totaled hundreds of hours, and she had taken copious and continuous notes on everything from sleep practices, discipline, feeding, elimination, and other people in the family to clothing, home decor, and manners—basically everything she had observed. And now, from her desk in Baltimore, reading and rereading her handwritten notes, seeking both patterns and surprises, she had to find her way through it all.
She had started to sort the children and their mothers into three categories. Around 57 percent (sixteen pairs) were what she called “secure,” meaning the children appeared to know how to use their mother as a secure base in their explorations of the world. And explore they did.
Juko (thirty and thirty-two weeks) did much cheerful exploring, even out the front door of the house. Especially at thirty-two weeks, he kept returning to his mother and seeking the breast before taking off on another jaunt.
Around 25 percent (seven pairs) were what she called “insecurely attached,” meaning the babies found it difficult to relax into reliance on their mother, thus making it more difficult for them to experience the world with the zeal of the others.
Sulaimani (forty weeks) cried immediately when his mother put him down but stopped when she picked him up again. Again she tried to put him down; he screamed, and did not stop this time even when she took him up again. Later he permitted her to set him down on the floor but he played in a desultory way and protested whenever she moved away.
And just a handful (five pairs) were “not yet attached,” meaning they didn’t seem to have a “special” relationship with their mother at all, a phenomenon Ainsworth came to understand very differently later.
The twins manifested very little attachment behavior during the period of investigation. At twenty-three weeks they were described as lifting their heads up and vocalizing (as though they wanted to sit) in response to the approach of a person, but the response seemed quite non-differential…Neither baby cried when the mother left the room and both were described as behaving toward the mother no differently than they behaved toward others.
While we often think that babies not responding differentially to one special caregiver is a good thing, a sign of a certain degree of self-sufficiency, Mary began to believe that the opposite was true—that being able to discriminate one’s attachment figure from all others is in fact the infant’s first step in developing an attachment. While secure babies may well enjoy the affections of many people, attachment theory posits that children benefit from having one special relationship, even in cultures where children are cared for by many loving people. More and more, Mary saw that babies who didn’t seem particularly affected, at age one, by an attachment figure’s departure and return might actually be responding to various upsets and anxieties in their family’s life that negatively impacted the attachment process, and she was very sympathetic about this.
Struck by the differences in attachment behavior she was seeing in her notes, Mary wondered why. Was there more going on than what she could see on the surface? Aside from the expected emotional ups and downs of family life, why would some relationships create a more secure feeling for babies than others?
As she searched her notebooks for clues, she noticed that there were “a few gaps in information.” Even though she had made every effort to visit each family for roughly the same amount of time, she began to see that she had significantly less information about some babies than others. She realized that in some cases this was due to a “reluctance to cooperate,” which made perfect sense: if a mother wasn’t very forthcoming, then of course Mary wouldn’t have a lot of data about that baby. But in other cases, the families had been visited with regularity and she was left feeling that the mothers were totally cooperating, so much so that Mary found it “unbelievable that the data could be incomplete.”
She noticed that the babies about whom she had the least data were the ones she deemed insecure or “not yet attached.” This was big—an observation that is at the root of all the attachment research that would