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

  “He was ahead of the curve,” says Steve Suomi, “by at least thirty years. He was the first to look seriously at social behavior as it emerges in a developmental sense. He was interested in the layers of relationships, between mothers and infants, infants and other parts of the social world. His work preceded substantially the current argument over who is more important, parents or peers.” As an example, Suomi cites the well-publicized 1998 book, The Nurture Assumption, which argued that peers and peer pressure could outweigh parents’ influence. Like many researchers concerned about early childhood development, Suomi is wary of the book’s message. He thinks the author, Judith Rich Harris, took the modern emphasis on the non-nuclear family to a risky extreme. It’s true that his old mentor, Harry Harlow, believed that childhood friends mattered hugely, in part as a trial run for adult relationships. The playful days of early friendships do teach us some of the subtleties of building the social safety net. They can also buffer us against a dysfunctional family. Like Suomi, though, Harry would not have agreed that friendships make our first connections, mother-to-child, unimportant. His perspective was more complex than that.

  In essence, Harry said, one good relationship opens the way to the next. There may be phases of our lives when friendships or partnerships seem more powerful than our original families. But our ability to make those later relationships may well depend on what each child gets from his or her parents. We learn about love and connection starting in the first microseconds of our lives. For better or for worse, those lessons last us a lifetime.

  In their paper “Learning to Love,” Harry and Peggy Harlow wrote that one outstanding quality of the good primate mother’s behavior is “total or near total acceptance of her infant.” In this model, the infant can do no wrong. The mother anxiously supervises his beginning sallies beyond the protective reach of her arm. She will scoop the child back if necessary. The baby, as a result, gains confidence in her protection and “total, tender, loving care.” Harry agreed with that central tenet of attachment theory—if we don’t have a secure attachment as a child, we may struggle throughout our lives to feel secure in all relationships. John Bowlby himself used to express great exasperation with the Western notion that dependency was a bad thing. Bowlby sometimes worried that we push our children away so fast, we value the model of independence so ridiculously much, that we rarely pause to acknowledge that dependence can also be both good and natural. A child depends on her caretaker, a dependency starts with simple survival and grows into real affection; and throughout our lives, we always depend on the affection of others. A part of any good relationship—child or adult—is the secure base, Bowlby argued, and if we are lucky, we may happily spend our lives exploring the world but never doubting our warm welcome at home. The best adult life, he once said, consists of explorations secured by a loving relationship—no different, really, from the young child on a playground who is fascinated by the new possibilities but still looking over her shoulder to see that mother or father is standing by. “On this foundation, it seems, the rest of [a child’s] emotional life is built—without this foundation there is risk for future happiness and health,” Bowlby wrote.

  In rhesus monkey society, those first loving bonds are almost exclusively the responsibility of the female. And Mendoza was also ahead of her time in realizing that they resonate within. Scientists have learned ways to measure the internal biology of that relationship. There is singularly comforting body chemistry to being hugged by a parent who loves you. If a mother monkey scoops a baby close against her chest, heart rates drop—even more beautifully than when a baby looks away from frustration. Researchers have measured the same peaceful response in both boy and girl monkeys. Their stress hormones drop, their entire systems seem to relax and smooth over. An identical reaction can be seen in human children. A child tucked against his mother’s shoulder seems lulled into that easy chemistry of contentment.

  It was this lovely image—Madonna and child in perfect tranquility—that helped foster the “Velcro-mother” idea, as critics called it, the notion that mother and child needed skin-to-skin bonding time after a birth. Two pediatricians, Marshall Klaus and John Kennell, proposed the idea in the early 1970s. It may seem simplistic today. At the time, though, there was a compassionate logic to it. Klaus and Kennell were justifiably exasperated by hospital rules. The two doctors crusaded against regulations that forbade mothers to keep newborn babies close by and parents to stay with sick children. In the matter of newborns, the pair suggested that real harm could come from such policies. Perhaps there might even be a critical minute, as it were, in which mothers had to be there, had to hold and cuddle. If a mother missed that moment, they argued, she might not bond with her child. Klaus and Kennell pointed out that although only 7 to 8 percent of babies born at the time were premature, from 25 to 41 percent of battered infants were carefully isolated preemies.

  The pediatricians wondered whether the blame lay with hospital practice. Perhaps medical administrators were doing real harm. By enforcing separation, hospitals were causing mother and child to miss that all-important moment of connection. Klaus and Kennell wondered, for instance, whether mothers who were allowed to stroke and cuddle their babies loved them more. If so, they might be less likely to be abusive. It was a wonderfully appealing idea for many people. If they were right, of course, they could cure all kinds of dysfunctional families. The mother-child “moment” caught on almost instantly. Baby bonding videos flooded the market; in 1978, the American Medical Association made a formal statement in support of early cuddling. Of course, women who weren’t able to snuggle, following general anesthesia c-sections, for instance, were tumbled into needless guilt and worry.

  Needless because the mother-child relationship once again proved more complicated. As charming as it may seem to some, one near-magical bonding moment would make us even more vulnerable than we already are. A species such as ours, which must protect and nurture its young for years, would hardly be limited to developing love only in the first few minutes after birth. “Human infants are so helpless that it would be far more likely that bonding would be very flexible,” says Meredith Small, a professor of anthropology at Cornell University and author of Our Babies, Ourselves. “It doesn’t happen in half an hour. The connection to a child is a process.

  “The good thing that Klaus and Kennell did is that they helped open people’s eyes to the idea that the baby should be with the mother most of the time. So they were a little radical, sure. But sometimes you have to burn a bra to get people to look.” It’s not that connected minute that is so important. It’s the stable and reliable connection. The emphasis on bonding at least reminded people that the best relationships begin early, and their beginnings are intense. The rest of the story is that the relationship needs to stay intense, perhaps for years to come. Children need attention in a long-term, not a short-term, sense.

  When she lectures on childcare, Small still reinforces her position with Harry’s mother-child experiments. She shows slides of the baby monkeys cuddling against cloth mom. She hasn’t found a better visual example of the simple need to hold and be held. She wishes she didn’t still need that evidence, that we already had learned the lesson. But, Small says, even today people argue against that weight of commitment. “American culture is built on individual achievement. You’re told to be independent, self-reliant, get through life on your own. And that’s in direct conflict with how humans are designed, evolutionarily and biologically. If you look at other primates, little kids and even adults are meant to be together. We’re not like a bunch of wildebeests on the savanna. We’re supposed to be dependent on each other, children especially. That’s what all the evidence showed, that’s what Harlow’s work showed, that it’s natural for the little rhesus to be connected to its mother. Being disconnected is like being punished.”

  With his usual direct approach, Harry also acknowledged the difficulties of balancing the needs of the child against the needs of the professional woman.
“The working mother probably doesn’t help the structure of the nuclear family,” he said. “It is difficult for anyone to substitute for a mother.” Who do you find who loves the child the way you do? If we are honest, the answer is usually no one. In Western culture, most of us no longer have someone from an extended family system to watch over our fledglings. We turn instead to paid day care and we promise ourselves that we will make it up to the child, in the evening and on the weekends. Sounding surprisingly twenty-first century, Harry Harlow said many years ago that such balancing acts are not unreasonable. He thought that if a mother carefully chose good childcare, and, when she came home, put in undivided time, children could still grow up with a strong sense of love and security. He cited Peggy’s work on this point; the nuclear family studies, he told one newspaper reporter, made it clear that a mother is part—but not all—of a whole family support system. To raise a secure and emotionally healthy child, she needn’t physically hover every minute—if there are other dedicated caretakers, if children grow up in a close network of friends and family. But when the mother comes home, Harry emphasized, she really needs to be there. No stacks of paperwork, no constant phone calls to colleagues. On evenings and weekends, he said, she should be a mother first, a company employee second or third, or somewhere even lower on the list.

  In 1947, when Bowlby began making his case for motherhood, just 12 percent of mothers with young children (under the age of six) worked outside the home. In 1997, that number had risen to 64 percent. If we still believed in John Watson’s dictate of distance between parent and child, there would be nothing troubling about those statistics. They might be said to indicate a healthy trend. But thanks to all the scientists who changed that perception—Spitz and Robertson and Bowlby and Ainsworth and Harlow and countless others—we worry that the distance is an emotional void. We worry that extended childcare is a social experiment. We worry about the risks inherent in any experiment. There looms the fear that we are raising a generation of children so loosely attached to their parents that they will become socially adrift. Of course, you could argue that if John Watson didn’t engineer that disconnect, perhaps no one could.

  In the early 1990s, the National Institute of Child Health and Development (NICHD) began a study to investigate the possibility of a link between children who spend much time in day care and children who are insecurely attached to their mothers. Does day care influence children to grow up without faith in the affection and security of home? The institute made a major commitment to this question. The study is still ongoing. It involves twenty-four scientists and 1,360 children from families fanned through a range of neighborhoods and income levels. The first research phase compared three groups of children: toddlers who stayed home; those who spent ten hours or less in child care centers; and those who were thirty hours or more a week in day care.

  The early results, published when the children were three, were the kind that Harry always liked best. They served up answers that seemed to be built out of solid common sense. When the scientists looked at all the families, the parents, the children, the centers, added everything up, and watched the interactions, once again it was relationships that made a difference. What mattered was the connection between mother and child, the affection between a child and the other caretakers. If babies from a loving home with responsive parents went into day care, most seemed as secure going out as coming in. If a child had a strained relationship and indifferent home setting and then went into an equally indifferent day care setting, the emotional distance between mother and child often became wider—and colder.

  There were some interesting—not entirely surprising—complications to this picture. The study also showed that it was usually the already securely attached mothers who worked hard to find good day care. The mothers with insecurely attached children tended to be less fussy about where their toddlers spent time. There was a kind of feedback system. The mothers of the more insecurely attached children didn’t show the kind of protective behaviors associated with security. They didn’t watch as Harry’s good monkey mothers watched. They didn’t pay attention the way a good Bowlby mother would. Their children—put into Ainsworth’s strange situation test—didn’t seek comfort the way most of the others sought it. When their mothers returned at day’s end, as Ainsworth had consistently found, these children did not run to them in joy and relief. Neither did they coax affection from alternate adults. Ainsworth had also shown that mothers of insecurely attached children don’t like to be touched. Growing up in those households, it seemed, their children had never learned the comfort factor of a cuddle.

  You might ask whether relationships improved for the insecurely attached child who ended up in a terrific day care situation? Could regular days and warm, affectionate teachers tip that balance? Could the child learn to reach out more, repair his patchy support network? And the answer is another one that is less than perfect. Good day care didn’t fix the relationship with mother. It could, however, improve other connections. Children did become easier with others, a little friendlier. That in turned helped them to build other relationships. Children could learn social skills from their caretakers that they might not learn at home. In that sense, affection, even in an institutional center, can help build a stronger foundation. As Bill Mason found with his mobile surrogates, we social species can draw a lot of benefit out of a very small amount of interaction and support.

  Of course, big studies like this tend to project the big picture. Child by child, no two looked exactly alike. Not every dedicated parent had a securely attached child; not every harried and indifferent mother produced insecure attachment in her toddler.

  In other words, attachment is complicated; in other words, Bowlby was often right and sometimes wrong. You can separate mother and child, even every day, and not break the bond. You can keep the child at home and still end up with insecure attachment—as the NICHD study also found. “Clearly, then, attachment theory is by no means without flaws, holes and huge unanswered questions,” wrote psychologist and Bowlby biographer Robert Karen. “Various studies suggest that we cannot be as confident as we once were about the parenting styles that lead to [insecure] attachment or the degree to which inborn and cultural mores may also play a part.” It might be that the child is irritable, Karen pointed out. Or that the mother is too nervous, overreacts to the baby’s signals, ignores the look-away signal that says he needs downtime. Perhaps it’s the baby who pulls back and not the mother. Perhaps, too, attachment theory is hopelessly mother-centric, demands too much of one parent, doesn’t make room for the help of others. One question raised by the various mothers in Harry’s laboratory was: Which is better: a bad mother or no mother?

  One of the more interesting twists on “the right mother” or parent or guardian comes from Steve Suomi’s research. Suomi was one of Harry’s favorite graduate students. A stocky, fair-haired man, he brought to research the kind of single-minded intensity that Harry possessed himself and admired in others. Suomi considered a lifetime career at Wisconsin, after Harry had retired. He held a faculty position at the university for twelve years before the National Institutes of Health “made an offer I couldn’t refuse” in the early 1980s. He still works at NIH, where he studies both monkeys and humans in his job as director of the Laboratory of Comparative Ethology in Poolesville, Maryland.

  Suomi’s study raised a deceptively simple question: Is the biological mother always the best mother for the baby? To evaluate parenting styles, he compared biological monkey mothers to foster monkey mothers. He chose with care. The foster mothers were “supermoms,” picked for their nurturing style and—perhaps as a result of that style—their securely attached offspring. The comparison mothers were not rejecting or abusive. They were just a little less interested in their children, less devoted. The children in their care, though, were identical in nature. Both sets of mothers had to care for some high-maintenance babies, unusually nervous and jittery little monkeys.

  The little monkeys
simply did best with the most loving mother. Under tender loving care, the jumpy little monkeys grew visibly less stressed and, eventually, into nurturers themselves. “Those high reactive kids, reared by supermoms, now have kids of their own,” Suomi says. “And they are supermoms themselves. It appears to be a nongenetic means of transferring behavior to next generation.”

  The little monkeys with their less engaged mothers did not show such a dramatic temperament change. The NIH researchers intensified the study. They selectively bred for highly charged monkeys. Those monkeys had super-charged children. Again, those infants were either kept with their high-intensity parent or placed with a loving foster parent. In this study, you could watch the nervous parent create the nervous child. The high-stress parents weren’t unkind. They were just so jangled and distracted that it was difficult for them to really concentrate on the child. They were absorbed by jumping and responding and fretting. And so, it turned out, were their children. They were unnerved by the slightest change. They clung desperately to their mothers, apparently even afraid of inching away to explore. If the scientists provided new toys, altered the dinner menu, changed anything, the babies appeared instantly threatened. The cage would explode in a cacophony of alarm screeches—mother and child echoing each other in dismay. The difference in the foster families, thus, was almost deafening by contrast. There was plenty of conversation but not much screeching. There was no evidence that these babies had been nervous little infants when they were born. They grew up calmer, this time mirroring their foster mothers’ personalities. The infants acquired other benefits from growing away from the natural nest. The cross-fostered monkeys were often adventurous little animals. They explored with energy, made friends easily. They were unruffled by small changes. Foster mother and child alike remained unfazed if served oranges rather than apples for dinner.