The Better Angels of Our Nature: Why Violence Has Declined

  So the sense of empathy that gets valorized today—an altruistic concern for others—cannot be equated with the ability to think what they are thinking or feel what they are feeling. Let’s distinguish several senses of the word that has come to be used for so many mental states.10

  The original and most mechanical sense of empathy is projection—the ability to put oneself into the position of some other person, animal, or object, and imagine the sensation of being in that situation. The example of the skyscraper shows that the object of one’s empathy in this sense needn’t even have feelings, let alone feelings that the empathizer cares about.

  Closely related is the skill of perspective-taking, namely visualizing what the world looks like from another’s vantage point. Jean Piaget famously showed that children younger than about six cannot visualize the arrangement of three toy mountains on a tabletop from the viewpoint of a person seated across from him, a kind of immaturity he called egocentrism. In fairness to children, this ability doesn’t come easy to adults either. Reading maps, deciphering “you are here” signs, and mentally rotating three-dimensional objects can tax the best of us, but that should not call our compassion into doubt. More broadly, perspective-taking can embrace guesses about what a person is thinking and feeling as well as what he is seeing, and that brings us to yet another sense of the word empathy.

  Mind-reading, theory of mind, mentalizing, or empathic accuracy is the ability to figure out what someone is thinking or feeling from their expressions, behavior, or circumstances. It allows us to infer, for instance, that a person who has just missed a train is probably upset and is now trying to figure out how to get to his or her destination on time.11 Mind-reading does not require that we experience the person’s experiences ourselves, nor that we care about them, only that we can figure out what they are. Mind-reading may in fact comprise two abilities, one for reading thoughts (which is impaired in autism), the other for reading emotions (which is impaired in psychopathy).12 Some intelligent psychopaths do learn to read other people’s emotional states, the better to manipulate them, though they still fail to appreciate the true emotional texture of those states. An example is a rapist who said of his victims, “They are frightened, right? But, you see, I don’t really understand it. I’ve been frightened myself, and it wasn’t unpleasant.”13 And whether or not they truly understand other people’s emotional states, they simply don’t care. Sadism, schadenfreude, and indifference to the welfare of animals are other cases in which a person may be fully cognizant of the mental states of other creatures but unmoved to sympathize with them.

  People do, however, often feel distress at witnessing the suffering of another person.14 This is the reaction that inhibits people from injuring others in a fight, that made the participants in Milgram’s experiment anxious about the shocks they thought they were delivering, and that made the Nazi reservists nauseous when they first started shooting Jews at close range. As these examples make all too clear, distress at another’s suffering is not the same as a sympathetic concern with their well-being. Instead it can be an unwanted reaction which people may suppress, or an annoyance they may try to escape. Many of us trapped on a plane with a screaming baby feel plenty of distress, but our sympathy is likely to be more with the parent than with the child, and our strongest desire may be to find another seat. For many years a charity called Save the Children ran magazine ads with a heartbreaking photograph of a destitute child and the caption “You can save Juan Ramos for five cents a day. Or you can turn the page.” Most people turn the page.

  Emotions can be contagious. When you’re laughing, the whole world laughs with you; that’s why situation comedies have laugh tracks and why bad comedians punch up their punch lines with a bada-bing rim shot that simulates a staccato burst of laughter.15 Other examples of emotional contagion are the tears at a wedding or funeral, the urge to dance at a lively party, the panic during a bomb scare, and the spreading nausea on a heaving boat. A weaker version of emotional contagion consists of vicarious responses, as when we wince in sympathy with an injured athlete or flinch when James Bond is tied to a chair and smacked around. Another is motor mimicry, as when we open our mouths when trying to feed applesauce to a baby.

  Many empathy fans write as if emotional contagion were the basis of the sense of “empathy” that is most pertinent to human welfare. The sense of empathy we value the most, though, is a distinct reaction that may be called sympathetic concern, or sympathy for short. Sympathy consists in aligning another entity’s well-being with one’s own, based on a cognizance of their pleasures and pains. Despite the easy equation of sympathy with contagion, it’s easy to see why they’re not the same.16 If a child has been frightened by a barking dog and is howling in terror, my sympathetic response is not to howl in terror with her, but to comfort and protect her. Conversely, I may have exquisite sympathy for a person whose suffering I cannot possibly experience vicariously, like a woman in childbirth, a woman who has been raped, or a sufferer of cancer pain. And our emotional reactions, far from automatically duplicating those of other people, can flip 180 degrees depending on whether we feel we are in alliance or competition with them. When a sports fan watches a home game, he is happy when the crowd is happy and dejected when the crowd is dejected. When he watches an away game, he is dejected when the crowd is happy and happy when the crowd is dejected. All too often, sympathy determines contagion, not the other way around.

  Today’s empathy craze has been set off by scrambling the various senses of the word empathy. The confusion is crystallized in the meme that uses mirror neurons as a synonym for sympathy, in the sense of compassion. Rifkin writes of “so-called empathy neurons that allow human beings and other species to feel and experience another’s situation as if it were one’s own,” and concludes that we are “a fundamentally empathic species” which seeks “intimate participation and companionship with our fellows.” The mirror-neuron theory assumes that sympathy (which it blurs with contagion) is hardwired into our brains, a legacy of our primate nature, and has only to be exercised, or at least not repressed, for a new age to dawn. Unfortunately, Rifkin’s promise of a “leap to global empathic consciousness and in less than a generation” is based on a dodgy interpretation of the neuroscience.

  In 1992 the neuroscientist Giacomo Rizzolatti and his colleagues discovered neurons in the brain of a monkey that fired both when the monkey picked up a raisin and when the monkey watched a person pick up a raisin.17 Other neurons responded to other actions, whether performed or perceived, such as touching and tearing. Though neuroscientists ordinarily can’t impale the brains of human subjects with electrodes, we have reason to believe that people have mirror neurons too: neuroimaging experiments have found areas in the parietal lobe and inferior frontal lobe that light up both when people move and when they see someone else move.18 The discovery of mirror neurons is important, though not completely unexpected: we could hardly use a verb in both the first person and the third person unless our brains were able to represent an action in the same way regardless of who performs it. But the discovery soon inflated an extraordinary bubble of hype.19 One neuroscientist claimed that mirror neurons would do for neuroscience what DNA did for biology.20 Others, aided and abetted by science journalists, have touted mirror neurons as the biological basis of language, intentionality, imitation, cultural learning, fads and fashions, sports fandom, intercessory prayer, and, of course, empathy.

  A wee problem for the mirror-neuron theory is that the animals in which the neurons were discovered, rhesus macaques, are a nasty little species with no discernible trace of empathy (or imitation, to say nothing of language).21 Another problem, as we shall see, is that mirror neurons are mostly found in regions of the brain that, according to neuroimaging studies, have little to do with empathy in the sense of sympathetic concern.22 Many cognitive neuroscientists suspect that mirror neurons may have a role in mentally representing the concept of an action, though even that is disputed. Most reject the extravagan
t claims that they can explain uniquely human abilities, and today virtually no one equates their activity with the emotion of sympathy.23

  There are, to be sure, parts of the brain, particularly the insula, which are metabolically active both when we have an unpleasant experience and when we respond to someone else having an unpleasant experience.24 The problem is that this overlap is an effect, rather than a cause, of sympathy with another’s well-being. Recall the experiment in which the insula lit up when a participant received a shock and also when he or she watched an innocent person receiving a shock. The same experiment revealed that when the shock victim had cheated the male subjects out of their money, their insulas showed no response, while the striatum and orbital cortex lit up in sweet revenge.25

  Empathy, in the morally relevant sense of sympathetic concern, is not an automatic reflex of our mirror neurons. It can be turned on and off and even inverted into counterempathy, namely feeling good when someone else feels bad and vice versa. Revenge is one trigger for counterempathy, and the flipflopping response of the sports fan tells us that competition can trigger it as well. The psychologists John Lanzetta and Basil Englis glued electrodes to the faces and fingers of participants and had them play an investment game with another (bogus) participant.26 They were told either that the two were working together or that they were in competition (though the actual returns did not depend on what the other participant did). Market gains were signaled by an uptick of a counter; losses were signaled by a mild shock. When the participants thought they were cooperating, the electrodes picked up a visceral calming and a trace of a smile whenever their opposite number gained money, and a burst of sweat and the trace of a frown whenever he was shocked. When they thought they were competing with him, it went the other way: they relaxed and smiled when he suffered, and tensed up and frowned when he did well.

  The problem with building a better world through empathy, in the sense of contagion, mimicry, vicarious emotion, or mirror neurons, is that it cannot be counted on to trigger the kind of empathy we want, namely sympathetic concern for others’ well-being. Sympathy is endogenous, an effect rather than a cause of how people relate to one another. Depending on how beholders conceive of a relationship, their response to another person’s pain may be empathic, neutral, or even counterempathic.

  In the previous chapter we explored the circuitry of the brain that underlies our tendencies to violence; now let’s see the parts that underlie our better angels. The search for empathy in the human brain has confirmed that vicarious feelings are dimmed or amplified by the rest of the empathizer’s beliefs. Claus Lamm, Daniel Batson, and Jean Decety had participants take the perspective of a (fictitious) patient with ringing in his ears while he got “treated” with an experimental cure consisting of blasts of noise over headphones, which made the patient visibly wince.27 The pattern of activity in the participants’ brains as they empathized with the patient overlapped with the pattern that resulted when they themselves heard the noise. One of the active areas was a part of the insula, the island of cortex that, as we have seen, represents literal and metaphorical gut feelings (see figure 8–3). Another was the amygdala, the almond-shaped organ that responds to fearful and distressing stimuli (see figure 8–2). A third was the anterior medial cingulate cortex (see figure 8–4), a strip of cortex on the inward-facing wall of the cerebral hemisphere that is involved in the motivational aspect of pain—not the literal stinging sensation, but the strong desire to turn it off. (Studies of vicarious pain generally don’t show activation in the parts of the brain that register the actual bodily sensation; that would be closer to a hallucination than to empathy.) The participants were never put in the kind of situation that evokes counterempathy, like competition or revenge, but their reactions were pushed around by their cognitive construal of the situation. If they had been told that the treatment worked, so the patient’s pain had been worthwhile, their brains’ vicarious and distressed responses were damped down.

  The overall picture that has emerged from the study of the compassionate brain is that there is no empathy center with empathy neurons, but complex patterns of activation and modulation that depend on perceivers’ interpretation of the straits of another person and the nature of their relationship with the person. A general atlas of empathy might look more or less as follows.28 The temporoparietal junction and nearby sulcus (groove) in the superior temporal lobe assess another person’s physical and mental state. The dorsolateral prefrontal cortex and the nearby frontal pole (the tip of the frontal lobe) compute the specifics of the situation and one’s overall goals in it. The orbital and ventromedial cortex integrate the results of these computations and modulate the responses of the evolutionarily older, more emotional parts of the brain. The amygdala responds to fearful and distressing stimuli, in conjunction with interpretations from the nearby temporal pole (the tip of the temporal lobe). The insula registers disgust, anger, and vicarious pain. The cingulate cortex helps to switch control among brain systems in response to urgent signals, such as those sent by circuits that are calling for incompatible responses, or those that register physical or emotional pain. And unfortunately for the mirror-neuron theory, the areas of the brain richest in mirror neurons, such as parts of the frontal lobe that plan motor movements (the rearmost portions above the Sylvian fissure) and the parts of the parietal lobes that register the body sense, are mostly uninvolved, except for the parts of the parietal lobes that keep track of whose body is where.

  In fact, the brain tissue that is closest to empathy in the sense of compassion is neither a patch of cortex nor a subcortical organ but a system of hormonal plumbing. Oxytocin is a small molecule produced by the hypothalamus which acts on the emotional systems of the brain, including the amygdala and striatum, and which is released by the pituitary gland into the bloodstream, where it can affect the rest of the body.29 Its original evolutionary function was to turn on the components of motherhood, including giving birth, nursing, and nurturing the young. But the ability of the hormone to reduce the fear of closeness to other creatures lent itself over the course of evolutionary history to being co-opted to supporting other forms of affiliation. They include sexual arousal, heterosexual bonding in monogamous species, marital and companionate love, and sympathy and trust among nonrelatives. For these reasons, oxytocin is sometimes called the cuddle hormone. The reuse of the hormone in so many forms of human closeness supports a suggestion by Batson that maternal care is the evolutionary precursor of other forms of human sympathy. 30

  In one of the odder experiments in the field of behavioral economics, Ernst Fehr and his collaborators had people play a Trust game, in which they hand over money to a trustee, who multiplies it and then returns however much he feels like to the participant.31Half the participants inhaled a nasal spray containing oxytocin, which can penetrate from the nose to the brain, and the other half inhaled a placebo. The ones who got the oxytocin turned over more of their money to the stranger, and the media had a field day with fantasies of car dealers misting the hormone through their showroom ventilating systems to snooker innocent customers. (So far, no one has proposed spraying it from crop dusters to accelerate global empathic consciousness.) Other experiments have shown that sniffing oxytocin makes people more generous in an Ultimatum game (in which they divide a sum while anticipating the response of a recipient, who can veto the deal for both of them), but not in a Dictator game (where the recipient has to take it or leave it, and the proposer needn’t take his reaction into account). It seems likely that the oxytocin network is a vital trigger in the sympathetic response to other people’s beliefs and desires.

  In chapter 4 I alluded to Peter Singer’s hypothesis of an expanding circle of empathy, really a circle of sympathy. Its innermost kernel is the nurturance we feel toward our own children, and the most reliable trigger for this tenderness is the geometry of the juvenile face—the phenomenon of perception we call cuteness. In 1950 the ethologist Konrad Lorenz noted that entities with measurements ty
pical of immature animals evoke feelings of tenderness in the beholder. The lineaments include a large head, cranium, forehead, and eyes, and a small snout, jaw, body, and limbs.32 The cuteness reflex was originally an adaptation in mothers to care for their own offspring, but the triggering features may have been exaggerated in the offspring themselves (to the extent they are compatible with its own health) to tilt the mother’s response toward nurturance and away from infanticide.33 Species that are lucky enough to possess the geometry of babies may elicit the awwwww! response from human beholders and benefit from our sympathetic concern. We find mice and rabbits more adorable than rats and opossums, doves more sympathetic than crows, baby seals more worthy of protection than mink and other weaselly furbearers. Cartoonists exploit the reflex to make their characters more lovable, as do the designers of teddy bears and anime characters. In a famous essay on the evolution of Mickey Mouse, Stephen Jay Gould plotted an increase in the size of the rodent’s eyes and cranium during the decades in which his personality changed from an obnoxious brat to a squeaky-clean corporate icon.34 Gould did not live to see the 2009 makeover in which the Walt Disney Company, concerned that today’s children expect “edgier,” more “dangerous” characters, unveiled a video game in which Mick’s features had de-evolved to a more ratlike anatomy.35