6
THE MILK OF HUMAN KINDNESS
HER WIDE, PALE belly scrapes across the sand as she heaves herself forward. Her thin, formless legs struggle for purchase, but her shoulders are strong. Again and again she pulls herself onward, gaining a few inches each time. The going is slow. The sand slopes unrelentingly upward, and her body weighs over 200 pounds. It leaves a ragged trench in the sand behind her, one that is easily visible even by moonlight. She stops often to rest. At last, about half an hour after first emerging from the ocean, she reaches her destination: the edge of the north Florida beach where the flat-packed sand yields to sugary heaps shaded by stalks of sea oats and creeping purslane. She has seen this spot before, although not for many years, and never from quite this vantage point. But she’s certain it’s the right place. After a pause to recoup her energy, she settles herself in and begins to dig.
She isn’t seeking treasure, nor would she know what to do with it if she found any. She’s here to build a nest. Inside, she will deposit dozens of rubbery, Ping-Pong ball–shaped eggs containing the embryos of her children, just a few of the thousands of loggerhead sea turtles she will produce over her lifetime.
The nest ready, she positions herself above it. The leathery tube of her cloaca bulges as the first egg of the clutch pops from it, shiny with fluid, bouncing off the side of the pit before coming to rest at the bottom. Dozens of siblings tumble out after it in succession, soon burying it. She never gives the gleaming pile so much as a glance. When her oviduct is empty and the pit is full, she flips sand back over the nest in a chaotic spray that coats her own head and shell as well, until the eggs are hidden from view and protected from the sun and wind and gulls and crabs. Afterward, she tamps the pile into a firm lump with the underside of her shell. It doesn’t look like much, but all this spraying and tamping is a significant moment in the lives of her offspring, as it represents the only care they will receive from their mother. Her job complete, she drags herself laboriously back down to the sea, where she disappears beneath the surf along with the last thoughts she will ever give to her babies.
Assuming this is a typical loggerhead nest, it holds about 115 eggs. Many of them will never hatch, some because they were never fertilized. Others will succumb to any of the various hazards that can befall a two-inch-long baby developing unattended in a shallow pit in the sand for several weeks, including being killed by fire ants or crabs or being discovered by poachers or overheating or getting flooded during a storm surge. Of the ones that do hatch and manage to dig their way straight up through a foot of eggs, fellow hatchlings, and packed sand, plenty will never make it to the water. Some will become disoriented by the bright lights of beachfront hotels and die of dehydration. Others will get picked off by shoreline predators like raccoons and gulls. Those that do reach the ocean will have a fighting chance at survival, but they remain so tiny and so vulnerable and face so many obstacles that nearly all of them will ultimately die young too. It’s estimated that only one of every 1,000 loggerhead babies reaches adulthood—that’s a one-in-ten chance that even a single hatchling from this nest will survive to maturity. No wonder their mother didn’t bother getting sentimental about them.
I encountered these hatchlings several weeks later and was not as lucky. If anything, “sentimental” underplays my response to them. This fact—that the same baby turtles could be forgotten and left quite literally for dead by their own mother and also be the object of profound and faintly embarrassing sentimentality from a passing human—is deeply intertwined with, and essential for understanding, the origins of humans’ capacity for altruism.
It was early July when my family and I recently visited my in-laws in Ponte Vedra Beach, Florida. That happens to be hatching season in a prime loggerhead nesting area. Turtles who lay eggs there are fortunate to have all of their nests recorded and monitored by the stalwart volunteers of the Mickler’s Landing Sea Turtle Patrol, a local conservation group. Every spring and summer, volunteers patrol four miles of coastline for signs of turtle nesting. When they spot telltale crawl marks ending in packed-down lumps of sand, they pound wooden stakes into the sand around the nest and scrawl on them the estimated date the eggs will hatch. The patrol’s policy is to wait until three days after the first hatchlings emerge from a nest, then dig up anything that remains to log how many eggs were laid and the fate of each. Was it fertilized? Did it hatch? Did the turtle successfully exit the nest? The best part of a dig is that anywhere from a few to a few dozen turtle hatchlings that have not yet dug their way out are usually unearthed, and are then released by the patrol to make their way to the sea.
My mother-in-law Krista brought me to watch this process with our young daughters on a sweltering Saturday night when a loggerhead nest was scheduled to be dug up. It was about the saddest thing I’ve ever witnessed. Two patrol members found the appointed spot in the sand, and we all circled around. A few hatchlings had emerged three days prior, so hopes were high. One volunteer carried a small red pail into which any new hatchlings would be placed for monitoring before being released. The other volunteer began to dig. Her gloved hands gently carved through the sand as she scraped away layer after thin layer, like an archaeologist, or a sculptor. When the first glimpse of smooth, pale shell appeared, we all held our breath. But it was a dud—an egg that had never developed. Its saggy profile and yellowish tint gave it away. After a few more duds, the first sign of hope appeared. The fragment of a clean, white shell was a sign that at least one hatchling had emerged alive. More fragments followed, as well as some fertilized but unhatched eggs, which the volunteer placed to the side. Eventually, the volunteer’s sliding hands revealed the thing we had most hoped to see: a thin sliver of black no longer than the first joint of my finger—the flipper of a baby loggerhead. But the flipper did not flip. It lay motionless in the sand, lifeless as the shell fragments surrounding it. The volunteer covered it with her hand, hiding it from the ring of hopeful children gathered around, then pried the rest of the body from the sand with her fingers. As she lifted it up, the flipper swung out from under her curled thumb. Carefully, she moved the body to a trough carved in the sand for counting purposes. Another trough already held the eggs, duds, and fragments that had been uncovered. This trough would hold only death.
After a while, there was no point trying to hide the bodies anymore. One still, dark form after another emerged from the sand, turtle upon turtle in various stages of decay—perhaps ten in all. A few were putrid and hardly recognizable. But several, like the first, were still perfect in every feature. My six-year-old daughter, to her credit, was not frightened by them, only curious, so we moved around to the trough to get a better look. I stared for a long time at the first one to be dug up. Its tiny chin rested peacefully on the sand beneath its beaky nose and the large dark whorls of its closed eyes. Its shell was a mosaic of irregular black pentagons fit neatly together. Its front flippers were outsized, like the long limbs of a foal, and their trailing edges bore intricate notches and ridges that I imagine would somehow improve hydrodynamics. Such a perfectly formed thing it was in every detail; a sculptor would weep with joy to create an object of such beauty.
But of course a mere sculpture’s marvels are all external. What lay inside this turtle was, if anything, even more wonderful. Within its chest blossomed an intricate four-chambered heart and lobed lungs, life-giving machines beyond the reach of anything science can yet replicate. Behind its closed lids lay eyes of such complexity that Darwin himself wrote that it seemed “absurd in the highest possible degree” to imagine such a thing evolving through natural processes. A wholly formed brain had pulsed inside its skull—millions of neurons linked together in intricate webs ready to support this turtle’s ability to escape its shell and dig and breathe and learn—maybe even feel. All built for nothing. All built so that the sum total of all of these wondrous parts could lay moldering under the setting sun, a checkmark in the patrol’s “dead hatchlings” column. The “live hatchlings” column stayed empty. The horrible waste of it all was overwhelming.
We returned the next evening with hopes for something less depressing, but the first of the two nests to be dug up was even worse. Whereas the first nest had at least born signs of hope and life, including intact fertilized eggs and many more shell fragments than dead hatchlings, the second was an awful avalanche of rotting hatchling carcasses, so many so badly decayed that it seemed unlikely that more than one or two had possibly made it out alive. The verdict was that fire ants from a nearby colony had attacked the nest and killed all the eggs and hatchlings, some still only partway out of their shells.
“This,” said the stoic volunteer by the time she had finally dug to the bottom of the whole horrid mess, “is not a happy nest. I’m so sorry.”
One more nest remained to be dug up before we were to return home to Washington, DC. I hardly had the heart to watch the whole thing again, but the girls begged to stay, so we did. I watched stonily, trying to detach myself from the process this time. The first scrap to appear in this last nest was positive: a shell fragment. More followed. No duds yet, no dead hatchlings. “Don’t get your hopes up, don’t get your hopes up,” I muttered to myself sternly. Then—unmistakable movement! The crescent of a tiny dark flipper flung itself from the sand and scrabbled the air. A thumb-sized head followed, craning toward its first glimpse of the sky, its round, black eyes blinking back the light and crumbs of sand. Jubilation ensued. Children shrieked. My daughter and sister-in-law laughed and hopped up and down. I wanted to laugh myself. It was as winsome and intricate as all the poor motionless ones we’d already seen, but to see its limbs flapping with energy and life was pure joy. More soon followed—eight live hatchlings in all. That is a smallish number, but it felt like a cornucopia. The crowd welcomed each turtle into its sandy new world with a cheer. The volunteers placed them one by one into the red bucket, where they stretched their wrinkled necks toward the setting sun and stepped on each other’s faces while the patrol volunteers finished digging out the nest and making final tallies. They were impossibly cute.
Then it was time to release them. The volunteers aim to interfere minimally with the natural process, so they wouldn’t bring them to the ocean, a thirty-second walk away at the most. Instead, they gently tipped the bucket up next to the nest. The hatchlings had to do the rest themselves.
Oh, the agony and delight of watching them struggle to reach the water! They paddled over the sand like windup toys, their flippers moving in perfect metronomic rhythm—flip-flip, flip-flip, flip-flip, flip-flip. They slowed only slightly as they moved over and around the flotsam and hillocks they encountered. Their certainty of purpose was comically at odds with their miniature bodies. They fanned out as they moved toward the ocean, and the volunteers struggled to keep an eye on all of them and the phalanx of eager watchers ushering them onward, many of us holding our arms out at protective angles. One poor hatchling lost his way repeatedly, heading each time for a thicket of children’s legs. More than once he narrowly avoided being crushed by an errant flip-flop. Each near-miss yielded groans of panic from the onlookers and shouts of: “Move back! Move back!” It was so hard to keep track of all eight. Any attempt to avoid stepping on one risked crushing another. My stomach churned at the thought of one being trodden on—or worse yet, treading on one myself by accident. I didn’t think I could stand seeing another of their tiny bodies go limp, not after having watched them struggle so mightily to reach the lip of the vast churning ocean that was their only hope of survival.
Despite our efforts, tragedy befell the hatchling that kept getting lost. He moved too far from the group, and a gull seized its opportunity to snatch him up. He was carried high in the air while a few people who’d seen it happen screamed invectives at the bird. Was it our screaming? Did the turtle put up a fight? For whatever reason, the gull dropped him. He plummeted down to the sand, fifteen feet at least. A collective gasp. Then, “He’s still alive!” shouted a triumphant volunteer. Not only alive but undeterred, he righted himself and returned to his mission. The onlookers closed ranks around him, ushering him the last few yards of his journey. He was the last one in, but he too made it to the ocean at last. When he reached the waterline and the first gush of bubbling froth hit him full in the face, I swear I saw a look of astonishment in his eyes. He froze for a moment at the novelty of it; yet another incredible event overwhelming his newly formed senses on this, his first day in the world. He quickly recovered and started to paddle, and soon the rushing tide had carried his small form from view.
Walking back to the car afterward, the four of us felt foolishly proud of the hatchlings for making it to the ocean, for having already beaten so many of the odds against them. Everything about this feeling was ridiculous, of course. They weren’t ours. They were reptiles who had been buried in the sand by a turtle, were monitored and dug up by patrol volunteers, and would have made it to the ocean on their own whether or not we had been personally cheering them on and yelling at people not to step on them. If I had happened to visit Florida on a different weekend, I would have never known they existed. So why had I—and thirty other beachgoers—become so instantly protective of and invested in them? So willing to spend time and energy taking what feeble measures we could to ensure that they made it to the water alive—shouting at each other and even at a sea gull (who was just hungry; why did we take the turtle’s side?), holding our arms out like beach bouncers as though this would somehow keep the turtles safe? Why did I feel so delighted watching the hatchlings’ paddling flippers and their sweet, beaky faces, and so terribly hopeful that they might continue to beat the odds and survive? Why did I feel so—there is no other way to put it—sentimental about them when their own mother spared not a single thought for any of them after she had tamped her final tamp over their nest and most likely would have ignored them if she had happened to be on the beach again that evening and seen them scramble past?
The answer traces back many millions of years. The short version is that I am a descendant of creatures called cynodonts, and loggerhead turtles are not. The longer version of this answer may provide the remaining pieces of the puzzle that is extraordinary altruism.
Loggerhead turtles are an ancient species, and a successful one in the scheme of things. They have been leaving their eggs under lumps of sand on beaches around the world for some 40 million years. They and six other existing species of sea turtles, collectively known as chelonioidea, can trace their ancestry back to the progenitors of all modern turtles who emerged more than 200 million years ago during the Triassic Period. That means turtles predate even dinosaurs and birds, who are the dinosaurs’ last surviving descendants. (I know the idea that birds and not turtles are the true heirs of the dinosaurs can seem a little silly—especially if the birds you’re thinking of are pigeons or parakeets—but just watch a great blue heron stalking fish in a stream sometime, its scaly legs and clawed feet parting the reeds while its fierce, pointed head swivels atop its long neck, and it will seem much more obvious.)
Humans and other mammals are the descendants of neither dinosaurs nor turtles, but of hamsterlike creatures called cynodonts whose lineage diverged from other four-limbed animals about 250 million years ago. Like modern mammals, cynodonts were furry and warm-blooded, but they also laid eggs. Together, these traits left them backed into something of a corner when it came to reproducing. To produce warm-blooded babies able to maintain their own body temperatures after hatching would require that the babies be very large—so large that gestating and excreting eggs big enough to contain them would have killed their tiny mothers. The only other option was to produce very small eggs, out of which very small and developmentally immature babies would hatch. But such babies would be so immature that they would be incapable of even supporting their own metabolism without someone keeping them constantly warm and nourished. They would be, in other words, altricial, which is the biological term for babies born immature, helpless, and dependent. That this word sounds like “altruism” is not a coincidence. Both words are derived from the Latin alere, meaning “to nourish.” Altricial babies can be contrasted with those who are precocial, or developmentally mature and self-sufficient. Human babies are quintessentially altricial, as are the babies of many other mammals and nearly all birds, whereas sea turtles and other reptiles and fish and any other animals capable of showing anything like “certainty of purpose” within minutes of being born are precocial.
So this was a pickle. How could cynodont mothers simultaneously keep their newly hatched babies both warm and fed around the clock? There are only two viable solutions to this problem, and cynodonts came up with one of them. Everything that we know about modern and ancient mammals points to the conclusion that the critical evolutionary development that allowed cynodonts to flourish was that cynodont mothers developed the ability to keep their tiny, altricial young both warm and fed by turning their bodies into food.
In other words, they made milk.
Milk is among the more spectacular evolutionary developments ever to come along. It’s easy to take for granted that mammalian mothers literally dissolve their own flesh and bones and excrete it as food through their nipples because we don’t know any differently. But imagine if you one day discovered that you could shoot hamburgers out of your armpits at will. That’s basically how incredible lactation is. Except milk is even better nourishment than a hamburger. It contains water and salts essential to sustaining life; nutritive fats, proteins, and sugars; various other vital substances, from calcium and phosphorus to the indigestible oligosaccharides that serve as prebiotics in the gut—all served perennially fresh, liquefied, and prewarmed directly from mom.
This elixir is directly responsible for the emergence of all the diverse modern-day mammals that populate the earth today, humans included. We could never have survived without it. It’s little wonder that mammals were named for the very glands that produce it. Milk allows metabolically greedy mammalian babies to be born small and underdeveloped and to survive their vulnerable infancy by receiving all of their nourishment from their mother while staying warm and expending almost no effort—not even chewing. It allows mammalian babies to thrive in nearly any ecological niche, since they need neither to compete with adults for food nor to find an alternative—and likely inferior—food source. This makes milk a likely reason that so many mammals survived the aftermath of the cataclysmic asteroid strike that killed off the dinosaurs. Milk also allows mammals to sustain unusual and beneficial growth patterns, like disproportionately rapid growth of the head and brain soon after birth. And finally, milk ultimately transformed the social life of the mammals who consume and produce it as well. Consuming milk is no mere luxury for mammalian babies but a necessity, so milk is the reason that mammalian babies remain dependent on and attached to their mothers for weeks or months or years—and the reason that their mothers remain attached to their offspring as well.
Milk is, in other words, the prime mover behind many of the psychological, behavioral, and social features that separate mammals from all the creatures who came before them, and from many who came after as well. This was an inevitable outcome, since the ability to produce milk would be a useless adaptation in isolation. It’s only worthwhile if accompanied by a suite of other changes that ensure that infants actually benefit from the milk.
What is the first and most essential of these changes?
Love.
Or, if it makes you uncomfortable to think of the same animals that we eat for dinner and experiment on in laboratories experiencing love, call it caring, which is love’s behavioral expression. It doesn’t really matter what you call it—it all boils down to the same thing.
And that thing is that producing milk to sustain altricial offspring is useful only if the mother stays in frequent and very close contact with her offspring to feed them the milk—literally pressing her body against them or cradling them or hovering directly above them for stretches of time every day for weeks or months to allow them to drink their fill. And no mother is going to do this every day, without fail, for every infant or litter of infants she produces, breeding season after breeding season, unless some powerful motivational tether keeps pulling her back to her babies again and again when she could be out doing something easier or more fun or interesting than hanging around being drained of her own liquefied flesh and bone. And that tether is love.
That love underlies the urge to be in physical contact with the newborn offspring as much as possible. To smell their intoxicating smell, to gaze at the endless wrinkles and curves of their small, strange bodies. To find their simple presence wildly reinforcing, and to feel the fierce longing of an addict when separated from them. To want to stroke them or cradle them or lick them or nuzzle them (depending on the species and available appendages) and protect them from the amorphous terrible dangers of the world. And of particular interest, to care about their welfare: to fly into action when they are distressed or in danger, and to feel pleasure in their contentment. Human mothers share these urges with every mammalian mother alive today—all thanks to our cynodont ancestors.
I am a mother myself, twice over, and I am all too aware of how heady and overpowering maternal love can be. That said, the births of my daughters did not represent the first or only times I have experienced fierce, heady, gripping love. You may recognize many of the features of maternal love from other loving relationships you’ve had—romantic love in particular has many of the same intoxicating qualities. There is a reason for that. It is thought that the capacity for love and caring of all kinds, from romantic to filial to sibling love, even love of friends or pets, grew out of the capacity for maternal love. Once the proto-mammalian brain was equipped with the wholly novel and evolutionarily necessary capacity to care about the welfare of other beings outside the self, there was no limit to what other kinds of love could theoretically be felt. It’s little wonder that the ethologist Irenäus Eibl-Eibesfeldt viewed the emergence of maternal nurturing as “a turning point in the evolution of vertebrate behavior—one of those celestial moments that [a poet] would call a star hour.”
I don’t think that is an understatement. As far as we know, the vast, cold universe existed for billions of years without love existing anywhere within it. Then, following the need of a few furry protomammals to keep their young fed and warm, it did. It was an explosion into being as magnificent as the birth of any star.
One of maternal love’s most impressive properties is its capacity for spontaneous combustion. Ideally, a burst of love goes off like gunpowder inside every new mother’s brain when she gets her first look at or whiff of her offspring and feels their first fumbling efforts to feed. The love needs to come on rapidly and powerfully in part because it needs to override or eliminate any fear she might otherwise have of the strange new creatures before her. Altricial babies need food and warmth right away—there is no time for a protracted warming-up period.
If it strikes you as silly to be afraid of a baby, that’s only because you are the descendant of cynodonts. Objectively speaking, altricial mammalian babies look funny and smell funny and make horrible, strange noises, and to top it all off, they are by definition complete strangers. Their mother has never encountered them before in her life. Normally, a little healthy fear of anything strange and new is a good thing. But a mammalian mother can’t be put off by her babies’ funny looks or smells or newness. Her love must draw her near to these little strangers immediately and keep her there, giving them food and warmth despite everything. And, as long as she’s going to commit all these resources to her babies, it would really be ideal if her babies survived to make her investment worthwhile. So she needs to not only stay physically close but to pay close attention to the welfare of each baby too. Is it content? Or is it upset? Does it need something? More food? Warmth? Did it somehow stray from the nest or lose track of her? Is it trying to get back to safety? Does it need cleaning? Is it hurt? Is it in danger?
That all of these new features—altricial young, milk production, attentive maternal care, and intense investment in each offspring—came bundled together might seem improbable bordering on miraculous. But this bundling is in perfect accordance with something called life history theory. According to the theory, the reproductive strategies that a given species can adopt lie on a continuum. At one end of the continuum sit species like loggerhead turtles that are r-selected, meaning that their reproduction is limited by the available resources. R-selected species tend to produce precocial offspring, invest few resources in them, and provide them with little or no care. As a result, most of these offspring die young. But r-selected species produce so many offspring that only a few need to survive to keep the species going. It’s a mass-production approach to reproduction that seems terribly heartless and wasteful to us because our ancestors shifted so dramatically away from it. Cynodonts and most of their descendants follow more K-selected strategies, which you could think of as an artisanal approach to reproduction. Mothers who follow a K–selected strategy produce altricial offspring and so must devote time and energy to keeping them alive. As a result, they can’t have nearly as many of them. Instead, they lavish the few babies they have with nourishment and care to maximize their odds of surviving to adulthood. We humans, with our small numbers of desperately needy and altricial offspring who require well over a decade of parental time and energy before they reach maturity, are about as K-selected as an animal can get.
The changes in reproductive strategies that our small, warm-blooded ancestors underwent to keep their babies alive beautifully capture the reason why a mother loggerhead could lay her eggs in the sand and then disappear without giving the progeny they contain a further thought while I found myself overwhelmed by anxiety and protective feelings for her hatchlings. Her chelonioidea brain simply could not conceive of love. It doesn’t have the requisite wiring, because it never needed to. But my K-selected mammalian brain comes prepared to love and dote on babies, to treat each one like a precious gem that requires abundant care and nurturing and protection.
“Okay,” you might reasonably retort, “that would make sense if you were talking about caring for your own babies. But why on earth would your K-selected mammalian brain have prepared you to care for turtle babies?”
In truth, many mammals would not care at all about turtle babies. They would be much more likely to step over them or on them, or to eat them, than to usher them to the water or heckle gulls on their behalf. These differences stem in part from how precisely tuned a species’ parental nurturing response is. All mammalian species must come prepared to care for their own babies. But the degree to which they will care for other babies varies widely. Ruminants like sheep, for example, are usually uninterested in any babies other than their own. After a newborn lamb first drops to the ground in an amniotic clump, his mother will spend the next several minutes assiduously licking him off, then nudge him to begin nursing. Their licking and nudging and nursing allows the mother and lamb to learn each other’s unique smells and to imprint upon one another. After a few hours, the imprinting window will close, and it is very difficult to get it to open again. Thereafter, the ewe will reserve all of her nurturing and milk for her own lamb and skirt away from or even butt away any other lambs who try to intrude. If a lamb is orphaned, it is unlikely to be adopted and will probably starve, even in a flock full of bulging udders. You can bet that a ewe who would complacently allow a herd-mate’s lamb to starve to death would not give a fig about a turtle hatchling.
Compare this to the open-armed mothering that can be observed in the humble rat. Rats are devoted mothers who will work strenuously to stay close to their pups. They will cross painful electrified metal grids—the rat equivalent of walking over coals—if their babies are marooned on the other side. Mother rats will tolerate more such suffering to reach their pups than hungry rats will undergo to reach food, or thirsty rats to reach water. But perhaps more remarkably, rats will struggle and suffer even for babies that are not their own, even for babies they have never seen before.
My colleague Stephanie Preston, a psychologist at the University of Michigan, recently unearthed a long-forgotten 1968 study by William Wilsoncroft that demonstrated this fact. Wilsoncroft and his students selected five pregnant rats for their study and taught each one that when she pressed down on a bar in her testing box, a piece of Purina rat chow would come tumbling down a nearby chute into her dish. If pregnant rats experience the same rapacious hunger I did when I was pregnant, I’m sure they were all too happy to learn this task.
Then, after the mother rats gave birth, they were given one day to bond with and nurse their pups before the researchers whisked the pups from their nests. When the mystified mothers emerged from their empty nests to return to bar-pressing, they once again found that chow came tumbling down the chute. At least, it did after the first six bar presses. The seventh time a mother pressed the bar it was followed, not by the dry rattle of a piece of rat chow, but the soft thumping of a pink, hairless pup. Down the chute tumbled one of the mother’s lost pups, right into her dish! I leave it to you to imagine what the mothers made of this development. Two of the five immediately did what attentive mother rats do, which was to pick up the pup gently in her teeth and carry it back to the safety of the nest, some three feet away. The other three pressed the bar a few more times first and found that after each press another pup tumbled down. Soon all the mothers were on the same page—pressing the bar and toting their lost pups to their nest.
But the experiment wasn’t over. On the thirteenth bar press, something new happened again: this time the pup who tumbled down the chute and into the cup wasn’t one of the mother’s own pups but a stranger. She had never encountered this pup before in her life. Now what? If mother rats were like ewes, the strange pups would be doomed to starve in the dish. Fortunately, mother rats take a different approach to mothering, which is a pup is a pup is a pup. None of the five so much as hesitated. They all picked up the strange pups, carried them back to the nest, and deposited them among their own pups. Then they went back to the bar and pressed again—and again and again, bringing pup after pup back to the nest, both their own pups and pups that were strangers.
How long do you think this process went on? Ten minutes? An hour? The surprising answer is that it never did end. The ones who put a stop to the parade of pup-toting were the researchers, who after three hours of refilling the hopper with pups ended the experiment in exhaustion. During this time the supermom of the group, Rat 5, retrieved an astonishing 684 pups from the dish (although in reality it was the same set of 20 or so pups, recycled many times over). That’s a pup every fifteen seconds for three hours, without a break, carted back to the nest for a total distance of over 2,000 feet. But even the least energetic of the mothers retrieved a total of 247 pups back to the nest. Moreover, all the mothers were equally attentive when confronted with both their own and strange pups, which suggests that for some mammalian mothers, anything that trips the “baby!” alarm is a call to maternal action (even if, again, this alarm did not always sound equally urgently in all the mothers).
This take-all-comers approach to mothering is not unique to rats. Many other group-living mammals excel in allomothering, which literally means “other mothering,” or taking care of infants other than one’s own. Allomothering can include anything from toting around other mothers’ babies to protecting them from danger to nursing them. In some cases, allomothers even foster or wholly adopt orphaned animals. Allomothers are often sisters, aunts, and other adolescent or adult female relatives, but not always. Females of many species allomother unrelated infants as well. And despite what the term implies, males also allomother, contributing to nearly every aspect of parenting other than nursing across a variety of species.
The emergence of allomothering within a species is largely driven by the neediness of the species’ young. It’s a behavior that is at least three times more likely to arise in animals that bear altricial babies. Hence sheep, which are born woolly and able to stand and walk within minutes, do not allomother, whereas rats, which are born naked, blind, and totally helpless, do. Among the many other mammals renowned for their energetic allomothering are meerkats, seals, sea lions, jackals, wolves, domestic dogs, and lions. Consider for a moment the fact that these highly nurturant species are also all predatory carnivores. Their ferocity on the hunt makes for a striking juxtaposition with the gentleness and devotion that female jackals, wolves, and lions show not only their own but other females’ young. The same lioness who runs down and eviscerates a wildebeest one minute may well groom and nurse another female’s cubs the next. It also makes for an interesting contrast with sheep, who are often depicted as quintessentially innocent and gentle creatures but can be callous or downright cruel toward each other’s lambs. It’s a good reminder that there is nothing unnatural or even surprising about the capacities for genuine ferocity and genuine nurturing care coexisting within a single species, or a single individual.
One spectacular demonstration of this is a Kenyan lioness dubbed Kamunyak, or “Blessed One,” who in 2001 was discovered in close proximity to an oryx calf that, to the astonishment of the locals who discovered them, was not dead. Oryxes are antelopes, which are favored prey for lions. So what was going on with Kamunyak? Nobody knows for certain, but it is thought that she may have chanced upon the calf and its mother, and the mother had bolted. Ordinarily that would have been the end of the calf. But not only did Kamunyak not attack the abandoned calf, she adopted it. For days she could be seen lying or walking by its side, occasionally even grooming it with her rough tongue. She responded attentively when it cried out and was fiercely protective of its safety, keeping it away from humans who ventured too close and at one point chasing off a leopard. She wouldn’t leave the calf long enough to find food for herself, so she must have been very hungry. Yet still she cared for the little calf until it ultimately met a sad end—it was killed and eaten by another lion after it ventured out of Kamunyak’s sight for a moment. By all appearances, Kamunyak grieved the death of her charge, rushing the lion who had attacked it and then, when nothing more could be done, wandering around roaring mournfully. But it wasn’t long before Kamunyak was spotted with another calf—the second of six oryxes that she would ultimately attempt to adopt. Her mothering skills improved over time; starting with her third adoptee, she briefly allowed the calves to nurse from their mothers before chasing the mothers away again.
Although her behavior was undoubtedly unusual—visitors to her park flocked to see a lion quite literally lying down with, if not a lamb, a similarly vulnerable antelope calf—Kamunyak was no singularity. Another lioness in Uganda was recently seen toting around an antelope calf by the scruff of its neck as though it were an unusually gangly cub; later she was spotted allowing it to try (fruitlessly) to nurse. And in 2014, photographers in Botswana observed yet another lioness kill an adult female baboon that was carrying a baby at the time. The infant tried to escape but wasn’t yet strong enough to climb a nearby tree. The lioness walked over to investigate and, spotting the baby, pawed at it gently with her massive forefoot. Eventually, she picked the baby up—as though it was one of Wilsoncroft’s tumbling rat pups—and carried it a short distance away, whereupon she lay down in the shade and deposited it between her front feet. The photographers captured images of the baby trying to suckle from the lioness’s bristly chest, and of the lioness charging a male who attempted to encroach. Retrieval, caressing, nursing, defense—the full suite of mammalian mothering behaviors—were all shown by a lion for a baby baboon!
Despite how bizarre it might seem for a wild lion to tenderly nuzzle and care for a baby antelope or baboon, this behavior is not difficult to explain in the context of allomothering. To effectively allomother each other’s young, lions must maintain a relatively relaxed threshold for what sets off their “baby!” alarm and triggers their urge to provide protection, affection, and care. And just as is true for rats, both the threshold and sensitivity of the “baby!” alarm and the energy with which maternal care is provided inevitably vary among individuals, just as surely as individuals vary in size and coat color and countless other individual traits. Extreme variation in maternal sensitivity appears to lead some unusually motherly lions to adopt baby oryxes and baboons, both of which do resemble lion cubs in many respects: they’re of similar size and coloration and have infantile features that are common to most species, like large eyes and foreheads and small noses and jaws.
Although perhaps only the very most motherly lionesses engage in cross-species adoption, this behavior is near-normal for species that are even more allomotherly, like domestic dogs. Stories abound of dogs who have of their own accord taken up the challenge of mothering infants of diverse species, including those that are dogs’ natural predators or prey. One of my favorites is Mimi, a grizzled ten-year-old female chihuahua owned by a Florida woman named Jeanette Young. In 2007, Young took in a litter of four squirrel pups that her son-in-law had found in a downed nest. Initially she didn’t let Mimi near them, which seemed reasonable—there may be no wild animal that elicits more crazed canine hunting behavior than squirrels. But, as Young told CNN, Mimi became progressively more obsessed with the pups, continually approaching them and whining and, as Young put it, “carrying on.” So finally, although still somewhat fearful that Mimi might try to attack or eat the pups, Young put the nest on the ground to see what Mimi would do. To her surprise, Mimi almost immediately adopted the pups as her own. Right away she began licking them vigorously, and soon she became protective of them, refusing to venture far from them or to allow Young to approach them. Most spectacularly, after a short while watching over “her” pups, Mimi began producing milk for them and allowing them to nurse. This despite not having given birth to pups of her own for four years!
This is one of roughly a billion stories about domestic dogs caring for animals of other species. In some zoos, it’s actually protocol to give orphaned animals like tigers or cheetahs to dogs to raise. An Australian shepherd named Blakely has been designated the “resident nursery companion” at the Cincinnati Zoo, where he (yes, he) has helped to raise dozens of zoo animals that were abandoned or orphaned, including cheetahs, a warthog, and a skunk. His primary job is teaching his charges mammalian social behaviors by gently roughhousing and playing with them. In other zoos, female dogs nurse orphaned baby animals of various species; for instance, a golden retriever named Izzy at a Kansas zoo raised three newborn Bengal tigers along with her own pups. The first time they were introduced, the cubs took immediately to their foster mother, and Izzy to them, and she nursed and nurtured them successfully until their first birthday approached and zoo owners finally separated them. But at no point did either dogs or tigers exhibit hostility toward each other. On what would be their last day together, photographs show Izzy nuzzling the face of one of the 140-pound tigers she raised. Many other examples abound of domestic dogs nursing or otherwise caring for infant tigers, lions, cheetahs, red pandas, deer, wild dogs, pigs, ducks, owls—you name it. There seems to be no limit to the kinds of animals that dogs will allomother.
Although carnivores like lions and dogs can be surprisingly good allomothers, primates are, as a group, even better. From tiny tamarins and marmosets to siamangs, many primates are wonderful allomothers (although, interestingly, great apes like chimpanzees and orangutans allomother relatively little).
But the real allomothering superstars are humans. As the anthropologist Sarah Hrdy describes in her classic book Mothers and Others, we humans owe our survival as a species to the vigor and promiscuity with which we nurture each other’s children. Whereas other great ape mothers are jealously possessive of their newborns, human mothers traditionally have sought and received sustained help with their babies right from the start. Hrdy and other anthropologists have studied members of modern foraging societies throughout Africa, Asia, and South America for clues as to how our ancestors most likely lived—and hence how our species evolved. They have found that at the low end of the human allomothering spectrum are the !Kung of southern Africa, among whom infants are in the care of someone other than their mother roughly one-quarter of the time. In other societies, like the Hadza of Tanzania, babies are cared for by someone other than their mother a whopping 85 percent of the time during their first days of life. Although mothers take over more as time goes on, Hadza children still spend roughly one-third of their infancy being carried by allomothers. Central African Aka and Efe women share their babies, cooperatively holding, comforting, washing, and even nursing them collectively. This is no anomaly; women provide children who are not their own with milk in nearly 90 percent of modern foraging societies around the world. Humans in modern cultures extensively allomother as well, although allomothering takes different forms across various cultures and subcultures. Most modern human infants don’t receive care from twenty different people every day like the average Aka baby, but they will nonetheless be allomothered by a great many adults and older children between infancy and adulthood. This includes all the feeding, cleaning, cuddling, protection, and entertainment that infants receive from their earliest days onward from fathers, siblings, grandparents, aunts and uncles, and cousins—as well as from genetically unrelated caregivers like doctors, nurses, neighbors, and babysitters. It also includes the care, instruction, and resources that older children receive from teachers, coaches, and other adults as they mature. And of course, children who are raised by adoptive parents, foster parents, or godparents are, biologically speaking, allomothered as well.
I sometimes get the sense that some people view it as a necessary but unnatural evil that human children receive so much care from individuals other than their mothers. The belief that children should spend the maximum amount of time possible specifically with their mothers for proper socioemotional development is common and not entirely without basis in the scientific literature. Developmental scientists tracing back to the psychiatrist John Bowlby have historically emphasized the importance of a secure attachment to a single primary caregiver (nearly always the mother) for a child’s social and emotional well-being. Given this, the assumption that a child will inevitably be at least a little worse off when cared for by anyone other than his or her mother is common. This assumption underlies many public debates about whether child care for working mothers should be affordable and accessible, and it probably influences the thinking of the 60 percent of Americans who continue to believe that children are better off if their mother stays home to care for them. Mothers who are not working get, if anything, even more flak if they leave their children in others’ care. I was on maternity leave after having my second daughter when I mentioned to an older relative that we’d hired a night nurse to help care for her. His somewhat skeptical response was to ask why I wasn’t doing it “the old-fashioned way.”
The idea that it is preferable for a human mother to be her child’s sole caregiver around the clock is rooted in modern postwar views of family life that have drifted far afield from our species’ evolutionary roots. Mothers getting ample help from experienced caregivers is the actual old-fashioned way. This way is vastly more natural and sustainable—and beneficial for children—than for one or even two frazzled and inexperienced parents to try to manage twenty-four hours of daily care for some of the most altricial newborns on earth. As the historian Stephanie Coontz has put it, “Children do best in societies where childrearing is considered too important to be left entirely to parents.” Allomothering not only relieves mothers of the massive burden of caring for and rearing needy, altricial, resource-hungry children alone but also fosters strong bonds between children and a wide array of supportive adults and provides children with opportunities to learn the many skills they will need as they mature—not least among them the opportunity to learn to love and trust widely rather than narrowly.
I should mention, by the way, that in focusing on mothering and allomothering, I by no means intend to give human fathers short shrift—far from it. Most mammalian fathers play little or no direct role in raising their children, but human fathers are among the more devoted and indispensable “allomothers” in the animal world. Between fathers’ direct care for their children and the indirect care they provide for mothers while mothers nurse and tend to children, it is clear that humans’ survival as a species depends, without question, on fathers’ committed care.
Solo care of human infants is essentially impossible. It’s so difficult that mothers in some foraging cultures (as well as mothers of other allomothering species) will abandon their newborns if they perceive that they will not receive sufficient allomothering support. The prevalence of postpartum depression has much less to do with postnatal hormones (a common myth) than with how legitimately depressing it is to care for a baby without enough help. Inadequate social support is a top risk factor for postpartum depression; it’s a bigger risk factor than poverty or having medical complications. My own experience bears this out. My husband and I tried to take care of our first daughter without enough help, and the effects on our mental health were grim. We were not living in poverty, and I suffered no complications, but we were grossly inexperienced at taking care of babies, had no close family living nearby, and were miserable. We were much smarter about paying for extra allomothering the second time around. The salary we paid Marie, our lovely night nurse, to help care for my second daughter may be the best investment I’ve ever made, and it opened my eyes to the importance of allomothering in raising human children.
None of this negates the importance of close bonds between mothers and their own children, of course. Allomothers need not prioritize the welfare of all infants equally, or be equally invested in their care. But humans’ deeply ingrained capacity for allomothering explains why most adults find it so difficult to disengage from the sight or sound of children in need, even when those children are total strangers. This is why charities often use images of distressed children in advertisements to such powerful effect. Depictions of suffering or injured children can galvanize action from strangers in a way that hordes of suffering adults rarely do. Think of the famous “napalm girl” photograph in which a naked and burned nine-year-old Phan Thi Kim Phúc and several other children are running, terrified and screaming, from a napalm attack on their village. This searing image was credited with turning the tide of public opinion against the war in Vietnam. Think also of the awful, heartrending image of Aylan Kurdi, the Syrian toddler whose sweet, round body was found facedown and lifeless amid the lapping waves on a Turkish beach. The sight of this one boy stirred compassion in millions of strangers and was widely credited with causing large shifts in opinion about Syrian refugees. It also massively increased charitable donations to help them. One charity, the Migrant Offshore Aid Station, which rescues refugees’ boats in the Mediterranean, saw a fifteen-fold increase in donations in the twenty-four hours after the photograph was published. The average number of donations to the Swedish Red Cross increased one-hundred-fold in the week following the photograph’s publication relative to the week immediately before.
That intensive allomothering is a fundamental part of human nature helps to explain why we humans have unusually low and relaxed thresholds for what will trip our “baby!” alarms. We put lions and dogs to shame. But our alarms work fundamentally the same way as theirs: the urge to care is tripped by the perception of what ethologists call “key stimuli” that characterize babies and young children. These features include a large head, large eyes, and a small chin and jaw, features that set children apart from adults across nearly every vertebrate species and that elicit care from adults very effectively. The reason that babies look this way is fairly straightforward: their brains and the tops of their skulls grow fast and early, whereas the lower halves of their faces grow slowly. These patterns are particularly pronounced among mammals, in whom milk supports the growth of babies’ ballooning craniums. The resulting babyish proportions, termed kindchenschema by the ethologist Konrad Lorenz, create an appearance of cuteness and cuddliness that draws in adults’ attention and causes them to respond with increased care.
For example, Jim Coan and his colleagues at the University of Virginia found that after adult men and women simply view pictures of baby animals, they make slower, more careful body movements. This is particularly true when the images depict unusually cute, babyish-looking animals. My own research has found that viewing images of human infants triggers the motivation to approach—to literally draw closer. With my student Jennifer Hammer, I presented forty-five subjects with images of unfamiliar adults and infants on a computer screen and asked them to alternately categorize each picture by moving a lever toward or away from themselves. The goal of the study was to measure the degree to which different images caused respondents to either want to approach what they saw—to draw closer—or to avoid it. Approach and avoidance are among the most primitive emotional responses that we have, and nearly any emotionally significant stimulus will trigger the desire to do one or the other (or occasionally both). Approach and avoidance can be measured in the laboratory using a lever, which in our case was a joystick hooked up to a computer. When we compared how quickly subjects pushed or pulled the lever in response to each face, we found that they pulled and pushed equally rapidly in response to adult faces, meaning that they felt no particular motivation to either avoid or approach. But when subjects (both men and women) gazed at babies’ faces, they were suddenly able to pull the lever toward themselves—indicating the motivation to approach—much more quickly than they could push it away. We also found that this pattern was linked to psychopathy. Subjects who were less psychopathic—in other words, more caring—had stronger approach responses to babies, suggesting an intrinsic link between this response and compassion.
The psychologist Leslie Zebrowitz and others have documented many other ways in which a cute, appealing, babyish appearance elicits caring responses from adults, whether the possessor of the babyish appearance is an actual baby or an adult with a babyish-looking face, or even a babyish-sounding voice. Adults with babyish features are viewed as more worthy of concern and care and as needing more of it. They receive lighter judicial sentences for minor crimes and are more likely to receive help from strangers—in one study, adults with babyish-looking faces were more likely to have lost résumés mailed back to them than were less babyish-looking people. These effects are not simply due to babyish-looking people being more attractive. Even controlling for variation in attractiveness, more babyish-looking or-sounding people evoke more concern and protectiveness from others who encounter them. Note that the urge to nurture and care for that which is babyish doesn’t stop with human adults but extends to babyish-looking cartoon characters, toys, and nonhuman animals, all of which attract attention, approach, and the urge to care.
This fact helps to explain all the abundant alloparental care that humans provide to nonhuman animals as well. What do you think domestic dogs and cats are but unusually infantile-looking wolves and wildcats that people provide with the full suite of maternal care, including retrieval, cleaning, feeding, and protection? Over half of all American households at any given time routinely engage in these behaviors, a fact that researchers have struggled to explain. There appears to be no genuinely rational reason that so many people across a wide variety of cultures keep and care for pets. Although some of them fulfill nominally useful functions like protection or catching vermin, household pets, particularly those in industrialized countries, are mostly an enormous net drain on time and resources, and they require constant care—not unlike human babies. It may be that we just can’t help ourselves. We are literally built to parent, and we have fairly lax thresholds regarding who or what we will apply our allomothering energy to. It has been speculated that one of the causes of rising pet ownership in industrialized nations is the decline in these nations’ fertility. It’s as though we seek out pets to satisfy allomaternal urges that aren’t met by human babies.
Humans are also responsive allomothers to wild baby animals of all stripes. Just think back to the crowd on the Florida beach where I shepherded the baby loggerheads to the sea. That day I was part of a glorious assemblage of men and women and boys and girls all allomothering our hearts out together. Those turtles were just a few of the thousands upon thousands of wild animal babies helped or rescued by Americans every year—like the squirrel pups adopted by Mimi the chihuahua, who would never have met their new allomother had they not first been retrieved, protected, and fed by two humans, Jeanette Young and her son-in-law. Even in an urban metropolis like Washington, DC, wildlife allomothers abound. I became one of them yet again shortly after setting out for a run through my neighborhood one recent September.
I had just turned on to a large, busy boulevard when I encountered a man and a woman on the sidewalk staring up into a tree. I paused long enough to spot tufts of gray, fluffy feathers and a slate-gray beak poking out between the curled fingers of the man’s hand. My curiosity got the better of me, and I stopped to ask what was going on. The man held out his fist to show me a baby blue jay cupped inside. It was quite still, but alert and unharmed; it gazed at me with glittering black eyes. The man told me he had been driving by and saw it hunched in the middle of the road, still wearing the downy feathers of a nestling and unable to fly. He pulled over, leapt from his car, and ran into the street, evading squealing rush hour traffic, to retrieve it and carry it to safety. But now he didn’t have the faintest idea what to do with it; he and the woman had been trying to locate a nest in the tree above. But they were both on their way to work and couldn’t stand there indefinitely. Isn’t it interesting that none of the three of us ever even considered simply leaving the little creature on the sidewalk to fend for itself? Unthinkable. What could I do but offer to take it? So I did. Gently, the man transferred the bird’s warm, cloud-soft body into my hand, then thanked me and drove away. I carried the jay home, its scaly dinosaur toes clutching my fingers and its heart pounding against them. It never made a sound, and its eyes never left my face. Once home, I put it inside a small box lined with a cloth, then drove it to a local shelter called City Wildlife, where it was raised with foster siblings brought in by other local allomothers before being released back into the urban forest. I think of it sometimes still. I hope it is doing well.
Surely you see where I am going with all of this.
Allomothering, particularly the forms that involve retrieval and protection, is all but indistinguishable from extraordinary altruism. What my heroic rescuer did for me in 1997, an anonymous neighbor here in Washington, DC, did (in an only slightly less extreme form) for a baby blue jay: he spotted a vulnerable creature in distress, felt an instantaneous urge to help, then risked his own safety by swerving to the roadside and running into traffic to rescue it. An even more comparable animal example can be seen in a recent video I encountered in which a Russian motorist risked his life to rescue a kitten stranded on a busy freeway. The urge to respond with protection and care to the young and the vulnerable, even at significant risk and cost to ourselves, is our birthright as mammals. The urge to respond this way even to youngsters who are not our own is our birthright as group-living social mammals who bear altricial young. But a birthright we share with only a very few other species (domesticated dogs among them) is the urge to respond to creatures of a wide range of ages and species—even if we have never seen them before, even if we might in other circumstances consider them predators or prey or pests—if they manage to trigger our highly sensitive and generous “baby!” alarms. It is this urge that lays the groundwork for the emergence of extraordinary altruism.
Allomothering is altruism, really. Species that allomother are perennially attuned to vulnerability, distress, and need, and they are primed to respond with nurturing and care when they spot it, even if the object of their care is unfamiliar or unrelated to them. A rat pup rescued by a strange adult female, an oryx calf protected from leopards by a lioness, or a blue jay saved from the street owes its life to the person who rescued it just as surely as I owe my life to my roadside rescuer, or Priscilla Tirado owes her life to Lenny Skutnik, or Zina Williams owes her life to Cory Booker, or thousands of patients in kidney failure have owed their lives to anonymous donors. And it may be exactly the same neural mechanisms that impel all of these behaviors.
Direct evidence that allomothering provides the basis for altruism came from a recent study of humans and other primates that was conducted by Carel van Schaik at the University of Zurich. He and his colleagues were seeking to identify evolutionary causes of altruism, or what they termed proactive prosociality, in which one individual spontaneously helps another without receiving any gains in return. They tested the tendency of twenty-four distinct species of primates, including lemurs, monkeys, apes, and humans, to engage in such behavior using a simple task in which one individual could work to provide other individuals with food while gaining nothing personally. The researchers then examined what other factors were associated with rates of helping across the various species. They considered factors like brain size (a proxy for intelligence), overall social tolerance, frequency of cooperative hunting, and whether the species tends to form strong pair bonds. They also examined rates of allomothering. They found that across all the primate species, including humans, the single best predictor of altruistic behavior was allomothering. Among macaques and chimpanzees, who allomother very little, altruism among adults was almost nonexistent. Among tamarins and humans, who have very little in common except extensive allomothering, altruism was frequent. Most of the other variables the researchers considered ceased to be related to altruism at all once allomothering was entered into their statistical model. The moral: species who care for one another’s babies are also far more likely to help each other out—to be altruistic—even when they don’t stand to gain anything from it. The researchers concluded that “the adoption of extensive allomaternal care by our hominin ancestors thus provides the most parsimonious explanation for the origin of human hyper-cooperation”—or, in other words, altruism.
This explanation makes a lot of sense. The evolution of mothering is widely agreed to be the origin of the capacity to care about the welfare of any being outside the self; the evolution of allomothering represents the origin of the capacity to spread that care far and wide rather than hoarding it greedily for one’s own progeny.
But can the relationship between allomothering and altruism also help us understand the rarer phenomenon of extraordinary altruism in humans? I believe it can.
In all the species I’ve described, individual variation in allomothering responsiveness is evident. Both the impulse to care for infants and the threshold for what triggers the urge to care vary considerably, not only across species but within each species as well. Wilsoncroft studied only five mother rats (who were all first-time mothers), but found clear variation in their mothering ability and motivation. Most lions kill any baby antelopes or baboons they encounter, but a few do not, and some, like Kamunyak, even go to great lengths to care for and protect them. And of course, humans also vary considerably in their allomothering interest and ability, as well as in their altruistic tendencies. The key question is this: what, if any, direct evidence is there that variation in allomaternal responsiveness underlies extreme acts of human altruism? I have already described it to you.
Recall that my own research, and that of others, finds that one of the best predictors of altruism is responsiveness to fearful facial expressions. Individuals at the very low end of the caring continuum—psychopaths—are notably insensitive to these expressions, probably as a result of dysfunction in the amygdala. They fail to recognize fearful expressions and fail to show appropriate emotional or behavioral reactions to them. Whereas fearful expressions seem to inhibit aggression and elicit empathic concern in most viewers, people who are psychopathic are relatively impervious to their effects. Altruists, on the other hand, are unusually sensitive to these expressions. They recognize them better and show enhanced emotional reactivity to them.
The reason this is so incredibly interesting is that, of all the expressions that a human being can make, the one that reconfigures the face to most resemble that of a baby is fear.
Fearful eyes are wide and large, just like a baby’s eyes, the visible portion of which has already attained adult size by three months of age. Fearful brows are high and angled upward, while the mouth is rounded and low and the jaw is small and receding. Together, these features make fearful faces appear vulnerable, submissive, appeasing, and infantile. Without a doubt, if you were trying to maximize an adult human face’s resemblance to a baby, you’d make it look fearful. That fearful expressions do look babyish has been empirically demonstrated. Some time ago, I published a paper with my undergraduate adviser Robert Kleck and my colleague Reginald Adams Jr. that showed that adopting a fearful expression causes a face to appear more babyish in every sense. Viewers described an array of fearful faces we showed them as infantile and dependent and as possessing all the “key stimuli” of actual babies’ faces, including large eyes, high brows, a small jaw, and a rounded appearance. They perceived fearful faces as babyish even when the faces were altered to retain their key appearance features but to be no longer recognizable as expressing fear, demonstrating that it is the physical appearance of the expression that makes the expresser appear babyish. This may explain why, as my student Jennifer Hammer and I recently found, people respond to fearful expressions with the same patterns of approach that they show infants’ faces, and why this urge to approach is also strongest among people who are the most compassionate (and the least psychopathic). The similarity in this study between how people respond to babies’ faces and how they respond to fearful expressions was striking.
That fearful expressions evolved to look the particular way they do is almost certain. These expressions, like happiness, anger, and other basic facial expressions, are displayed and recognized by members of many cultures around the world. A meta-analysis conducted by my graduate mentor Nalini Ambady and my colleague Hillary Anger Elfenbein made this clear: they found that in hundreds of studies conducted across dozens of cultural groups around the world, viewers can reliably interpret the meaning of fearful, angry, happy, and other expressions displayed by members of even distant cultures. Considering that some of our nearest primate relatives also express their fear, anger, and happiness using similar facial behaviors, it is very unlikely that these expressions are a purely culturally learned or socialized behavior. Instead, these expressions may serve vital functions that have caused them to be evolutionarily conserved across the generations.
The most powerful evolutionary explanations apply across multiple species—much as van Shaik and his colleagues were able to capture the relationship between allomothering and altruism by looking across multiple species of primates. So it is useful to consider the functions that fearful behaviors serve in other highly allomaternal species, like dogs and wolves. Dogs and wolves employ distinctive behaviors when fearful of attack. They crouch down or roll over, fold their legs and tail close to their bodies, and flatten their ears. They may whimper, lick the jaws of the aggressor, or even urinate. Together, as explained earlier, these behaviors usually trigger the observer’s Violence Inhibition Mechanism and save the cringing, crouching wolf from harm. But why do these behaviors inhibit violence? Or, to alter the emphasis slightly, why do these behaviors inhibit violence? Because, together, they cause the fearful wolf or dog to take on the appearance and other key traits of the one creature that social mammals who bear altricial young are very unlikely to attack: a baby. Key stimuli that set wolf pups apart from their parents are their small size, supine posture, and flattened ears. Pups also emit high-pitched cries, lick their parents’ jaws to request food, and sometimes pee on themselves. That is how these cues inhibit violence: in combination, they very effectively trip not just the aggressor’s “baby!” alarm but the even louder and clangier “oh my God, a baby is in trouble!” alarm, which rapidly suppresses the urge to attack and replaces it with care.
Human fearful expressions, it seems, may do precisely the same thing—as do other fearful cues, like crouching, frightened body postures, or shrill, high-pitched, fearful screams that echo the high-pitched cries of infants. By signaling both distress and infantile vulnerability, these expressions of fear are tailored with extraordinary precision to move those who encounter them to care. The moral philosopher Adam Smith seems to have intuited this when he wrote: “The plaintive voice of misery, when heard at a distance, will not allow us to be indifferent about the person from whom it comes. As soon as it strikes our ear, it interests us in his fortune, and, if continued, forces us almost involuntarily to fly to his assistance.” This description strikes me as uncannily similar to the way in which many altruists I’ve interviewed described their “almost involuntary” urge to help once they had been alerted to another’s misery. Smith’s only mistake, perhaps, was to infer that this is a response that is equally strong in everyone. That is almost certainly not the case. Rather, those individuals among us who are the most sensitive to these powerfully care-evoking cues are also—not remotely by coincidence—the most altruistic.
I am willing to bet that by now you can easily guess what part of the brain is considered the entry point into the parental care system. Yup, it’s the amygdala. Of course, the amygdala isn’t solely responsible for parental care any more than it is solely responsible for any other cognitive or behavioral outcome. But it is essential for getting parental care going.
Incoming sensory information of any motivational significance gets channeled inexorably toward the amygdala. When key baby stimuli are detected—like the large head and eyes and small lower face that create babies’ classically cute appearance—off to the amygdala this information goes. Leslie Zebrowitz and others have found that any face that carries these appearance cues reliably engages the amygdala, regardless of whether the cues belong to an actual baby or an adult who just looks like a baby. Babies’ cries are directed to the amygdala as well, and listening to them results in more activation there than do non-cry sounds of even very similar pitch and loudness. This fits in with the amygdala’s perennial alertness to signs of distress, like the bright pop of a wide, frightened eye or the ragged sound of a scream. The reason for the amygdala’s strong response to fearful cues, then, is likely twofold: fearful expressions and screams not only signal distress but carry infantile features that are important to the amygdala in their own right. The really interesting question is, what happens next? What happens when stimuli like fearful expressions or screams that signal both babyishness and distress arrive in the amygdala together? How does the signal processing in the amygdala lead to the urge to care?
We already know a little bit of the answer, of course. Seeing or hearing distress cues leads to an internal simulation of the distressed state. Externally, this registers as a slight uptick in heart rate, blood pressure, and sweat on the palms. This is an empathic response, in the sense that it helps the person experiencing these changes recognize and understand the other person’s distress. But another change that occurs when people see someone in distress is a little counterintuitive. Usually, events that cause mild fear reactions in people, like the sight of a snake or a gun, also trigger the urge to flee, to escape. But we know that, in the average person, this is precisely what people do not desire to do in response to others’ distress. The lever tasks I’ve run show that most people respond to others’ fear with approach, not escape.
This suggests that something approaching alchemy happens within the amygdala in response to others’ distress. Although it enables a viewer to conjure up a little internal simulation of another person’s frightened state—to empathize—this simulated fear results in the urge to approach, which is a behavior totally incompatible with escape but quite compatible with providing care and protection. And this response kicks in very rapidly—in roughly a second. The instant urge to approach someone who is frightened and vulnerable almost certainly results from that person’s similarity to an eminently approachable infant. In our lever studies, people responded the most quickly to both babies and fearful expressions when they categorized the two kinds of faces together instead of separately, suggesting an implicit association between them. What seems to be happening, then, is that after the sight of a fearful face results in a simulated fear response, somewhere within the brain—likely the amygdala, which regulates approach and avoidance behaviors—the rumbling train of behavior that follows switches tracks in the busy railyard of neurons within it. Because a fearful face carries key infantile stimuli, the train is redirected down an entirely separate track toward caring, protective behaviors. Empathy has been transformed into caring.
But who, or what, is doing this switching? In all likelihood, the switchman responsible for turning social lead into gold is not any single brain structure but rather a brain chemical that changes the activity of multiple brain structures—the amygdala included—simultaneously. This chemical is a neurotransmitter composed of nine linked amino acids produced in only one place on earth, which is the hypothalamus of all living mammals. This molecular alchemist is called oxytocin.
It is difficult to specify exactly when or how the first molecule of oxytocin came into being, but it almost certainly was in the brain of a cynodont, as all the cynodonts’ descendants—and only their descendants—produce it. Oxytocin—and its sister hormone vasopressin—probably originally cleaved from an older hormone called vasotocin that fish, reptiles (including sea turtles), amphibians, and birds still produce today, and which differs from oxytocin by only one amino acid.
But what a difference an amino acid makes!
Oxytocin is responsible for two key physiological functions that mammals require to reproduce. The first is stimulating contractions in the smooth muscles of the uterus to get babies out of it. If you or someone you know has had labor induced by a drug called Pitocin, you know how effectively oxytocin brings the walls of the uterus crashing inward. Pitocin is just a Frankenstein form of oxytocin created in a laboratory. One of my two labors was induced with Pitocin, and after the nurse inserted the IV and sent it flowing through my veins, I went from no contractions to wall-clawing labor within two hours. I imagine I would have found it fascinating from a scientific perspective had I not been preoccupied by the sensation that I might be about to explode.
The second key mammalian function that oxytocin performs is enabling nursing. It’s not involved in milk production itself, but in making sure the milk can be drunk. Milk that is inside a breast will more or less stay there until a baby latches onto the nipple and starts to suck. That strange drawing sensation is transmitted from the nipple up to the hypothalamus, where it prompts a few small clusters of cells to start churning out oxytocin and relaying it to the nearby pituitary gland. From there, the pituitary releases the oxytocin into the bloodstream, where it filters down to myoepithelial cells in the breast, alerting them to release the milk they contain into the nipple. And voilà, out it flows. This neat little process is called the milk ejection reflex, and it has been keeping baby mammals alive for many millions of years.
Of course, again, the physiological ability to lactate is not all that neat in isolation. Producing milk is only useful if it is accompanied by all the behavioral, cognitive, and emotional changes that allow the young to access it and benefit from it. These changes include the desire to spend lots of time in close proximity to the offspring, a lack of fear of them, and various nurturing and protective behaviors that keep them fed, clean, and safe. This seems like an enormous number of new adaptations for mammalian mothers to have acquired in a short period of time, and indeed it was. But incredibly, they are all supported by oxytocin, the same chemical that gets the babies born and the milk flowing. If milk and maternal care are the defining characteristics of mammals—and they are—then you could say that we mammals owe everything we are to this one little clump of nine amino acids. It gives me goose bumps when I really stop to contemplate it.
Oxytocin’s importance was initially discovered in studies of rats. You might remember that Wilsoncroft’s wonderfully maternal subjects were all first-time mothers. That is an important detail, as rats who have never had babies before act quite differently toward pups. They are worse than ewes, believe it or not. Virgin female rats find the smell and cries of rat pups highly upsetting and generally go out of their way to avoid them. If forced to remain in close proximity to pups, they sometimes attack or even cannibalize them. “Ew,” you can almost hear them saying to themselves. “I really, really, really don’t like rat pups.”
But one thing can turn these callous, cannibalizing monsters into attentive mothers willing to spend hours tirelessly rescuing rat pups from a dish, and it can do so almost instantaneously. That thing is oxytocin.
In the days and hours before a rat gives birth for the first time, neurons that produce oxytocin begin to multiply in her hypothalamus. Receptors for the oxytocin molecule also proliferate throughout her brain, sprouting up where none had been before in regions like the olfactory nucleus, which moderates responses to smells; the hypothalamus; the stria terminalis, a ribbon of fibers connecting the hypothalamus and amygdala; and the amygdala itself. These changes seem to set the stage for caring behavior to flourish.
Up through the 1970s, many efforts had been made to identify the neurotransmitters responsible for caring maternal behavior, and they had largely failed. Estrogen, progesterone, prolactin—all are hormones involved in female reproduction and seemed likely candidates for motherliness, but when they were injected into the brains of virgin rats, their aversion to pups remained unchanged. But when Cort Pedersen and his colleagues injected oxytocin into female rats’ brains, their responses to pups were transformed in minutes.
At the start of one experiment, the researchers divided up more than 200 virgin female rats into groups. One group was randomly selected to have an inert saline solution injected into the fluid-filled cavities inside their brains. Then they were placed in the center of a cage that also held three squirming pups arrayed in a triangle, three inches apart from each other. Most of the rats did what virgin female rats usually do, which was to ignore the pups. Fewer than one in five showed signs of maternal behavior—although those that did showed the full suite within the hour, including picking up the pups and putting them into a little pile together, licking them clean, building a nest for them out of any materials handy, crouching over them, and retrieving back to the makeshift nest any pups who wriggled free.
For comparison, the researchers tried injecting a number of other chemicals into other groups of rats, including vasotocin, vasopressin, and estrogen. Still no change. But when they injected oxytocin into the rats’ brains, the change was fast and profound. Now nearly three-quarters of the rats started to care assiduously for the strange pups—a 400 percent change over baseline. This showed that oxytocin can singlehandedly promote not just mothering but allomothering, as all the pups were strangers to the female. But again, note that even the powerful oxytocin didn’t affect all the rats equally. About 20 percent of the rats given oxytocin still failed to show any maternal behavior, and another 7 percent killed at least one pup—as did about 7 percent of the rats in every group.
A tidal wave of research has since fleshed out the critical roles that oxytocin plays in mothering and allomothering. Oxytocin has been shown to induce maternal (and allomaternal) care in a wide variety of species, including rats, mice, rhesus monkeys, meerkats, goats, and sheep—and thus we can reasonably assume that it similarly affects all the other mammals that haven’t yet been tested. Female rats whose brains produce more oxytocin tend to be better mothers. Less devoted mothers get better, though, if you give them more oxytocin. In contrast, they do much worse if you chemically block oxytocin receptors in their brain, which eliminates maternal care almost completely. Oxytocin can induce sheep to begin tenderly caring for unfamiliar lambs, something they would normally never do, thirty seconds after it is injected into their brains. This is not terribly useful information for a rancher, as most sheep ranches are not equipped to perform brain surgery on their livestock. But the knowledge of oxytocin’s importance has given ranchers another tool for inducing ewes to accept orphaned lambs, which is—and I know this is gross, but it works—to poke either their hand or a special balloon up into the ewe’s birth canal and massage her cervix. Fun, huh? But this stimulation triggers a wave of oxytocin production, much the same way a baby suckling does, and causes the motherly lightbulb in even a sheep’s fairly dim brain to flicker to life.
Oxytocin acts throughout the brain, and the specific locales in which it acts vary somewhat from species to species. But across species, the amygdala is a central locus of oxytocin’s effects. Oxytocin seems to act in the amygdala to reduce any aversion to the unfamiliar smell, sight, or sounds of an infant, preventing avoidance or aggression and opening the door—or switching the track, if you prefer—for care. Very recent research suggests that this is true in people just as it is in other mammals.
Studying oxytocin’s effect on people is more challenging than studying it in rats or sheep. Researchers inject oxytocin directly into these animals’ brains because oxytocin is a very big molecule—so big that it is thought to have trouble crossing the blood-brain barrier if it’s injected into the bloodstream or swallowed. Researchers usually avoid injecting hormones directly into living human brains, so efforts to investigate oxytocin’s effects on humans were stymied for a while. But a simple method discovered in the 1990s solved the problem: squirting oxytocin up the nose as a nasal spray. There it gets absorbed through the thin, porous skin lining the sinuses and into the brain itself.
When I began my postdoctoral work at NIMH, I was dying to set up an intranasal oxytocin study to test how oxytocin affects care-based responses in humans. I started getting the paperwork drawn up in 2004, before almost any human oxytocin research had been published. Unfortunately, the internecine bureaucracy of the NIH prevented me from actually running the study until 2006, by which point a wave of human oxytocin research had gotten under way, accompanied by huge swells of hype. Early studies found that oxytocin increased the amount of money people offered strangers in economic games, or the amount of time they spent focusing on their eyes. Oxytocin was quickly dubbed the “cuddle hormone” and the “love hormone.” News articles suggested that car dealerships should pipe it through their HVAC systems to increase sales (really). Psychiatrists speculated that it might cure autism (it doesn’t, sadly). The research that generated all this hyperbole ended up getting a fairly bad rap, understandably, and later studies have questioned whether some of these early findings were even true. In truth, oxytocin is not a panacea for making all social interactions more cuddly and lovey. Why should it be? Its essential purpose is supporting the care of vulnerable offspring. (In some species, it has also since been exapted to serve related functions, like pair bonding and social recognition.) Thus, sometimes it will promote sweetness and cuddling and other times it will promote wariness and aggression against intruders—these are all forms of maternal care supported by oxytocin.
In an effort to explore oxytocin’s care-related functions in humans, my research assistant Henry Yu and I spent two years asking people to squirt doses of oxytocin or a saline placebo up their noses in the Clinical Center of the NIH, just upstairs from where I was scanning the brains of psychopathic adolescents on the weekends. None of our oxytocin subjects tried to cuddle with us, but their behavior did change in ways that were consistent with increased parental care. In one study we found that a few squirts of oxytocin up each nostril increased subjects’ preference for infants’ faces, but decreased their preference for unfamiliar adults’ faces. This is just what you would expect if oxytocin influences not just parental care but alloparental care: increased preference for unfamiliar babies, whose faces carry key infantile stimuli, and wariness of unfamiliar adults who might do a baby harm. Think of sweet Mimi the chihuahua fiercely defending her pups from her owner once her squirrel babies got her oxytocin production humming, or the lioness who attacked other adult lions when they threatened her baby baboon.
We also found—in keeping with the fact that there is always variability—that the degree to which oxytocin increased subjects’ preference for babies’ faces depended on variation in a gene called OXTR, which affects the activity of oxytocin receptors in the brain. We found that people who carried the “A” version of a particular segment of the OXTR gene preferred babies’ faces no matter what we gave them, but that people who carried only the “G” version of this polymorphism preferred infants’ faces only after a hit of oxytocin. These findings remind me of Pedersen’s rat studies, which found that a few rats are maternal even without extra oxytocin, but that most need an oxytocin boost to become fully maternal. Finally, we found that oxytocin also increased subjects’ ability to recognize happy facial expressions, although only happy expressions that were fairly subtle and hard to recognize. This was an interesting result, but several other researchers have since reported results that I find much more interesting: oxytocin has even stronger effects on increasing accuracy for recognizing fear. One study found that oxytocin improved people’s ability to recognize fear (and only fear) by about 7 percent. Two more studies, conducted in Israel by Meytal Fischer-Shofty and Simone Shamay-Tsoory and their colleagues, found selective fear-recognition improvements of 13 and 20 percent, respectively.
That oxytocin strongly increases sensitivity to fearful faces—the same faces that psychopathic people fail to recognize, that highly altruistic people recognize with exquisite sensitivity, and that evoke approach and caring in people who see them—is, to me, remarkably clear evidence that oxytocin underlies the power of these vulnerable, infantile expressions to cause those who see them to care.
Now, for this to be true, oxytocin would need to be able to accomplish two things simultaneously. It would need to promote a strong empathic response to distress cues like fearful expressions to enable these expressions to be interpreted. But it also would need to inhibit the urge to avoid or escape in favor of approaching and caring for the fearful. As it happens, the findings reported in a 2016 study of rats strongly suggest that oxytocin can accomplish exactly this delicate balance. When researchers gave oxytocin to rats who were under threat, the rats showed all the usual physiological signs of fear—like elevated heart rates—that are part of an empathic fear response. But the rats didn’t show any of the avoidance or freezing that normally accompanies fear. This striking pairing—intact fear physiology but not fear behavior, which would enable an animal to feel scared but not act scared—was mediated by oxytocin’s effects in two separate groups of cells within the central nucleus of the amygdala. These findings help to explain what might otherwise be a puzzling phenomenon in rats: anxious rats make much better mothers, including being braver when defending their pups from harm. Their courage seems to result from their unusually vigorous amygdala response coupled with a rush of oxytocin in this structure when their pups are in danger. I felt a thrill of joy when I first encountered these findings, which represent an essential piece of the strange puzzle of parental care, and of altruism more broadly.
So, although no current technology can directly test this hypothesis in humans, here’s what I think is going on. Once a signal arrives in a human amygdala (the basolateral region, specifically) that somebody is frightened, two things happen. First, the basolateral nucleus responds vigorously, reflecting the importance of what it has detected. It then transmits the signal it receives to the central nucleus, which gins up an empathic response. For example, it tells the hypothalamus to increase physiological fear responding—pounding heart and sweaty palms and spiking blood pressure. Simultaneously, the vulnerable, infantile qualities of the expression are also being processed, setting off a wave of increased oxytocin production in the hypothalamus. When the oxytocin reaches the central nucleus of the amygdala, it triggers a response in the huge number of oxytocin-sensitive neurons that populate the lateral part of this nucleus. Neurons here suppress fear-relevant activity in other regions of the amygdala in response. These neurons may signal other cells in the central amygdala to inhibit what would otherwise have been an avoidant fear response in favor of a caring approach. These behaviors are regulated through the amygdala’s connections with various other structures in the parental care network, like the striatum and the periaqueductal gray, both of which are densely packed with oxytocin receptors.
The sum total of all this activity is a signal from the parental care system to the rest of the brain that there’s a sweet, juicy baby out there in the world who’s in trouble and needs your help, so don’t be a nervous Nellie—go out and get it!
I’d bet money on this system being fouled up in psychopaths (as well as other systems, I should add—psychopathy almost certainly represents not one single dysfunction but a constellation of them). Dysfunction throughout the amygdala prevents people who are psychopathic from registering others’ fear strongly in the first place, and even when they do, their oxytocin system probably isn’t set up to generate the urge to care, perhaps owing to observed abnormalities in their OXTR gene or other yet-to-be-discovered causes. In altruists, on the other hand, it’s likely that both of these systems are exquisitely sensitive. We know already that altruists’ amygdalas are highly sensitive to signs of others’ distress. Although so far we have data only about how they respond to fearful facial expressions, I would venture that a much wider array of other signs that a person is vulnerable and distressed—screams, frightened body language, crying, or other forms of helplessness or suffering conveyed verbally—might affect them similarly. Many of the altruistic kidney donors I’ve worked with say that they first felt the spontaneous urge to donate a kidney when they saw, heard, or read a news story about someone suffering from kidney disease. One read a harrowing Reddit post in which a stranger described what life in kidney failure is like and decided to donate that day. Several have been moved to donate after watching someone close to them suffer on dialysis. Harold says that one of the motivations for his donation was seeing an obituary for a child who had died of blood cancer because no bone marrow donor could be found. Lenny Skutnik was moved to dive into the icy Potomac after hearing the chilling scream of a drowning woman. I’ve always wondered whether my own rescuer caught a glimpse of my frightened face through the windshield. I’ll probably never know. Altruists uniformly seem to get a blast of empathic distress in response to these cues, but perhaps owing to highly responsive oxytocin-producing cells in their hypothalamus, or an unusually high density of oxytocin receptors in certain regions of the amygdala, they don’t respond by avoiding or escaping what they’ve seen. They dive right in with care.
If all this is true (and again, I’m betting it is), it would finally explain the fact that altruists are both sensitive to fear and brave in the face of others’ distress, and that their bravery is instinctive and intuitive. A complicated chain of events within some of the deepest recesses of their brains cause them to act on some of humankind’s most primitive, atavistic urges—urges that trace back to our earliest mammalian ancestors, whose babies needed food and care and would never have survived unless their very babyishness had prompted an unconditional desire to provide care and protection.
A high degree of variation in the specificity and sensitivity of these systems results in a small fraction of the population being so acutely sensitive to others’ distress and vulnerability that they will respond with the same conviction and certainty when saving the life of a stranger as an ordinary person might feel only when the life of their own child (or mother) is at stake.