape.
The key point is that formation as a person takes place amid a variety of dimensions that, depending on the ecology, make use of divisions by race, language, ethnicity, gender, political affiliation, and wealth. When differently formed groups with different social statuses interact, there are often clashes of value and of ways of being human, including norms and scripts for emotional expression.
Ecologies, microecologies, and micro-microecologies can be conceived as landscapes that receive, channel, and interact with the organisms who enter the landscape, and who change and are changed by that very landscape. Here is C. H. Waddington’s picture of a simple individual, represented by the ball, about to enter a fairly simple, variegated landscape (see fig. 1). The probable trajectory of the ball and its probable end point are easy to see. Now, for realism’s sake, imagine a multiplicity of nonsimple individuals, represented by balls with variegated dimples, carrying their own individual and social histories in the pattern of dimples, and entering into a much more complicated and variegated ecology from every direction with different initial spin. Such a picture represents the ecological situation across the earth and also the situation in many multicultural, cosmopolitan locales, places where lucky people come for new opportunities and unlucky people come to escape degradation, poverty, war, and genocide.
A Waddington landscape can help us think about variation inside a culture or between cultures, as well as about circumstances where interactions take place between people from different valleys. If one wants to, one can imagine two persons who, by the circumstances of birth, end up living in the leftmost and rightmost valleys. They are Mexican and American, Basque and !Kung, or New Guinean and Rwandan, say, and they are constituted to some significant extent by values particular to the ecology in which they were formed. Then circumstances of immigration lead the people from different valleys to intermingle. What happens then? What should happen then? My “What happens?” questions are focused on what happens and should happen to their values and how these are affectively colored, constructed, and enacted to do work in creating the conditions of a meaningful life. It is known, for example, that on average, Mexicans are happier and experience more positive emotions than Americans.11 If they meet, who will change and why? If they meet, who should change and why?
Citizens of Ecuador, Paraguay, and Uruguay lead the world in feeling that they are treated respectfully by their compatriots. Will we in the United States learn their rules for emotional expression, ideal emotional tone, and respectful interaction if they immigrate, or will they acquire our less respectful modes of interaction? Or, considering another case, the Japanese in Japan, even those who score high on the anger trait, do not score high in identical circumstances, such as the experience of road rage, like Americans do (McLinton and Dollard 2010). If a Japanese person moves to America and is caught in a traffic jam or is yelled at on the freeway, what will happen? What should happen?
The first question invites causal prediction. The second question invites wondering if, and when, reasons for or against a certain way of being should be brought to bear on the otherwise decisive, and possibly entirely arational, causal circuits. Thinking in terms of Waddington landscapes helps us realize that a fertilized egg identical to that of Martin Luther King Jr. or Mahatma Gandhi or Sojourner Truth or Rosa Parks or Albert Einstein, born at an earlier or later time and in a different place, is a different person and not, as it were, one of those notable people, although, we hope, a good, worthy, and successful person in their own right. Prospects, fates, and ends depend on multiple sources, some in individuals but most in natural and social ecologies. Forms of life, visions of good lives, are cocreations of persons, communities, and natural and social ecologies with long natural and social histories. One consequence of thinking in terms of coevolution is that we detect the possibly disturbing truth that if we were born into different cultures, we would have different emotional experiences and different views about the proper norms and scripts for emotions. But perhaps this knowledge can be power. Can we leverage this knowledge of our contingency to motivate us to try to locate live options for changing ourselves for the better? One source of evidence about the possibility space can come from attentive study of cultural differences in emotion norms and scripts. For denizens of multicultures, the study can really be that of a “participant observer,” since we already live in such mixes.
For most of human history, we have lived in small groups, typically fifteen to twenty member bands, rarely larger than 150 people. Reputation tracking is relatively easy with numbers such as these. One could easily know who had reason to be proud, or ashamed, or to feel guilty. Starting twelve thousand years ago, with the inventions of agriculture and the domestication of animals, there was a rapid expansion of the size of social units and thus entirely new ecologies. New practices in agriculture, trading, waterworks, and architecture allowed urban development. There are, as I write, 30 cities in the world with more than ten million people, and that number is predicted to rise to 43 by 2030. There are 436 cities with between one and five million people, and another 550 cities with between five hundred thousand and one million people. With anonymity and distance between trading partners come difficulties in reputation tracking and the need for new strategies for doing disciplinary emotions. One form this eventually took over millennia was the change from up-close and personal reactions to normative violations to handing over punishment and rewards to impersonal institutions, the military, civil service, school systems, the state, and international courts.
Thinking of human development as taking place in multifarious ecologies allows, but does not require, us to think of the mechanisms that govern social and cultural evolution as operating according to similar mechanisms that operate on genes. Selection by consequences is the unifying idea that genes, ideas, norms, and social institutions increase their footprint, but perhaps only for a limited time in a particular place, when they lead to reproductive success, or to a successful solution to an ecological challenge, or to the flourishing of some population (those in power, the men, the whites, possibly, at the limit, everyone), as well as various serendipities, chance, drift, and randomness (Boyd and Richerson 2005; Richerson and Boyd 2005). One fancy capacity that humans, but not just humans, have is the capacity to create and maintain various kinds of normative order. The normative order uses both (1) the capacities of individuals to acquire reliable dispositions inside themselves—typically conceived as virtues or character traits—to do what is judged to be good, right, or expected, and (2) public institutions and structures, such as the government, the law, and tax codes, to accomplish, regulate, and enforce regimens of compassion, justice, forgiveness, and mercy, which individuals might not find easy to motivate from reliable inner resources. Emotions are considered to be integral parts of virtues, and they play a major role in marking what we value and don’t value (Kristjánsson 2018; Sreenivasan 2020). We, but again, not just us, are normative animals (de Waal 2006b; de Waal et al. 2014; Whitehead and Rendell 2015; Andrews 2020).
When Cultures Meet
Think again of the valleys in a Waddington landscape as inhabited by peoples who live in distinctive cultures. One might imagine that, in place of some of the hills between valleys, there are instead seas and oceans and impassable mountain ranges, and many different languages, cultures, and religions on the other side of these barriers. Imagining variation might help us consider the extent to which the ways we think about and enact emotions might be path dependent and culturally specific.
Anna Wierzbicka is a linguist who has spent much of her career warning anthropologists, psychologists, and cognitive scientists about the danger of taking the meanings of psychological terms in English as capturing psychological essences that can be mapped onto foreign terms without loss of meaning on either side. In Imprisoned in English (2014), Wierzbicka gives numerous examples of a lack of synonymy of emotion terms across languages. One might think that translation is straightforward between nearby languages like English and German, and especially for terms that some think name basic emotions. But this is not so. “Anger” is typically translated as Wut. But Wut refers to a very negative feeling (as opposed to simply negative) and involves “being out of control,” which “anger” in English does not. Wut also connotes destroying something rather than, say, retribution against someone for an act of injustice. So perhaps Wut is better translated as “rage” or “fury,” and Zorn is better for everyday negative, but not out-of-control anger. But Zorn is an increasingly uncommon word in German and is also often translated as “wrath” (2014, 79–82). So neither Wut nor Zorn accurately captures the English word “anger,” but it’s good enough for everyday work, and we let linguistic context fix meanings at a particular time so that, for example, if Wut is used to translate anger about a minor inconvenience, we know it is not normal German Wut, which is “furious.”12
In her classic book, Unnatural Emotions (1988), Catherine Lutz also discusses the difficulties of translating emotion terms across cultures. She explains that the Ifaluk word fago can be translated as romantic love, but it is better read as meaning something like love + compassion + sadness. Lutz considered the possibility that fago is one word that, depending on the context, names what for us are three separate emotions. But for good anthropological reasons, including applying the principle of charity in interpretation, she found that the love + compassion + sadness interpretation was the right one. This means that fago is a word that combines three emotions that we normally distinguish. As understood by us, fago is a linguistic molecule made up of three emotion atoms. This does not—without lots of further analysis—mean that it is an emotion molecule for the Ifaluk. It could be that what we take to be a combination of more basic emotions is taken as unitary by the Ifaluk. The greatly simplified story for this word—imagine it is uttered like our word “love” on one’s wedding day—is that the fragility of life among the Ifaluk is such that when one looks into one’s beloved’s eyes, one experiences love + solidarity with the lineage from which they both come and that contains much loss + the recognition that either loved one might be lost to the other before the proper time. So fago means “love,” except, well, as with “anger” and Wut, not really.
Emotion terms are one kind of value term. “Good” and “beautiful” are value terms with affective aspects. Some good things are not beautiful, and not everything that is beautiful is good. But what is good and what is beautiful are identified by the same word in twenty-seven different languages from eight different language families (Mayer et al. 2014). The close conceptual connection between what is good and what is beautiful reveals itself in the finding that Americans infer moral goodness and a healthy personality from physical attractiveness (Dion, Berscheid, and Walster 1972). The fact remains that most languages in most language families do not use the same word for “good” and “beautiful.” This does not settle the question of whether what is good and what is beautiful are conceptually closely linked in these languages but only that it is not so obvious. Cars and bikes could be named by one word. They are not. But the words “car” and “bike” are conceptually and semantically interconnected in ways that “car” and “cat” are not.
These sorts of linguistic relativities and variations matter because we will be talking about emotions using English words, but we can’t assume that we carve up emotions as other cultures do. And if we don’t, we can’t be confident that, when we discuss norms and scripts of emotions, we are comparing apples to apples and oranges to oranges. Tracking the meanings of emotion terms across languages is a promising way to find out what different peoples mean by their emotion terms—how emotion terms are connected in a language with other emotion terms—without presupposing that emotion terms mean the same thing across languages. The psychology of emotions is holistic. Even if emotions are distinct from one another, they interact with one another conceptually and causally in myriad, culturally specific ways.
Linguists classify languages into language families according to principles of descent. There are 135 language families from which emerge the 6,500 languages spoken in the world today. About 4,500 of these languages have substantial numbers of people speaking them. Mandarin Chinese is the first language of one billion people. English is the first language of about 380 million people, but it is the most spoken language in the world at 1.5 billion competent speakers. What do we know about how people across the earth conceptualize emotions?
A remarkable study by Jackson and others (2019), published in Science, reports on an analysis of the meaning of emotion terms in a sample of 2,474 languages from twenty major language families. The experiment tracks “cases of colexification, instances where multiple concepts are coexpressed by the same word form in a language.” For example, in Persian, the word aenduh is used to express both the concept of grief and the concept of regret, whereas in the Sirkhi dialect of Dargwa, the word dard is used to express both the concept of grief and the concept of anxiety. The key background assumption is that colexification can be used reliably as a proxy for conceptual proximity or distance as understood inside a tradition. Persians think of grief and regret as similar emotions, whereas Darga speakers think grief is more similar to anxiety than regret.13
Jackson and others write:
Our findings reveal wide variation in emotion semantics across 20 of the world’s language families. Emotion concepts had different patterns of association in different language families. For example, “anxiety” was closely related to “fear” among Tai-Kadal languages, but more related to “grief” and “regret” amongst Austroasiatic languages. By contrast, “anger” was related to “envy” among Nakh-Daghestanian languages, but was more related to “hate,” “bad,” and “proud” among Austronesian languages. We interpret these findings to mean that emotion words vary in meaning across languages, even if they are often equated in translation dictionaries. (2019, 1522)
They continue:
Despite this variation, we find evidence for a common underlying structure in the meaning of emotion concepts across languages. Valence and physiological activation—which are linked to neurophysiological systems that maintain homeostasis—served as universal constraints to variability in emotion semantics. Positively and negatively valenced emotions seldom belonged to the same colexification communities, although there were notable exceptions to this pattern. For example, some Austronesian languages colexified the concepts of “pity” and “love,” which implies that these languages may conceptualize “pity” as a more positive (or “love” as more negative) concept than other languages. The ability of valence and activation to consistently predict structure in emotion semantics across language families suggests that these are common psychophysical dimensions shared by all humans. (ibid.)
This is interesting in a host of ways. First, it supports my skepticism about the claim that emotion words name mental states that are typed by universally recognizable, narrow, phenomenal feelings. On such a view, “fear” names whatever inner state has the robust, distinctive, and unambiguous fear feeling; “anger” names whatever inner state has the robust, distinctive, and unambiguous angry feeling; and so on for “guilt,” “shame,” “happiness,” and “sadness.” In my experience, it is normally fairly easy to get oneself or another to second-guess what emotional state they are in when queried: “Are you sure you are angry at James? I think you are more scared or sad about losing your relationship with him than you are angry.” Second, these findings about valence and activation support Lisa Feldman Barrett’s (2017) and Joseph LeDoux’s (2018) view that linguistic communities teach the language of emotions by making inferences—for example, in a child’s case—about whether a reaction expresses a positively or negatively valenced state. If the child breaks a toy, we infer negative valence; if she starts crying, we infer high physiological activation and teach the child the language of being “sad” as opposed to the language of being “bored” or “depressed,” which are also negatively valenced but of low activation.14
Figure 2 visually maps some of the semantic relations among emotion terms in different language families (Jackson et al. 2019). In Indo-European languages, grief and anxiety can be expressed by many different terms, whereas it is not at all clear how they are expressed in Nakh-Daghestanian languages.15
FIGURE 2. Colexication of emotion concepts across all languages (top left) and the largest language families. The nodes represent emotion concepts, and node size represents the number of colexications involving the concept. The connecting lines represent colexications, and connecting line thickness represents the number of colexications between two emotion concepts. The node shape designates community. Adapted from Jackson et al. 2019.
Emotion Words Are Theoretical Terms
How does a child learn the language of emotions? One thought is that adults point to an emotion in the child and name it in the same way they point and say “apple,” “red,” “nose,” and “car.” But adults don’t see emotions in children, nor do children see them in themselves, although they have feelings, as we say. Adults and older siblings surmise that a child is sad because her toy broke and she is crying, or she is happy because she is smiling while eating M&Ms.16 The adults don’t point and say the right word; the adults infer, surmise, and offer the child a way of interpreting and speaking about their experiences.
The distinction between observation terms and theoretical ones is neither simple nor clean. Names for common objects and some of their properties can be thought of as paradigm case observation terms. “Apple,” “red,” “round,” “block,” “tree,” and “dog” are observation terms. Naming them is anchored to observable things, things even a novice can directly experience. “Love,” “friendship,” and “bully” are theoretical terms. These things surely exist. They are part of the ontological table of elements. But you can’t see them without understanding something akin to a piece of social theory. There are human relations and human behaviors that require positing phenomena such as love, friendship, and bullying. In science, we don’t see electrons, neutrinos, bosons, fermions, genes, or electrical fields without instrumentation, in some cases not even with instrumentation. These things are all real and required to explain the phenomena in their totality. It is the same with terms for psychological states with unobservable properties (the way you feel now) and terms that name complex social relations (bullying, friendship, true love).
In a paper on learning emotion terms, Shablack, Becker, and Lindquist (2019) write:
Much of the experimental work on children’s vocabulary development focuses on how children acquire words for object concepts, which are primarily labeled by nouns (Gentner, 1982; Huttenlocher and Smiley, 1987; Markman, 1990; Bloom, 2000). This emphasis is logical, as children’s earliest vocabulary items are largely nouns that label people and basic objects (Bates et al., 1994). However, words of different lexical categories (verbs, adjectives, etc.) tend to have very different kinds of meanings and are learned in very different ways. For instance, verbs often label actions and events, and adjectives, which modify nouns, typically label properties or attributes. Emotions are internal states that are most frequently labeled by adjectives in everyday speech (Shablack, 2017), and verbs and adjectives are conceptually more complex than nouns (Gentner, 1982). Moreover, while caregivers may label salient objects for children ostensively (e.g., “Look! That’s a dog!”), caregivers do this only rarely (if at all) with properties and states of being (Gleitman, 1990).
Words like “sad” or “happy” are introduced in complex episodes where what is marked is a relation consisting of the cause (of the feeling state) + feeling state with its associated content (what the feeling is about) + behavioral accompaniments and effects. Words like “sad” or “happy” are introduced and learned by way of meaning rules that are wide in scope and refer to functional syndromes that involve typical causes + feelings + effects. Never in any actual ecology has anyone tried to teach the child names for an inner phenomenal state that is sadness-as-such or happiness-as-such by ostension. I doubt that there is any such thing as an emotion-as-such.
Even in cultures in which there is lots of emphasis on what a person feels,17 and in which careful description of what one is feeling is prized (thus, in all therapeutic cultures), one is always being encouraged to describe a feeling state with its associated distinctive content (often with its causes as well).18 I’m sad that the candy is gone; I’m sad that big sister can stay up half an hour later than me. It is an interesting philosophical question whether there are any phenomenal properties that sadness in these two cases shares independently of their content. But it is not an ordinary question, nor is it easy to study or decide.
One thorny epistemic problem is about the identity conditions for narrow phenomenal feelings—anger-as-such, sadness-as-such, fear-as-such—assuming that there are any such things. Barrett (2014, 2017) and LeDoux (2019) claim that there is no brain signature that captures any common feature of conscious fear or anger, or any other emotion. Furthermore, introspection and/or phenomenology will not yield a high degree of confidence that fear of snakes and fear of being fired are of the same kind fearwise, or that being sad that the M&Ms are gone and being sad that Mommy said it’s time for bed are the same sadnesswise. People will express confidence about the differences between sad states and scared states. But this confidence may well have as much to do with information they receive from the situation, audience cues, and dispositions to act as it has to do with any direct information about inner phenomenal feelings (in which I include somatic/bodily feelings).
Suppose we came to really care about whether people use emotion terms like “anger,” “fear,” “sadness,” and “shame” to refer to the same inner qualitative states—anger-as-such, fear-as-such, sadness-as-such, shame-as-such, assuming for now that there might be such things as emotions-as-such. One might accept my view that emotion terms normally refer to functional syndromes, only one aspect of which involves inner feelings, but still legitimately note that I can get angry that the Red Sox lost the series and I can get angry that the president is a racist, and want to know whether there is any consistency to the phenomenal feeling aspects of anger across these two anger episodes.
We could divide the question into two, one intrapersonal, the other interpersonal or communal: Does some individual use words like “anger,” “fear,” “sadness,” and “shame” reliably over time, where “reliably” means that the inner phenomenal feeling is the same, or more or less the same, across usages? Do the members of some population use these terms to refer to the same, or more or less the same, inner state?
Suppose one started with the reasonable assumption that conscious emotions supervene on or emerge from complex, possibly global, brain and bodily states. If so, then sameness of phenomenal states depends on some kind of sameness at the level of brain and bodily states. Thus, in the intrapersonal case where we wonder whether some individual uses words like “anger,” “fear,” “sadness,” and “shame” to name the same feeling states, linguists should gather speech-act evidence, which would be checked against physiological measures of heart rate, blood pressure, cortisol level, and so on, which would then be checked against brain-imaging data and neurochemical assays. Then we’d look for matches in the bodily state profile with the intrapersonal use of the relevant emotion terms. The interpersonal case would be more difficult and not just because of the numbers of subjects required. Different human bodies are bound to display differences across every physiological and neural measure. Think about your annual physical when your doctor shows you your comprehensive metabolic and blood panel scores and declares you healthy. Every measure admits of a range of values that are all normal. Different bodies achieve the emergent property of blood and metabolic health in somewhat different ways. Likewise, there will be many levels of grain at which no two bodies are ever in exactly the same state or do anything—even visual perception or digestion—in exactly the same way. One consequence is that answering our two questions about sameness of phenomenal feelings requires more than matching similarity or sameness of bodily states with linguistic reports; it requires theoretical advances in consciousness science that tell us what kind of similarity or sameness at the bodily levels are sufficient to infer a high degree of phenomenal similarity or sameness in conscious emotional experience.19 Furthermore, if emotional states are the states that they are partly in virtue of their content, which in the normal case the person consciously knows, then sadness that the flowers died and sadness that Mom died are unlikely to be the same kind of sadness. This is not just because they differ in depth, intensity, and existential ramifications, but (1) because their content differs (dead flowers versus dead mother); (2) because the distinctive contents are partly constitutive of the two different sadness states; and (3) because these sadness states with their different contents must supervene on, emerge from, or be produced by a brain/bodily difference. Otherwise, it is a miracle that we can detect them and know what emotional state we are in. Where is the sadness-as-such? Probably nowhere.
Emotional Norms
So emotion words are theoretical terms, not observation terms. The community infers that the child is sad or happy or jealous or ashamed by way of typical causes and effects, including the child’s own behavior, and teaches the child the language of emotions. The child does not attach emotion terms to inner feelings alone but to inner feelings plus characteristic causes and effects.
Emotion terms also convey norms. That is, the child is introduced to emotion terms along with encouragement to learn the aptness conditions, rules, and norms for expressing that emotion. There are modes of emotional expression that are appropriate, decorous, warranted, and ethical, and those that are not. In a human life, there will be many times when one will feel sad, angry, scared, or ashamed, but ought not to. The oughts and ought-nots for emotions are offered normally as culturally certified norms for feeling and expression. The child who is learning words like “happy” or “sad” is learning about emotions, as well as when it is appropriate to be happy or sad or, what is different, to express happiness or sadness (advanced placement emotions training often involves learning what emotions you are permitted to have but not to share). It is okay to be happy when you are given M&Ms but not when your sister cries because they are “all gone.” It is okay to be sad when your puppy is sick or when you fall and scrape your knee, but not when the M&Ms have been evenly distributed between you and your sister and you want more than your fair share.
Anger and shame are generally even more implicated in normative life than emotions like sadness, fear, and happiness. As disciplinary emotions, they play central roles in an entire form of life. The Old Testament, for example, theorizes shame primarily in terms of unlawful sexual relations, especially mother-son sex and sibling incest. Modern children are introduced to varieties of shame and taught that some common sources of shame do not warrant real shame. For example, shame is appropriate for bad things you have done, bad intentions and dispositions, but not for wrongs that have been done to you or features of your body or mind that you have no control over. The philosopher Eleonore Stump offers a helpful fourfold taxonomy of shame: “Shame resulting from one’s own wrongdoing; shame stemming from wrong done by others; shame following on some defect of nature; shame attaching to being a member of the human species” (2018, 347).
Each kind of shame is governed by norms: If I steal from the poor I should be ashamed. If I am raped I might be ashamed but shouldn’t be. If I am born with a mental or physical handicap, I might feel ashamed but shouldn’t. If I feel awkward, embarrassed, or ashamed for being naked in front of uninvited others, welcome to the human species. If I feel shame because my ancestors owned slaves, it is contested whether this is warranted shame or not, and thus whether I now have a special responsibility to make amends for what “my people” did then.20
Or consider anger. According to its fans, anger is a refined detector of injustice, and uniquely effective at getting oppressors to yield their evil ways. Liberté, égalité, fraternité, insofar as they have been achieved, have only come from the insistent activism of angry souls. Opponents of anger say that anger is undisciplined and sloppy. It is “greedy for revenge,” the unprincipled servant of a rapacious ego, and a miserable detector of the truth. Seneca, the first-century Stoic, offers this evidence of anger’s ultimate futility: “See the foundations of the most celebrated cities hardly now discerned; they were ruined by anger” (1928).21 And Mahatma Gandhi writes: “The only force of universal application can … be that of ahimsa or love.… It follows, therefore, that a civil resister, whilst he will strain every nerve to compass the end of the existing rule, will do no intentional injury in thought, word or deed [to another person] … a civil resister … will harbour no anger” (1993).
Different communities teach different norms and scripts. Stoic children, Buddhist children, and children of the Western Enlightenment learn what anger is, in large measure, by learning norms about when, if ever, anger is permitted, when it is expected or required, how to enact it, how to suppress or rechannel it, and how to respond if one is on the receiving end of another’s anger.22
Emotion terms are taught and learned in ecologically rich situations that typically involve episodes and events with complex causal structures, in which norms for the emotion are in play and being conveyed in the case of new language learners, or reinforced in the case of competent language users. When one is introduced to emotion terms and concepts, one is also simultaneously being inducted into a form of life, a way of being human, in which novices are taught about how we (in this family, school, culture, nation-state) do emotions, enact norms of appropriate expression and action, and so on. The child is inundated with information about what sorts of things typically make other kids and caretakers sad, angry, happy, and embarrassed. And the child can read off information about causes and permissions from the caretakers. “When I don’t share with my sister, Mom gets mad and sends me for a time-out.” The child also learns what kinds of emotional displays on her part get positive or negative uptake. If a child eventually claims to have an opinion about the narrow phenomenal feelings that are definitive of different emotions, she is engaged in a complex and fraught abstraction exercise, because she has received almost no direct information about the matter at all.
Path Forward
Let me sum up and at the same time describe the path forward. There is the predicament, a method to address the predicament, and a view of the nature, function, and plasticity of the emotions. The general version of the predicament is this: There are many philosophically and existentially significant things that humans may question about how they are living at a time and in a family, community, and nation-state for the sake of doing better. We can question our loves and hates, opinions, values, virtues and vices, general principles, commitments, aims and aspirations, and civic institutions and practices. One among the things that can be questioned, criticized, and possibly changed is how we do emotions.
The specific version of the predicament, which I claim has some urgency, at least where I live, is that we are too angry and don’t see our way out. We also live in a world in which shame is given a bad rap and is considered as a primitive and damaging emotion. I think we underestimate the good uses of shame. I do not intend to advocate for shaming but for a mature sense of shame, for the importance of equipping children with a strong sense of responsibility for the health and well-being of the practice of truth seeking and abiding the virtues of cooperative and respectful civic life. When children are inconsiderate, selfish, cruel, or dishonest, we say, “You ought to be ashamed of yourself.” This is not shaming or humiliation, which typically enact cruelty. It tells the child that they have been seen doing what is wrong, but it doesn’t claim that the gaze of others is what they need to be responsive to. “You ought to be ashamed of yourself” is an invitation to internalize certain values and norms so that the child eventually is ashamed of itself when it is tempted to do wrong or does do wrong. A mature sense of shame is good, possibly the best protection for such deep value commitments, and it might play for us something like the positive role it played—possibly that it still plays—in East Asia and South Asia.
My remedy for the predicament is to explore, excavate, and critically evaluate resources for doing anger and shame differently than we now do anger and shame. How do we see what we are doing with these emotions with a clear eye and, at the same time, explore alternative possibilities? There are several ways to undertake such a task. The method I follow here is a variation of what I call “the natural method” (Flanagan 1992): bring to bear any and all resources that pertain to the problem at hand. The question at hand is how to do emotions, specifically anger and shame. The question has both scientific aspects and normative ones. Who are the experts on the scientific aspects of anger and shame? Who has methods for describing and explaining emotions? Who should one go to in order to understand the nature and function of emotions? The answer is that there are entire subfields in biology, psychology, sociology, and anthropology devoted to the nature and function of emotions. I have read this research, so I bring it to the table.
The normative questions include the following: How do we tell when our emotional practices are not doing their job? What is their job? What are emotions supposed to do? How do we tell if and when there are morally problematic aspects to the way we do these emotions (either by the lights of the culture in question or from some external perspective)? How do we judge when we should readjust or modify our practices, norms, and scripts for anger and shame?
Who has methods for addressing these normative questions? My answer may surprise the reader. I do not think philosophers or religious leaders have any special expertise to answer normative questions except in one narrow respect. The narrow respect is that philosophers and religious scholars, but only as scholars, sometimes know a lot about the ways in which different traditions have conceived emotions, whether and how they are thought to support a good life, when they get in the way, and so on. As scholars, philosophers can bring knowledge, in the present case, of different traditions—for example, the way Aristotelians, Stoics, Buddhists, Confucians, Jews, Christians, Kantians, and utilitarians think or have thought about the proper role of emotions in human life, their role in virtue, and so on.
But I do not think that philosophers, including ethicists and social and political philosophers, have any special abilities in the normative spheres or regarding normative questions. All the scientific disciplines that study emotions also contain and deploy normative standards and criteria that are used to assess such things as an individual’s emotional well-being (psychology) or a subculture’s or culture’s emotional well-being (sociology, cultural psychology, anthropology, political science). There is no problem with that, no violation of some intellectual division of labor that restricts “oughtology” to philosophers and religious leaders. Ideas about living good human lives are entirely in the public domain. My method is to invite reflection on the variety of ways people have for doing emotions in service of improving our practices.
There is a background theory of emotions, which I express in summary form and without necessary hedges and exceptions. I intend the theory of emotions to be plausible based on the totality of evidence; it only partially reflects the folk psychology of emotions.
· Emotions are things we do. Emotions are enactments.
· Emotions typically involve exchanges between or among people.
· Terms for emotions, and conceptual categories for individuating emotions, are introduced along with norms for doing emotions.
· There is cultural variation in the norms and scripts for emotions.
· The norms for emotions express when, where, how, and (sometimes) why different emotional responses are acceptable; when they are expected or possibly required; how intense the feelings ought to be; and culturally certified scripts for emotional responses.
· Emotion terms normally refer to functional states or syndromes characterized by a quartet comprising typical causes + typical feelings + typical dispositions to act + typical actions.
· Virtues and vices involve emotions. Classic definitions of virtues depict them as reliable dispositions (to perceive, feel, think/judge, and act) in the way a situation requires. One element of a virtue is to feel correctly—for example, to have a proper sense of shame, not to be shameless or overly severe, to experience and express anger as an injustice warrants, neither too little nor too much. Acquiring the virtues requires in part cultivating the emotions, sometimes taming them (anger, fear), and sometimes expanding their scope (compassion) (Kristjánsson 2010; Sreenivasan 2020).
· Emotion terms do not normally refer to some narrow phenomenal state that is the emotion, to some narrow, natural kind that the terms “sad,” “scared,” and “angry” refer to across all instances of a kind (Barrett 2006, 2017; Scarantino 2012; LeDoux 2019). Emotions might nonetheless refer to well-behaved social kinds (typed in terms of wide functional syndromes) such that both local and global generalizations and predictions about emotions can be made in the human sciences.
· When individuals focus on the phenomenal feeling aspect of their emotions, we should expect there to be differences in feelings that are inherited from the ways an emotion is assembled by caretakers (who represent a culture or subculture). American guilt and shame are conceptually intermixed and have negative valence; American shame is associated with “feeling defeated,” which it is not almost anywhere else on earth (Sedgwick and Frank 1995; Wong and Tsai, 2007); Chinese, Japanese, and Hindu shame has both positive and negative valence. Dispositions to change behavior are associated with guilt in American samples, but they are associated with shame in Javanese and Raramuri (Mexican indigenous people) samples. Regarding anger: Americans feel like punching the person who makes them angry, whereas Japanese feel like leaving the room.
· As emotional norms are acquired, the initial untutored causal sequence (what some call a “basic affect program”) is restructured so what was the initial or typical sequence is remade into a warranted sequence, by which I simply mean the normatively permitted (apt or fitting) expression of the emotion replaces the original expression. If children are originally, naturally, or typically sad if they don’t get more than their fair share of M&Ms, they come to learn that sadness is unwarranted and adjust their desires, expectations, feelings, and responses. Remarkably, children who initially want more than their fair share of M&Ms (all children) eventually learn to be genuinely happy when everyone gets exactly the same amount!
· Cultures have control over how emotions feel insofar as how they feel depends on which other emotions are recruited to build the emotion, and by proximity or distance relations to other emotions. American shame is conceptually linked to anxiety and fear, whereas Japanese shame is linked to love and happiness (Romney, Moore, and Rusch 1997).
· Cultures also have control over the intensity settings and the scope of emotions. If a culture indicates that an emotion like shame ought to be tuned high and, what is different, be all-encompassing in the sense that it devalues the person who is ashamed as a human being, it can do that. Because cultures are not fixed, and because there are multiple models for how to do emotions on offer in multicultures, there are constant renegotiations about legitimate reasons for experiencing an emotion, for intensity tuning, and for scope. For example, shame can be constructed in such a way that I feel that I am a failed human being or, more narrowly, such that I feel that I ought to change some particular aspect of myself.23
My proposal is to renegotiate the dominant scripts for anger and shame. I will now proceed to examine the emotions of anger and shame from the perspectives of cultural psychology, anthropology, and cross-cultural philosophy. The exploration of some of the many different ways that these emotions have been done over the course of human history, and are being done across the earth now, allows us to imaginatively explore the possibility space. The aim is to use the evidence of variation as an invitation to think about how we do these emotions, to think of how we do these emotions as something we are in charge of and that we can change if we have reason to. The method of change might be direct and involve changing ourselves. More likely, it will involve some work on ourselves plus work to change social expectations and social institutions. My personal desire is to make the case that we could do better specifically by tuning down the anger and promoting a healthy sense of shame. Even if the reader entirely disagrees with me about the particular cases of anger and shame, they still might find the method of comparative philosophical anthropology valuable as a source of critical tools for thinking about how we do emotions, and how we might do them better. The method of comparative critique is entirely exportable to any emotion, value, or practice that is done differently by different people.