Love at Goon Park: Harry Harlow and the Science of Affection

by Deborah Blum

  Six baby monkeys were placed with dogs; six with hobbyhorses. All the monkeys showed visible affection for their companions. They all grew up into competent animals. But they were very different in some very important ways. Mason watched these monkeys for four years, even after they had been moved into a bigger colony. Over and over, he found that the dog-raised monkeys were more engaged with other animals and more interested in the rest of the world.

  In one test, he put the monkeys into a box with peepholes in the side. It was not like a Butler box. The monkeys didn’t have to do anything to see out. There were peepholes, open and available for the looking. The animals just had to be interested in the possibilities outside. If they did put their eyes up to the peepholes, they’d see a picture. The dog-raised monkeys were simply fascinated by the chance to see something new. If Mason changed the picture, they would crowd the peephole and curiously study the differences. The other monkeys tended to hang back. They were a little nervous, a little uncertain about that outer world. If Mason gave the animals a puzzle to solve, the same pattern developed. It was the dog-befriended animals that exuberantly tackled the problem. The monkeys raised with the hobbyhorses tended to quit if the puzzle was too difficult. It worried them. They’d go back and comfort themselves by holding onto their hobbyhorse again.

  The difference, Mason argues, is that wonderful underrated opportunity to interact heartily with your companion: fight, play, share food, hog the bed space—even the shoving match matters, the constant back and forth as one player influences the other. “Stationary surrogates and hobbyhorses surely provide few opportunities for the developing individual to experience the fact that his behavior has effects on the environment,” Mason says. Maybe more important, if you successfully snag the last piece of cake, if your bedmate gives you that extra inch or so of space, you learn that you can, sometimes, exert control over your environment. “Inert mother substitutes make no demands, occasion no surprises,” as Mason says, and thus teach us nothing about managing our surroundings and, occasionally, ourselves. You need not pay attention to them, but it is in paying attention to others that we acquire social skills, learn the “fabric of social interaction.”

  Harry also came to realize that a cloth mother’s impenetrably passive nature made her a hopeless parent. She might be as cuddly as a fleece, but fluffy availability was never going to be enough. And it wasn’t just that cloth mom didn’t hug back. It was all the other things she didn’t do: She didn’t teach, direct, or steer the baby toward others. From cloth mom, the baby really learned more isolation, separation from others. “Growing up to be a monkey is an intricate process involving both ties of feeling toward other monkeys and the learning of monkey behavior patterns,” Harry said in a newspaper interview. Real living breathing mothers, he added, were not yet dispensable. Not for monkeys—or for their human cousins.

  Because we are social animals, it seems that one companion serves to connect us with another. And it’s worth remembering that rhesus macaques are definitively social animals. In the wild, or in cages that are large enough, they spend their lives in big, tumbling, interactive troops. They play games, they schmooze each other, they groom each other. The females help care for each other’s children; the males plot and form alliances, triumph together and sulk together, according to the results. A troop is strictly structured, organized by hierarchy, by social awareness, by street smarts about who’s a friend and whose back really, really needs to be scratched.

  By comparison, surrogate-reared animals are like alien monkeys from the planet nowhere. The cloth-mother-raised babies didn’t engage in any of that all-important schmoozing. They didn’t play with other monkeys; they didn’t swing into the usual spring mating season. They had no idea what to do with other monkeys—as friends, as enemies, as potential mates, as casual companions. No one had showed them the social ropes and they simply couldn’t find them without help. “The surrogate mother can meet the infant monkey’s need for an object of affection,” Harry said. “But it cannot teach the infant to groom itself or others, as the real mother does. Nor can it replace the mother and the other members of the monkey group, young and old, in providing the variety of social rules that the young monkey needs to make its way in the monkey world.”

  The original concept of his surrogate mother was still grounded in that sterilized notion of a healthy baby—clean, fed, warm, disease free, isolated from harm. Once again that had been proved incomplete at best and destructive at worst. “Harry originally thought he could be a better mother than the monkeys were,” Levine says. “And he was wrong.” Harry hadn’t stepped all the way back to the complete Watsonian model of maternal indifference. He knew that touch and affection mattered. What he hadn’t appreciated was that this matter of mothering was so complicated. But Harry was learning fast. He and his students continued in their attempts to dissect motherhood, not just what made a good parent, but what elements made the cloth mother such a bad one. What exactly were they seeing in this collapse of the surrogate-raised monkeys? Was it an insecure attachment? A failure of maternal responsiveness? A lack of social education?

  The Wisconsin researchers would reach an all-of-the-above kind of answer. Love, Harry would eventually argue, was not built of one relationship but many. Our love lives, all of them, forge links in a healthy chain of normal development: maternal love, infant love, paternal love, friendship, partnership—one connecting to the next and then the next. The early attachment is the first link of that chain, the start of our ability to connect with others.

  Now cloth mom wasn’t all awful. She was always there and she was never rejecting, much as any mother of a very small child needs to be. Becoming a parent means that patience becomes an ever-elastic attribute, stretching farther and farther as needed—and that will be much farther than a novice parent first appreciates. No matter how much a baby wakes up in the night or throws up all over her mother or screams in her father’s face, most of us know that our job is to answer the cry, clean up the mess, comfort away the scream without anger. We learn to walk away, to take our exhausted frustration out on, preferably, the nearest inanimate object. And babies all need that rock-solid acceptance as well, Harry insisted. Even monkeys know this. Rhesus mothers almost never punish a baby monkey in the first three months of his life, no matter how he tugs or pulls or makes his mother’s life uncomfortable. Cloth mother was perfect in meeting this particular challenge. She never slapped, never rejected.

  And yet, Harry also came to believe that one of the most important things that the mother must do, not at first, but soon enough, is to nudge the child away. There were two problems, as he saw it, with good old cloth mom. One was that she didn’t groom, or talk, or make faces, or directly and indirectly cue the baby for relationships with others. And the other was that she didn’t cut him loose to engage in those relationships. The mother needs, absolutely, to be there for baby, but she needs to show him how to be there for others. In a social species, Harry said, one relationship is never enough. We build a world of connections. We weave them—contacts and friendships and family and loves—into something that we lightly call “a support network,” and which is really the safety net that catches us as we balance our way along the high wire act of every day life and from which all of us occasionally fall.

  The one thing that made cloth mom sound so appealing at first—her never pushing baby away—turned out to be one of her eventual liabilities. If you turn that passive acceptance around, it meant she never encouraged her child to let go, never gave him—you may see this coming—that gentle push out into the rest of the world. In this light, Harry reconsidered Konrad Lorenz and his adoring flock of goslings. Lorenz’s famous imprinting work with the graylag geese had shown that the youngsters were dedicated to the mother who first hovered over them. If the goose wasn’t there (having been removed by the scientists) and Lorenz was, well then, the goslings followed Lorenz with compulsive dedication. There are still wonderful photos of Lorenz, upright and gray-
bearded, marching through a meadow, trailing goslings behind him like beads on a string. Lorenz called this dedicated behavior “imprinting,” suggesting that the mother is imprinted, like words set into stone, into the baby’s consciousness. John Bowlby was a friend of Lorenz’s and an admirer of his work; that hard-wired connection between mother and child was a touchstone piece of evidence in the building of attachment theory.

  And yet, at some point, a gosling must be able to stretch that connection. Eventually, waddling, flying, climbing, and walking away from your mother is also a survival instinct. Even a bird needs to grow up, find its own mate, build a nest, and raise its own family. Or, in Harry’s still poetic view of it:How does an infant break away

  To be a goose himself some day?

  Psychologists had long considered that all of us—goose or child—require some independence. Babies need total acceptance, but as they grow, too close and cuddly a nest is not necessarily a good thing. New York psychiatrist David Levy, whose work would eventually help support attachment theory, also spent time trying to determine when a child should stand on his own. His 1943 book, Maternal Overprotection, sometimes reads like an ominous Brothers Grimm tale of mothers who hedge their children in and surround them, like that wall of thorns around Sleeping Beauty’s mythical castle. Levy’s book contains twenty case studies, every one a lesson in the destructive effects of denying your child room to grow. Levy tracked one gentle sixteen-year-old boy whose mother always went to the movies with him and explained all the action so that his mind wouldn’t be “poisoned” by the wrong ideas. Another mother told Levy candidly that she hoped to keep her son “her baby” until he was at least thirty-five.

  Some of those homebound children simply obeyed maternal orders, growing ever more withdrawn and deferential. Other children beat against the walls of their cells. One fourteen-year-old boy, whose mother wouldn’t let him play with other children, or play chess, or even read detective novels, took to deliberately tracking mud through the house, cutting holes in his clothes, and screaming at his parents and siblings. All twenty of the children in his study, Levy noted, had extraordinary difficulty in making friends. They were loners, and Levy believed their mothers had turned them into misfits.

  The question for Harry Harlow was whether the ever-available cloth mom exerted a similar warping influence. She wasn’t a perfect parallel to Levy’s suffocating parents. She never held the children back, trapped them with her. They could leave whenever they wanted. She just never encouraged them to take those first steps away.

  Harry did eventually see some parallels with Levy’s overprotecting mothers. Cloth mom, he said, by being so passively acceptant never nudges her charges toward other relationships and “thereby never encourages independence in her infant and [affectionate] relationships with other infants and children. The normal state of complete dependency is prolonged until it hinders normal development.” The question still remained, of course, of how exactly to promote healthy independence. Should mother or child let go first? Should the mother softly hint the child toward others? Or does the child need to push for herself, open those still-fuzzy gosling wings, tread air for a while? And when? Is there a right time to fly, an equally right time to be gathered back into the nest?

  Everything out of Harry’s lab, and elsewhere in primate research, says this is a delicate moment of negotiation, all timing and small steps. Robert Hinde, Bowlby’s good colleague at Oxford, saw it almost as a dance: When a child is very young, the mother works hard to pull the child close for her own safety. But as the child grows, is less vulnerable, the mother becomes a little less protective. She takes a step or so back. Then the baby’s first response is to move toward mother. Hinde reported that small monkeys would begin calling more, nestling tighter. If the mother kept pulling back—mother monkeys would now cuff an older child who pulled too hard on her fur—the youngster would learn to seek comfort elsewhere. She might scoot over to other young animals, play longer away from home. Eventually, she would have strengthened those friendly relationships as well, broadening her base of support.

  The bond between the specific mother and baby, though, provided the background music to this dance. If the mother was never very supportive, what you might call a John B. Watson–approved monkey who pushed the baby away from the beginning—then the little infant didn’t want to go elsewhere. The young monkey didn’t feel secure enough to risk the next relationship. Harry began to see a sequence in this, a surprisingly strict order. First, the baby needed to believe solidly in that relationship with his mother, to be securely attached. If he didn’t feel that kind of solid support, it was much harder for him to turn easily to other animals. There was an obvious common sense conclusion, Harry thought, to the construction of this chain of relationships. If the first one failed you, it was much harder to forge the next. So the little monkeys stayed close to home, trying to mend their link to mother, failing to build the links to others.

  Leonard Rosenblum, after he had graduated from Wisconsin, did some studies that beautifully illuminated this idea. He used another species of monkey—rhesus are not the only macaque species, after all, and cousins of all shapes surround them: crab-eating macaques, bonnet macaques, pigtail macaques, lion-tailed macaques, Barbary macaques, Japanese macaques. Macaques in shades of gray and brown and silver-gray and flaming gold, tree-climbers, water lovers, foragers, homebodies. Sweet-tempered macaques and evil-natured ones, good mothers and indifferent. Pigtail macaques tend to be exceptionally affectionate mothers. Bonnets are far less focused in their offspring. And what Rosenblum found was that pigtail macaque babies, cuddled and fussed over and protected, found it easier to move on to other relationships. But the bonnet babies were continually trying to repair to the home front. They were clingier. Instead of the independence you might have once expected from being pushed outward, they hovered near home longer. So here you had an apparent paradox: To create an independent child, you needed to allow the baby to be dependent.

  And then you had to know when to let that baby go. At one point, recognizing cloth mom’s failures, Harry wondered whether pairing baby monkeys together would answer their need for a real relationship. But babies, remember, are made to cling and hold on and gain comfort from a security figure. So what the scientists ended up with was not two secure monkeys but two monkeys who wouldn’t let go of each other. Harry explained it like this: “The clinging together for contact comfort overwhelms the babies. They don’t know when to stop ... they haven’t even enough sense without mother to know when to start playing.”

  The infants wouldn’t reject each other at any stage of development. They were almost worse than cloth mom. At least she didn’t keep a stranglehold on her baby. “Harry discovered that if you rear two infants together, it’s almost as bad as total isolation,” says Jim Sackett, now at the University of Washington in Seattle. “They develop a tight clinging behavior and it looks cute, but if you separate them they go to pieces. It’s deadly for later socialization. Nobody in their right mind, who knew Harry’s work, would raise rhesus babies in pairs. Adults in pairs, yes. Infants, no.”

  So the list of mother mechanics gets longer here: warmth, motion, affection, and now enough sense to know when to hold tight and when to nudge a child away. Psychologist Irwin Bernstein, at the University of Georgia, calls such behaviors—a sense of timing in relationships, an awareness of when to hold tight and when to let go—“social intelligence”; he notes that back in the 1960s, when Harry first started talking about this, it was not a topic on psychology’s radar screen. “It was Harry’s genius to recognize that the baby monkeys were abnormal emotionally and that it was ‘social intelligence’ that they lacked,” Bernstein says. “This was not an area much investigated in the middle of the twentieth century.”

  The mother-child bond, on the other hand, now had everyone’s attention. Psychologists were riveted by the notion of mother love, contact comfort, and attachment theory. Now, as Steve Suomi dryly notes, there was a “relat
ive preoccupation with mother-infant relationships by those in the mainstream of child psychology and psychiatry.” Thus the explosion of work that so inundated the editors of The Competent Infant. It was going to take a little time for psychologists, in general, to see the mother-child relationship in the wide-field context of “social intelligence.”

  But, as Suomi points out, a monkey colony almost forces the broader panorama of relationships upon you. The whole community is there, in effect, and if you put mothers and infants, juveniles and other adults together you see a society a tumble with small monkeys knocking each other over, chasing, exploring, arguing, zipping back and forth from friends to family to friends. There was no way to watch monkeys and believe that one relationship alone was enough. You needed social skills on a far bigger scale to survive. For one thing, Harry said, “Monkey groups can spot a stranger a mile away, and if the stranger does not recognize its predicament and display the appropriate submissive behaviors, it is almost certain to be threatened, repeatedly attacked, repelled and perhaps even killed. Knowing one’s friends has enormous adaptive import, even among nonhuman primates.”

  In rhesus society, with its rigid top-to-bottom hierarchy, knowing your friends, their place, your place, adds up to a basic formula for survival. No one is born with that knowledge, and yet, from very early on, a child needs to know where he fits. If social intelligence has to be taught, the suggestion is that every child—human or monkey—requires a dedicated teacher. “Monkeys are not honeybees,” said Harry, and preprogrammed responses are not going to get them through life. “The best rhesus monkey genes in the world do not guarantee that the individual possessing them will be socially competent,” as Suomi puts it. It’s that absolute requirement for social intelligence that also helps explain why cloth mother—who knew nothing and could teach nothing—in the end turned out to be such a bad surrogate.