UNDERSTANDING

ATTACHMENT
This page intentionally left blank
 UNDERSTANDING
ATTACHMENT
PARENTING, CHILD CARE, AND
EMOTIONAL DEVELOPMENT

JEAN MERCER

PRAEGER Westport, Connecticut

London
Library o f Congress Cataloging-in-Publication Data

Mercer, Jean.
Understanding attachment: parenting, child care, and emotional development /
Jean Mercer,
p. cm.
Includes bibliographical references and index.
ISBN 0-275-98217-3 (alk. paper)
1. Attachment behavior. I. Title.
BF575.A86M47 2006
155.4'18—dc22 2005019272

British Library Cataloguing in Publication Data is available.

Copyright © 2006 by Jean Mercer

All rights reserved. No portion of this book may be

reproduced, by any process or technique, without the
express written consent of the publisher.

Library of Congress Catalog Card Number: 2005019272

ISBN: 0-275-98217-3

First published in 2006

Praeger Publishers, 88 Post Road West, Westport, C T 06881

An imprint of Greenwood Publishing Group, Inc.
www.praeger.com

Printed in the United States of America

The paper used in this book complies with the

Permanent Paper Standard issued by the National
Information Standards Organization (Z39.48-1984).

10 98765432
 For Andrew—

one securely attached little gu

and his secure base people

This page intentionally left blank
Contents

Preface ix
1. What Is Attachment? The Study of Emotional Ties 1
2. Babies, History, War, and Politics: Early Work on the
Attachment Concept 13
3. The Growth of Attachment Theory: Connecting Ideas
through Research 33
4. Attachment, Age, and Change: Emotional Ties from
Birth to Parenthood 55
5. Families, Experiences, and Outcomes: What Difference
Does Attachment History Make? 79
6. Attachment, Mental Health, and Psychotherapy:
Emotional Ties and Emotional Disturbance 103
7. What Everybody Knows: Popular Views of Attachment 127
8. N e w Directions: Parents, Children, and
Attachment Concepts 149
Notes 173
Bibliography 181
Index 189
This page intentionally left blank
Preface

I
first thought about writing this book while I was sitting in a court-
room. I was observing the trial of two unlicensed therapists who had
managed to kill a child they were treating and who had done so partly
because they did not understand the facts about attachment (part of this
story is told in Chapter 7 of this book). Expert witness after expert witness
testified, and the word "attachment" was used many times—but no one
said what it meant. I looked at the jury to see if they looked confused. They
didn't seem to be having any problem. I waited for the judge to ask them
if they understood the term, for the prosecutor—for anybody at all—to
inquire whether the jury understood this important issue. No one did. No
member of the jury asked for a definition, either.
During a break, I sidled up to one of the prosecutors and asked her
whether anyone was going to tell the jury what attachment was. She smiled
pleasantly and murmured something, but her thoughts were obviously on
other issues, such as conviction. I minded my own business, then, because
I thought what the therapists had done was seriously wrong, and I wanted
them to be convicted, too. If the prosecutor could manage without having
the jury understand what attachment was, that was fine with me. (They
were convicted and are serving sixteen-year sentences, by the way.)
Nevertheless, I was puzzled. "Attachment" is referred to and discussed
daily in the twenty-first-century United States. As you will see in this book,
legal decisions about children are made on the basis of this concept. We
seem to be living in the "Age of Attachment" rather than the Age of
X Preface

Aquarius. Twelve members of that jury apparently thought they knew

what the word meant. Yet my experience as a professor and a lecturer is
that attachment is far more difficult to understand than most other psycho-
logical concepts. And my experience as a daughter, a mother, and a grand-
mother is that, once understood, the idea of attachment is of enormous
help in understanding family love and family relationships. I wanted to
make the attempt to pass on what is known about this fascinating topic.
I had seen at the trial and elsewhere the possible consequences of mis-
taken beliefs about attachment. I had also fielded some questions over the
years that showed how confused even child care providers are about attach-
ment ("Isn't there a disease about that? Something called separation anxi-
ety?"). But how to help people understand? The scientific and clinical
literatures about early emotional development and attachment are highly
complex. The number of published pages alone is a daunting challenge,
even though modern work in this area began only about seventy years ago.
No sensible layperson, and few professionals other than those working di-
rectly on the topic, would care to sit down and try to make his or her way
through all that has been said about attachment.
What I have tried to do in this book is to select the most important work
on attachment and to summarize it, stressing the concrete reality and ap-
plications of the ideas rather than research design, statistics, or even abstract
clinical concepts. In the course of this, I have included some material that
is important because it is probably wrong; readers need to know that not
every statement made about attachment is thoroughly validated. I have also
tried to show how our understanding of attachment has changed through
recent history—and how it continues to change. We have by no means
come to a complete understanding of this aspect of human life, and we
need to recognize this fact by leaving the end of this story open for specula-
tion, but not for wild fantasy.
I have tried to write about attachment in a way that will be accessible to
educated readers who have no background in child development or mental
health issues. Most of the language is nontechnical, and where I have had
to use technical terms, I have defined them. For those readers who are in-
terested, there is some material about research design and research conclu-
sions, but those who want to get on with reading about attachment will
probably not be much troubled if they skip those parts. There is a mini-
mum of biology included, and a small amount of discussion of animal be-
havior, because human beings are not alone in the animal kingdom in their
emotional connections to family. Most of the story of attachment comes
from the disciplines of psychology, psychiatry, and anthropology, however.
New parents sometimes inadvertently use the word "human" when
they mean "adult." Although they embarrass themselves by doing this,
 Preface XI

they are really quite right. Although an infant is a member of the human
species, he or she is not human in the full, mature, complete sense—the
sense that involves being a member of a group and caring about others. It
is the development of attachment that makes the baby a real human being
and prepares him or her to take his or her rightful place among other hu-
mans. How this happens to almost every individual is the real subject of
this book.
Our own family experiences cannot be the only source for understand-
ing emotional ties, but they can help us flesh out the sometimes dry bones
of systematic research. I have tried to include here some examples from my
own family life. In addition to personal experience, we can benefit from
the fact that attachment, separation, and loss have been the subjects of songs
and stories for generations. Literary treatments range from Biblical descrip-
tions of the prodigal son and the judgment of Solomon to modern forms—
the orphan status of Donald Duck's and Mickey Mouse's nephews, the
adoption of Superman, the lost father of Luke Skywalker. Orphanhood and
the lost or estranged parent have traditionally received the most emphasis;
as the blues song tells us: "Motherless children / Have a hard time / When
they mother is dead." More recently, the focus has been on family relation-
ships that are present but only partially satisfactory, as The Simpsons shows
us. The modern concern seems to be conveyed by Philip Larkin's poem,
"They you up, your mum and dad. / They do not mean to, but they
do." Artistic and literary descriptions of attachment can be vivid and
compelling, but they are not in themselves sources of information because
they are restatements of our own beliefs, and that is why we like them.
It may be valuable while reading to think from time to time of great writ-
ers' depictions of "disorder and early sorrow," or even of the mother of
James James Morrison Morrison Wetherby George Dupree, who went
downtown without consulting her three-year-old and, A. A. Milne tells us,
was never heard of since.
The hardest part about studying attachment may be abandoning old as-
sumptions and starting from scratch in our understanding of family emo-
tional connections. I hope this book will help readers accomplish this
important task and come to a new comprehension of a basic part of being
human, the source of some of our deepest pleasure as well as of our great-
est emotional agonies, without which we would not belong to the family
of man.
This page intentionally left blank
 chapter 1

What Is Attachment?
The Study of
Emotional Ties

"I don't want to breastfeed my baby, because he's going to have to go

to day care, and I don't want him to cry because he's attached to me,"
announces the mother of a newborn boy.
"How's Susie? Oh, she's doing fine, especially when you consider that we
never had a chance to bond with her," says the father of a lively toddler.

A nurse remarks, "A lot of my time at that job went into helping
mothers bond with their babies."
A mother tells her husband, "I like this day care center because they
keep changing Tommy's caregivers. I don't want him to get attached
to them."
A character in a detective novel comments, "He was ready to murder—
he never had any deep attachment after his mother."

Consoling a jilted friend, a woman explains, "You can't really blame

him for being commitment-shy after his mother deserted the family
like that."

A grandmother advises, "He just has that blanket because he's insecure.
You should take it away from him."
Another grandmother says, "We had never seen our grandson, but we
read him a story and we bonded right away."
A newspaper article describes a military interrogator as moving closer
to a prisoner "in order to bond with him."
2 Understanding Attachment

A book by a high-ranking intelligence advisor claims that terrorist

leaders can be detected on the basis of their childhood experiences.1
A judge orders a bonding assessment before making a child-custody
decision.

THE IDEA OF ATTACHMENT

What are all these people talking about? The specific situations are dif-
ferent, but there is a common theme: the emotional ties that exist between
human beings and guide their feelings and behavior. A term often used to
describe such ties is attachment. Attachment, in one form or another, is a fact
of every human life.
Every normal human being over a year of age likes some people better
than others. We choose to spend more time with the people we prefer, if
we can, and we seek them out when we are unhappy. When we are young
children, we feel safe with our special people; when we are adults, we try
to protect such people if they are young or vulnerable. If we are both
adults, we may choose them as sex partners, helpers, or companions. We
generally prefer people who are most familiar to us. If we are separated
from them for a long time, we are distressed, and if they die or desert us,
we grieve. We do not react in the same way to the loss of unfamiliar peo-
ple. This pattern of emotional relationships is present in one form or an-
other in all cultures and people. A person who feels and acts the same about
everyone is regarded as strange, and is strange.
A person's special preferences for others can be called liking or love. In
newspapers, television programs, and ordinary conversations, the words at-
tachment or bonding are often used instead of the meaningful but less ex-
citing everyday words. No matter which term is chosen, however, the
intention is to describe the fact that humans prefer to be with certain peo-
ple and are indifferent to—or even avoid or hate—others.

What Are We Talking about,

When We Talk about Attachment?
How can a word or two refer to all the different comments quoted on
the previous page? Can attachment be defined so broadly as to include all
those situations? In fact, attachment is not very easy to define. One highly
skilled and educated clinical psychologist, when asked to define this term,
replied, "I don't know, but I know it when I see it!"
Attachment has a multitude of meanings and implications, but its first
and most important aspect involves emotion. Human feelings about attach-
ment are among our most poignant and powerful, disturbing and gratifying
 What Is Attachment? 3

emotional experiences. Love, devotion, grief, mourning, jealousy, and

anxiety are all connected to attachment. From the two-year-old who
watches out the window for Daddy to come home to the lover who tor-
ments himself by imagining that his partner is with someone else, we hu-
man beings have intense experiences of attachment-related emotion.
Attachment is not just a matter of our personal emotional experiences,
though. In addition to attachment emotions, we have attachment thoughts.
These are our beliefs and ways of thinking about relationships with other
people. They can include not only our own individual love relationships
but also what we expect others to think and experience, even about rela-
tionships in which we play no role ourselves. Attachment emotions and
thoughts combine to form an internal working model of emotional and social
relationships, a set of feelings, memories, ideas, and expectations about
people's interpersonal attitudes and actions.
Adults may or may not want to talk about their attachment thoughts and
emotions, and young children may not have the ability to do so. Much of
what we know about attachment comes from the study of attachment be-
havior. The fascinated, engrossed gaze of the new father on his tiny baby,
the anxious pacing of the mother whose teenager is late coming home, the
wail of the toddler who sees his mother leaving the house—these are all
actions that tell us something about each person's attachment-related
thoughts and feelings.

Is Attachment the Same throughout Life?

The comments about attachment quoted at the beginning of this chapter
dealt with individuals of a variety of ages, from infancy to grandparent-
hood. All of those people were said to show attachment in some form, and
so they most probably do. But infants, parents, and grandparents do not
have exactly the same attachment behaviors, emotions, or thoughts. Like
many other aspects of human life, attachment develops; it changes as a re-
sult of age and experience.
The natural development of attachment is one of the reasons that there
seem to be so many events being described by a single word. During differ-
ent stages of life, attachment behaviors, emotions, and thoughts have differ-
ent age-appropriate characteristics.
Attachment is probably most easily observed in infants and toddlers from
about eight months to two years, and most readers will have seen at least a
little of these young children's characteristic behavior. Young children's
attachment seems to involve powerful feelings, the desire to be near the
preferred person, and the fear of separation or of the approach of an unfa-
miliar adult. (Unfamiliar children are not such a problem.) During this
4 Understanding Attachment

period of development, the child may be generally fearful and distressed by

loud noises, strange places, big dogs, and even loving but unfamiliar grand-
parents. Large groups of people may also seem threatening—as one two-
year-old complained tearfully, "Too many PEE-PUL!" Young children's
attachment emotions can be seen in behaviors like crying, showing fear, or
laughing and snuggling with a look of pleasure. Infants and toddlers are
likely to show stranger anxiety and separation anxiety as very obvious attach-
ment emotions.
Infants and toddlers show characteristic attachment behaviors, including
moving toward familiar people and away from strangers and making or
avoiding eye contact. Young children also use social signals like calling out
or raising their arms to ask a caregiver to pick them up. They may show
characteristic reunion behavior when a familiar person has been away, and
they may display mourning if the loss is permanent, with intense crying,
withdrawal, depression, and problems playing and sleeping.
As toddlers get better at crawling and then walking, they display a new
kind of attachment behavior, using a familiar person as a secure base. In se-
cure base behavior, the toddler explores a new environment or person by
briefly leaving the familiar person, then checking back by returning, getting
in the adult's lap, talking, or perhaps just making eye contact across a dis-
tance. In the absence of a familiar person who can serve as a secure base,
exploration does not occur so easily. Children of this age may use a transi-
tional object to help them explore or take comfort in a strange place. This
may be a familiar "blankie," a pacifier, or a stuffed toy; and, just as only a
familiar person brings comfort, only the right object serves the purpose.
The transitional object is not a substitute for the familiar person, and by this
age, the child probably will want both for maximum comfort.
During the later toddler and preschool periods, children respond to threat-
ened separation with aggressive actions, such as temper tantrums. Tantrums
are often a response to the child's perception of a loss of contact with an
adult attachment figure. (Our relatives, the baboons, throw tantrums when
their mothers refuse to carry them as they grow too heavy.) Anger continues
to be part of our response to loss throughout life, as does mourning behavior
that resembles the young child's. Although we tend to emphasize the posi-
tive feelings we have toward attachment figures and the longing we feel
when separated from them, attachment relationships also involve negative
emotions. This is very evident when an attachment figure has been lost and
the resulting mourning includes anger, frustration, resentment, and profound
sadness. Ongoing attachment relationships can also involve negative feelings
mingled with the positive ones. Ambivalence is evident in toddlers' responses
to their mothers, such as the rapprochement crisis2 in which the toddler
simultaneously demands help and pushes it away, the eye-rolling teenager's
 What Is Attachment? 5

sarcastic frustration with parental protectiveness, or the struggles of many

married couples to be "close enough but not too close." It may well be the
case that the importance of attachment relationships comes from the meld-
ing of strong but contradictory feelings and behaviors.
Most of our basic attachment behaviors are in place by the end of the late
toddler period. By that age, we are already inclined to try to stay close to
those we care for, to look at and talk to them when we are distressed, and
to cry or act sad when we are separated for a long time. Angry tantrums can
also be part of our attachment behavior. As we get older, more mature, and
more in control of our emotions, we learn to inhibit or to dissemble some
of our attachment-related impulses. The wife seeing her soldier husband off
on a deployment may want to cry or even rage at him, but she tries to keep
a brave face so he will think of her with pleasure in the weeks to come. The
child of divorced parents learns to keep the peace by acting as if he does not
mind when a parental visit is arbitrarily cancelled. The widow or widower
may not weep at the funeral service, but waits for a private time and place
before expressing attachment feelings.
With maturation, of course, we also learn about the world, and we un-
derstand, as a young child does not, the implications of some events for
attachment concerns. A diagnosis of serious illness triggers adult fears of loss
even before the sick person changes much in appearance or behavior. A
newspaper photograph or television news report can bring up vivid emo-
tions of anxiety in adults, but would be dismissed by a young child as just a
picture. A wife may recognize that her husband is talking a little too often
to the attractive neighbor and may begin to respond with anticipatory grief
and anger, although really nothing has happened yet.

What Is Bonding?
So far, we have talked about attachment, love, and liking, and about the
connections between these feelings and thoughts about other people. What
happened to bonding? Why is it not mentioned? This term is frequently
used in people's descriptions of preferences for friends and relatives. "The
mother and the baby bonded." "We went fishing and did a little male
bonding." "My sister says her cat is unfriendly, but it bonded with me."
Television programs, newspapers, even courtroom discussions speak of
bonding as if the term is clearly understood.
The word bonding was originally used to refer to a father or mother's sud-
den development of positive, protective feelings toward a baby born to them
or a very young adopted infant.3 When people spoke of bonding in that
sense, they usually meant to imply certain ideas: (1) that bonding involved
the feelings and behavior of adults toward babies, not of babies toward adults;
6 Understanding Attachment

(2) that bonding involved a dramatic, irreversible shift in the adult's emo-
tional life; and (3) that bonding was needed for the child's sake because it
enabled the adult to do the hard work involved in early parenthood.
As time went on, however, there were some questions asked about these
assumptions, and the term bonding became popularized and used more fre-
quently with less and specific meaning. Today, you will often hear people
talk of bonding and attachment as if they were a single process. Nursing pro-
fessionals sometimes use bonding as a name for the things they do to foster a
good relationship between a new baby and mother—things like encouraging
skin-to-skin contact or showing the mother things the baby can do.
Whatever meaning bonding once had seems to have been lost, as the
term has been used more and more loosely and carelessly. This book will
not use the word "bonding" without the accompaniment of a clear defini-
tion, and then only with respect to a very narrow range of human relation-
ships. It might be good for readers to think twice before using a word that
communicates so many different things to different people. Although at-
tachment has its own confusions, it still maintains a common enough mean-
ing to be useful—and we really have no substitute that carries all the
meanings of attachment emotions, attachment behaviors, and internal
working models that grow out of early relationships.

APPLICATIONS: ATTACHMENT AND REAL LIFE

Don't people just naturally understand attachment? Don't we all have
intuitions that help us manage our relationships with other people? Why do
we need to think about these natural parts of life?
Certainly most of us have learned a lot about emotional connections
through observation, but modern life often places demands on us that are
more than we can deal with on the basis of our casual experience. We
would like to be sure that we are doing the right things when we make
decisions about difficult matters like child care or adoption. Although hu-
man beings are resilient, we would like to feel confident that our decisions
will have the best possible outcomes.
There are a great many practical situations in which people need to apply
an understanding of attachment. Whether that understanding is correct or
not can have a real impact on the success and happiness of human lives,
especially those of children and families.

Daily Problems of Childrearing

Especially in the infant and toddler period, problems related to sleep are
often connected with what appear to be attachment issues. Fearfulness and
 What Is Attachment? 7

tantrum behavior have similar connections. Such behaviors are usually

worsened if a caregiver is sick or depressed, or in cases where there has been
some change, like moving into a new house or the birth of a new baby.
Applying attachment concepts to these practical problems can be helpful
for parents or other caregivers.

Child Care
When children are cared for in groups outside their family homes, their
experiences can play an important role in their development. Regulations
governing day care centers should include applications of what we know
about attachment. For example, the numbers of children cared for by indi-
vidual day care staff members, staff turnover, and the ways that caregivers
interact with children are all related to attachment issues and can help or
hinder early emotional development.
Parents choosing day care arrangements need help in understanding the
role these experiences have in determining their children's development.
Motivated and educated parents are a critical force in bringing about regu-
lations that produce higher quality child care. A survey done by the orga-
nization Zero to Three 4 showed that many U.S. parents believe it is a good
thing for young children to have several different caregivers. Parents with
this misunderstanding may opt for child care arrangements that are poten-
tially harmful.

Foster Care
Foster care programs are an essential factor in the lives of thousands of
children whose parents are not able to care for them effectively. These
children may be placed in foster care because they have been neglected or
abused, or because the parents are unable to provide adequate housing or
other necessities. The parents may be young, poorly educated, or mentally
or physically handicapped, and drugs or alcohol may have played a part in
the family situation.
Although it might seem logical to expect that children in foster care
would welcome and appreciate a home without neglect or abuse, in fact,
these children suffer from being separated from their parents, and they need
sympathetic responses to their reactions to loss. The age at which foster
placement occurs is an important factor here. Foster parents need to be able
to work with children reacting to separation, and as we will see later, they
may need to be chosen for personality characteristics that enable them to
provide a good emotional environment for children.
Foster care programs need to take into account attachment issues, such
as the need for familiarity, consistency, and continuity of care. Frequent
8 Understanding Attachment

changes of foster care placement do not give necessary support to early

emotional development. Along the same lines, foster care placement needs
to involve permanency planning: the attempt to guarantee a secure long-
term placement for the child as soon as possible rather than an open-ended
arrangement without a sense of a secure home.

Kinship Care
For some children, a substitute for placement in the foster care system is
kinship care. When the parents are not successful in providing an appropri-
ate home, a child may be placed with a family member, not informally, but
with the approval and supervision of the state's child protective services
agency. Attachment is an issue in kinship care just as it is in other foster care
placements, unless the child has had a longstanding familiarity with the
family member. As we will see, the genetic relationship is no substitute for
the experience of a familiar relationship, and movement in and out of
kinship care needs to be planned as carefully as any other form of care.

Adoption
Of all practical situations, adoption is the one that seems to have the
greatest relevance to attachment. When adoption occurs, children are
asked to undergo and adjust to separation from familiar people, then to
develop a new attachment for the adoptive family. Adoptive parents do
not just want to care for and rear an adopted child; they want to create a
new family relationship and new emotional ties between themselves and
the child.
How well the new connections work, and what the adoptive parents
need to do, depends strongly on the age of the child at the time of adop-
tion. As we will see, applying the abstract concept of attachment to real-
world situations requires attention to age differences. Adoptive parents
need to understand how a child's emotional ties change with development
and how the family needs to handle children in different circumstances.
Even if they have raised other children successfully, adoptive parents may
need help and training in order to do well with adoptees.
A highly demanding situation for adoptive families involves foreign
adoptions. Children adopted from other countries can be difficult to com-
municate with because of different language backgrounds and what may
have been serious experiences of neglect and abuse. They may have had
little opportunity to form attachments in the past, and they may also have
physical and mental problems that interfere with emotional connections.
Adoptive families who are trying to cope with all these problems can benefit
 What Is Attachment? 9

from a clear understanding of attachment and the role it plays in children's

lives.

Handicapping Conditions
Whether they are adopted or not, children with certain handicapping
conditions may offer a challenge in terms of emotional development, as
well as mental or physical growth. As we will see later, young children
cement their emotional ties with their parents by communicating their in-
terest in familiar people. They do so with the sounds they make, through
eye contact, by following, and by expressions of pleasure or displeasure.
Children with handicaps may not be able to communicate in the emotional
ways their parents expect. For example, a visually impaired child may not
look at the parents' eyes or facial expression and respond to it as a normally
sighted child would, and the parents may interpret this as a lack of emo-
tional connection. Parents of children with handicaps may also need to
make a practical application of information about attachment.
Children with handicapping conditions or serious illnesses may spend
significant periods of time in hospitals, receiving treatment or recovering
from it. Information about attachment can help clarify their emotional
needs during these periods, as well as the ways in which medical staff and
parents can be most supportive.

Domestic Violence and Terrorism

When children are exposed to domestic violence, their emotional lives
and ties to others may be distorted as a result. Mothers who live in fear of
violence may be unable to create normal emotional connections with their
children. For parents, foster and adoptive parents, or child care providers,
an understanding of attachment processes may help in the care and guid-
ance of children who have been traumatized by violence. The same is true
for those who work with or adopt children who have been orphaned by
civil wars and acts of terrorism.

Divorce and Custody

The frequency of divorce and remarriage in modern life means that large
numbers of children are subjected to separations and the possibility of hav-
ing to make new emotional ties. Application of information about attach-
ment can make these situations somewhat smoother for children. It is
important for parents to be aware of the emotional impact of these experi-
ences on their children and to take this impact into account as they plan
10 Understanding Attachment

further life changes. It is also important for attorneys and judges to be aware
of the effects of custody decisions on children's emotional lives.

School Issues
Although most readers will think of child care as a definite factor in early
emotional development, many will assume that by school age, children are
no longer vulnerable to separation or affected by other attachment issues.
However, separation and loss can have an effect on children's schoolwork,
play, and friendships, and schools can apply information about attachment in
ways that might reduce the impact of divorce or changes in living situations.
Such applications could be especially helpful for children in foster care.

Psychotherapy
Forms of psychotherapy that stress attachment are used in treatment of
both children and adults. It is possible, although not certain, that some
emotional problems are a direct result of difficulties in early attachment. It
is evident that diagnosis and treatment of emotional disturbance must in-
corporate the appropriate application of well-established principles of at-
tachment, or they risk doing harm rather than good.

UNDERSTANDING ATTACHMENT: USING THIS BOOK

Attachment and emotional development are not yet completely under-
stood, but much is known and will be discussed in this book. This first
chapter has presented some ideas about attachment and has offered defini-
tions of some important terms used in the discussion of emotional develop-
ment. Chapter 2 will provide some historical background and identify
important events in the growth of the attachment concept as we know it
today. In Chapter 3, we will examine one of the critical steps toward effec-
tive research on internal working models of social relationships: the inven-
tion and implementation of ways to measure attachment behaviors,
emotions, and thoughts. Chapter 4 will describe the natural history of at-
tachment and discuss some of the attachment changes characteristic of dif-
ferent stages of human life. In Chapter 5, we will look at work on the
outcome of attachment experiences, examining the evidence that different
personality outcomes result from different attachment histories. Chapter 6
will look at the possibility that mental illness can result from disturbed at-
tachment experiences. It will discuss the possibility that attachment prob-
lems can be treated through psychotherapy and the use of the general
concept of attachment in the therapeutic setting. Chapter 7 will describe
some ways in which the attachment concept has been used and misused in
 What Is Attachment? 11

popular discussions and in practical legal applications. Finally, Chapter 8 will

address the changes in social life that may lead to differences in attachment
experiences and the n e w directions in theory and research that may lead to
a somewhat different view of attachment than was k n o w n in the past.

Outside the Scope

T h e r e are some basic matters connected with the study of attachment
that are simply b e y o n d the confines of a b o o k like this one. G o o d research
design and statistical analysis are critical to the understanding of any devel-
opmental process, but there is space here only for the most essential points
about research evidence. Readers w h o want to k n o w m o r e can pick from
an excellent array of books on the subject, such as Keith Stanovich's How
to Think Straight About Psychology.
Infant determinism is another related issue outside the scope of this b o o k .
This m u c h - d e b a t e d notion posits that experiences during infancy have ex-
traordinary influence over personality development and override most later
life events. R a t h e r than arguing about the details of this complex issue, this
b o o k will be based on the assumption that individuals are shaped by both
early and later experiences.
Readers may be surprised to find little discussion of brain development
in the chapters to c o m e . It has b e c o m e fashionable to link all aspects of
child development to changes in the brain's structure and functioning. B o t h
cognitive and emotional developments are often discussed with respect to
brain growth, and facts about the effects of the e n v i r o n m e n t o n the brain
are used to advocate for childrearing and educational practices. This b o o k
will not follow that popular pathway. In spite of the frequency with w h i c h
these connections are made in the mass media and elsewhere, the fact is w e
do not have m u c h real information about connections b e t w e e n the brain
and emotional development. T o assume that w e do is to make an u n w a r -
ranted assertion. In fact, most of the information w e have is from animal
studies that may or may not be relevant to h u m a n development.
In one discussion of this matter, child development and policy expert
Jack Shonkoff 5 notes:

Translations of science that go far beyond what is known, or that

distort existing data, represent significant threats to its credibility in
the arenas of both public education and social policy. One egregious
example is the conclusion that young children who are abused or
who witness violence to others sustain irreversible changes to their
developing brains that result in permanent emotional damage and
inevitable violent behavior ... later in life [Tjhere is absolutely no
evidence to support [this].
12 Understanding Attachment

In a comment particularly relevant to the study of attachment and the

role of separation and loss, Shonkoff also states, "There are absolutely no
data that even begin to address questions regarding differential brain effects
related to the timing of an adverse experience, nor is there any neuroscien-
tific evidence regarding the reversibility of such hypothetical insults." It
is an attractive idea that everything about attachment might be explained
by looking at brain development, but it is one that will be avoided in
this book.
 chapter 2

Babies, History, War,

and Politics: Early Work
on the Attachment
Concept

H
uman beings have observed emotional development as long as
history has been recorded, and have paid a great deal of attention
to mothers' love for their children. The story of the judgment of
Solomon, like other tales, stresses the love of a mother for her child—in
this case so strong that the mother would rather lose the child and know it
to be well than keep it and have it harmed. Traditional stories of this type
gave less consideration to a child's preference for a familiar caregiver, but
our forebears did have some concerns about children's feelings.

M O T H E R LOVE
The mother's powerful preference and concern for her child were
sometimes considered to be an instinctive response, a biological charac-
teristic that humans shared with animals. (The difficulty of getting a
ewe to accept an orphan lamb when her own has died was well known,
for example.) At the same time it was recognized that an adoptive
mother could develop a strong emotional connection to a baby, who
would return her love. It was hard to argue that maternal love was
based on some biological response, built in and already present at the
time a baby was born. It was often assumed that the experience of
breastfeeding caused maternal affection to develop, and, although our
culture has largely forgotten the practice, adoptive nursing is possible for
humans.
14 Understanding Attachment

BABY LOVE
T h e idea that breastfeeding played a role in emotional preference was as-
sociated with the practice of wet nursing. T h r o u g h the eighteenth and n i n e -
teenth centuries, and even the early 1900s, it was not u n c o m m o n for a
family to hire a wet nurse—a lactating w o m a n w h o was paid to breastfeed an
infant. Some wet nurses lived with the family, and some took the child for
long periods of time to their o w n homes, possibly at a considerable distance.
This arrangement might have been made because the biological m o t h e r was
sick or had died; it might have allowed the m o t h e r to carry on earning
m o n e y with skilled work; it could have been based on the wish to allow the
m o t h e r to start another pregnancy quickly; or it might have resulted simply
from the wish for a fashionable figure and free social life.1 Discussion of wet
nursing usually stressed the possible dangers of a nurse with unwholesome
milk, or the possibility that undesirable traits could be taken in with the milk.
T h e r e was little concern about effects on the mother-child relationship.

Babies and Their Nurses

T h a t the w e t nurse could b e c o m e very fond of the nursling was taken for
granted, and some w e t nurses remained in the household for many years
after the child was weaned. T h a t a child could c o m e to prefer a nurse to the
m o t h e r was an observed fact, but usually not a source of concern, even
w h e n the nurse and the child w e r e abruptly separated u p o n weaning.
Eighteenth-century paintings showed toddlers shrinking from visiting
mothers and turning toward the familiar nurses they had c o m e to prefer. 2
Historical evidence of the impact of separation from a w e t nurse w h o
had b e c o m e an attachment figure was occasionally recorded. An account
written in 1841 of a two-year-old boy b r o u g h t to his parents' h o m e by his
wet nurse described h i m as enjoying " t h e best of health" during the t w o
weeks the nurse stayed, but w h e n she left:

[He] grew pale, sad, and morose. He showed himself unresponsive to

his parents' love and refused the dishes that had been his favorites only
a few days before The little boy whose failing became increasingly
apparent, spent whole hours in sad immobility, eyes fastened to the
door through which he had seen pass the woman who served him as
mother The doctor decided that the only way to save the child was
to call the nurse back immediately. 3

This strategy was effective; the child was allowed to spend another year
with the nurse, and a gradual separation at that time left h i m in good health.
T h e obvious cause of the children's preference for their w e t nurses was
the experience of breastfeeding, the w e t nurse's primary responsibility. It
 Babies, History, War, and Politics 15

was generally ignored that wet nurses also bathed, dressed, played with,
comforted, and even slept with their charges, all possible sources of emo-
tional connection. This view has not changed greatly, and in fact the aver-
age person in the United States is convinced that breastfeeding is desirable
because it creates mutual love, although, as we will see, there is little evi-
dence to support this idea.

Baby Love and Religion

For many eighteenth and nineteenth-century parents, a child's attach-
ment to mother or nurse was interesting, but usually inessential; it was cute,
basically harmless, and important only when the grieving child seemed in
danger of dying after separation from the loved adult. But some parents
during this period had a particular concern that their children have special
emotions connected with their parents—emotions that were thought to
play a part in their religious lives. For different religious groups in the
United States and European countries, the relevant feelings might involve
fear, respect, awe, love, or pleasure. These differing emotions were all
desired because they were expected to help achieve one goal: cheerful
obedience to parents at an early age.4 No doubt these parents, like modern
mothers and fathers, found an obedient child more pleasing than an unruly
one, but the parent's convenience was not actually the major concern.
Children were to obey the authority of parents as training for the obedi-
ence to God that would eventually save their souls—and the absence of
which would damn them for all eternity. Parents had the fearful responsi-
bility of shaping the child's attitude toward the goal of salvation, and they
were warned against yielding to their own affection for the child, an emo-
tion that might interfere with the work of child training. It was the child's
emotional response to the parents that was of critical importance; the par-
ents' love might be better suppressed than expressed.

FORMAL THEORIES O F A T T A C H M E N T

So far, our discussion on the history of the attachment concept has dealt
with informal, unsophisticated ideas shared by average parents and their
medical advisers. By the end of the nineteenth century, however, some
writers began to develop formal concepts of human emotional develop-
ment that included attachment processes.

Sigmund Freud
Freud based his thinking about attachment on the belief that feeding cre-
ates the child's emotional preference. Though development of the concept
6 Understanding Attachment

eventually surpassed this early idea, the foundation of attachment for Freud
was what our grandmothers called "cupboard love."
To discuss what Freud thought on a given matter is not as simple as it
may appear. Freud's ideas developed and changed in the context of a long
and eventful professional life. His conceptualization of the events of the
first years of life did not change greatly, however, even when some of his
ideas about human motivation altered.
To understand Freud's view of what we now call attachment, we need
to examine his ideas about the nature of infant personality as stated in about
1916. The newborn was seen as being driven by a combined need for food
and desire for the pleasure of sucking. Initially, in this view, the young
child's interests are narcissistic and focused on the self. Even when at the
breast, the baby concentrates on its own bodily sensations and gratifica-
tions. Nursing mothers take advantage of this by cleaning the baby's ears or
cutting nails while there is no resistance because the child is absorbed in
sucking. Gradually, though, that outside object, the breast, comes to be of
interest and is gratifying in and of itself. The mother becomes a love object,
not just a useful contraption for presenting the breast. Staying near the
mother then means security and happiness, and the threat of separation
becomes a cause for anxiety.
In order to disconnect the pleasure of feeding from the experience of
play and comfort, we would have to look for cases where one adult fed
the baby and another played and comforted it. In Freud's day, as in our
own, this would have been an uncommon situation. A baby may have
more than one caregiver, but each one generally provides a full spectrum
of care when on duty. Freud's view of feeding as a necessary cause of the
child's emotional preference emerged from the confusion of social and
biological factors (just as had earlier been the case when people thought
about wet nurses). The related fact that caregivers and feeders were almost
exclusively female led to the assumption that the mother was the focus of
the child's early emotional life. In addition, Freud stressed biological
drives like hunger because of their obvious connection with medicine
and the natural sciences—fields with respectable foundations of evidence
that reached far beyond what existed at the time for social and behavioral
studies.

Ian Suttie
Freud's many followers, and some of his opponents, maintained that
the emotional preference of the child for the familiar caregiver arose from
the gratification of being fed. However, this belief plays little or no role
in the modern concept of attachment.
 Babies, History, War, and Politics 17

In fact, late in Freud's life at least one thinker was already abandoning the
assumption that childhood emotional preference was founded on feeding
experiences. Dr. Ian Suttie argued that the child's need for the mother's
love was a primary necessity in itself and did not have to be connected with
the satisfaction of hunger. 5 Suttie was far more inclined than Freud to ex-
amine social and cultural factors and to note that not all groups of people
have the same childhood experiences, but, like Freud, he was not greatly
concerned with the smaller details of the child's development of emotional
preferences for people. Suttie's early death prevented him from elaborating
on these ideas.

The Ethologists
While Freud and Suttie were considering early emotional development
in humans, a quite different group of thinkers was approaching some re-
lated problems by studying the behavior and development of animals. The
German researcher Konrad Lorenz was a leader among them.
Lorenz worked in a discipline that attempted to explain behavior primar-
ily in terms of hereditary factors, with little emphasis on the role of learn-
ing. Focusing primarily on animal behavior, ethologists had little interest in
the effect of cultural differences on the individual. Their mission was to
describe behavioral characteristics of animal species, but there eventually
were some attempts to apply ethological concepts to human beings.6
Ethological theory had at least as strong an influence on modern ideas
about attachment as Freudian theory did and, thus, needs to be examined.
Before the twentieth century began, there had been many studies of ani-
mals using a natural history approach, but ethologists brought to their tasks
a clear theoretical system that went beyond simple observation. Certain
assumptions guided their work:

1. Ethologists emphasized the role of instinct in determining behavior.

Their definition of this term was rather precise and quite different from
the loose, everyday use today, when many English speakers say "in-
stinctively" to mean "automatically" or "involuntarily." For example,
someone might say, "When the dog ran across the road, I instinctively
slammed on the brakes." Although, from the ethologists' viewpoint,
instinctive behaviors were indeed both automatic and involuntary,
they were also inborn and unlearned, as opposed to the learned behav-
ior that is involved in driving a car. The concept of instinct included a
number of other assumptions as well.
2. Animals were assumed to inherit particular patterns of complex behav-
ior. Ethology was not concerned with a brief, simple reflex response
8 Understanding Attachment

like an eye blink, but with complicated actions, such as a mother mouse
caring for her offspring by following them and returning them to the
nest if they crawled out. Behaviors like this have many steps and must
be flexible and elaborate, guiding the mouse to follow her pups in dif-
ferent directions or to sniff them out behind an obstacle.
3. Ethologists assumed that inherited behaviors were species-specific.
Members of the species who are capable carry out these species-specific
behaviors (for example, a male opossum does not carry young in a
pouch, of course). There might be instinctive behaviors found only
among males or only among females of the species, and similarly there
might be instinctive behaviors that occurred only at a certain point in
development, not in older or in younger individuals.
4. Ethologists assumed that instinctive behaviors had some real connec-
tion to survival of the species. The types of behavior most likely to
involve instinct would be eating, drinking, defense, courtship, mating,
and care of the young; behaviors like intelligent learning and problem-
solving were much less likely to be instinctive. Of course, the most
important aspects of instinct from the perspective of this book are those
related to attachment issues: possible instinctive reactions of the young
to their caregivers, reactions that could mesh with the adults' instinc-
tive care behaviors in ways that facilitate survival of the young and of
the species.
5. Ethologists assumed that instinctive behaviors were usually directed
toward something in the environment and contributed to survival.
Generally, the objects that receive the animal's instinctive behavior
also release the behavior by providing the kinds of stimuli that cause
the behavior to begin. If the stimuli are the wrong ones, a normal in-
stinctive behavior may not occur. A baby animal that is sick or mal-
formed may not release its mother's instinctive caregiving behavior,
and a mother who behaves inappropriately may not release the baby's
instinctive responses to her.
To say that a behavior is instinctive does not mean that it will
occur in its perfect form under all circumstances. Nor does it mean
that the outcome of instinctive behavior is always appropriate. A
suspended, stuffed head of a female will release the male mating
behavior in some bird species. The males approach, position them-
selves, and copulate with her nonexistent body just as they would
with an intact live female.
6. The idea that instinctive behaviors are connected to their objects opens
the door to an important ethological idea about learning. By defini-
tion, instinctive behaviors are unlearned, but an animal might need to
learn something about the objects toward which instinctive acts can be
 Babies, History, War, and Politics 9

directed. Hunting and eating prey may be instinctive, but a wolf that
encounters a porcupine learns rapidly that a tender rabbit is a better
target than a painfully prickly beast. The ability to learn about the
potential objects of instinctive behavior is especially important for adults'
care of the young and for the young animal's appropriate response.
A female sheep, for example, quickly learns the smell of her own
lamb and will care only for that one, butting away any wandering or-
phan that tries to nurse. This devotion to her own lamb helps ensure
its survival, which would be endangered had the mother not learned to
recognize her own. Similarly, lambs, calves, and the young of other
species quickly learn to follow their own mothers, not just any female
of their species (most of whom would not care for the little one).
These animals' instinctive behaviors have to be combined with indi-
vidual learning for the best survival results, so the parallel with human
attachment is obvious.

IMPRINTING
Ethologists were especially intrigued by a rapid type of learning they
called imprinting. This type of learning was most obvious in young birds,
such as ducklings, whose early and later life experiences alter the object of
their instinctive behavior.
Young ducklings have the innate, unlearned behavior, in the first few
days after hatching, of following a moving object. In their natural life in the
wild, that object is most likely to be the mother duck, who has made her
nest and laid her eggs in a relatively solitary place. In other settings, how-
ever, the ducklings will follow other moving objects that happen to be
present: other kinds of birds, cats, dogs, people, or even toy trains. The
instinctive following seems to be released by the movement the duckling
sees, not by the kind of object that is moving.
Within a few days after this hatching and following, however, the young
duckling stops following just any moving object. It now restricts its follow-
ing to the kind of object that it initially saw and followed soon after it
hatched, whether this object is its own mother, a human caretaker, or even
something dangerous like a fox. The duckling is now considered to be
imprinted on the object it followed earlier and only that or similar objects
can release the instinctive following reaction.
Imprinting is, by definition, a form of learning because it involves a
change in behavior that depends on experience. It is an unusual type of
learning, though, because it only happens quickly and effectively during a
specific time period. (Ordinary learning happens in roughly the same way
at most times during an individual's life.) Imprinting has a different speed
20 Understanding Attachment

and pattern than other kinds of learning. To think of it another way, im-
printing is based not just on experience but also on an interaction between
the two basic mechanisms of development: experience and maturation.
Human attachment also involves both experience and maturation.

CRITICAL PERIODS
The time during which imprinting works best is called a critical or sensitive
period. Time is essential to the idea of imprinting, because an experience at
a particular time has an effect on development that it would not have had
earlier and will not have later. It is in this window of time that critical and
special processes appear to occur. Many critical periods happen early in
development, a point that we will need to keep in mind when we examine
modern thinking about attachment.
If you have been around farm animals, it probably does not surprise you
at all that ducks follow moving things, even things that are not at all duck-
like. What may be more surprising is that the duck's imprinting experiences
during an early critical period have a long-term effect on the duck, even
after it has passed the early following stage. As the imprinted duck becomes
mature and gets ready to mate, its preference for a partner is determined not
just by ordinary duck beauty and charm, but by the effects of the imprint-
ing experience. The duck remains extremely resistant to change and now
courts another duck, or a human being, or a toy train, the choice depend-
ing on the early following experience. Zookeepers must be very cautious
when hand-rearing birds that imprint, or they will find it impossible to
breed them as adults. Bird caretakers have been known to use a hand pup-
pet to attract the young bird's interest, as unrequited love for the zookeeper
may later cause the bird to reject mates of its own species.

THE FOUNDATIONS OF
A T T A C H M E N T T H E O R Y A N D W O R L D EVENTS
By the 1930s, much of Freudian theory and ethological theory were al-
ready in place. It would have been possible to blend these two views and
to produce the theoretical approach to attachment that emerged in the
1950s and later, but this blending did not occur.
World events in the 1930s and 1940s simultaneously postponed theo-
retical advances and piqued interest in early emotional life. The commit-
ment to "total war" during World War II delayed advances in theory.
However, due to circumstances resulting from the war, large groups of
young children were separated from familiar caregivers. This had occurred
for individuals in the past, but reactions could not readily be recorded.
 Babies, History, War, and Politics 21

Given the numbers of children involved during World War II, the possibil-
ity of observations on early emotional development increased, and research
interest grew.
Even before the war began, social and medical changes had already begun
to call attention to the effects of separation. Until well into the twentieth
century, young children who needed intensive medical treatment fell into
two categories: if poor, they were likely to die untreated; if affluent, they
received surgical or medical treatment in their own homes. By the 1930s,
advances in medical techniques and an increased awareness of infection
made it more likely that children would be hospitalized, but until the ad-
vent of antibiotics, fear of infection set severe limits on visits by parents.
During this period, large numbers of children experienced not only fear
and pain but also abrupt separation from their parents. One man who was
a child during this time recalls his tonsillectomy at age three. His mother
brought him to the hospital, handed him to a nurse, and then left, returning
for him as instructed ten days later. He did not speak again for a year after
this event.

Separation and Evacuation

As fathers joined the military during World War II, countless children in
the United States and European countries, as well as those in Asia, experi-
enced separation from their fathers. Their mothers' anxiety, depression, and
involvement in war work nearly amounted to a second separation. The death
of a father in combat made a child's loss permanent, and often exacerbated
the mother's preoccupation. Although wars throughout history have had
similar effects on children, the total involvement of combatant nations in
World War II made for an extraordinarily large number of affected children.
In the United States, though fears of invasion caused families to prepare
for evacuation of their homes, no serious evacuation program was ever
initiated. In Britain, however, constant German bombing of London and
of shipping and industrial areas caused immense damage to residences, and
consequently, many deaths of women and children. The British govern-
ment began programs to evacuate infants and children, with or without
their mothers, to safer rural areas, where they stayed for months or years,
often billeted in private homes. Evacuation of British children was abrupt
and hurried. One author wrote a few years later, "In a few days almost
three-quarters of a million children were separated from their parents
History has here made a cruel psychological experiment on a large scale."7
Observation of the evacuated children yielded confusing results about
the effects of separation, with frequent bedwetting as the most common
problem. In fact, it was quite difficult to understand the results of this cruel
22 Understanding Attachment

natural experiment to some extent because of the large numbers of children

and the chaotic conditions. Simply keeping track of trainloads of children
with their gas masks and identification papers had to take precedence over
data collection. And, of course, the primitive theories about emotional
development meant that most observers w o u l d have n o idea what to look
for in the children's moods or behavior.
Some famous observations have survived from this time period, h o w -
ever. Anna Freud, w h o w e n t to Britain with her father just before war
broke out, w o r k e d with D o r o t h y Burlingham o n observational material
they eventually published as War and Children* M u c h of their w o r k stressed
defense mechanisms—ways in w h i c h children protected themselves e m o -
tionally against the profound anxiety of separation from familiar people and
places. A w e l l - k n o w n report from Freud and Burlingham described the
case of Billie, a three-year-old separated from his mother. H e initially m a n -
aged his anxiety by repeating to himself that his m o t h e r w o u l d c o m e to get
him, that she w o u l d put on his coat and pixie hat, and so on. W h e n told by
a well-meaning or impatient staff m e m b e r not to keep saying these things,
Billie ended up nodding his head repeatedly in a tic-like way that made
sense only to those w h o had witnessed his earlier distress.
F r o m this period, w e also have Anna Freud's sympathetic descriptions of
the emotional concerns of y o u n g children in w a r t i m e — t h e i r terror of si-
rens and of people wearing gas masks, and especially of separation from the
familiar. O n e case states, for example:

Jim was separated from a very nice and affectionate mother at 17

months He formed two strong attachments to two young nurses
who successively took care of him. Although he was otherwise a
well-adjusted, active, and companionable child, his behavior became
impossible where these attachments were concerned. He was clinging,
over possessive, unwilling to be left for a minute and continually
demanded something without being able to define in any way what
it was he wanted. It was no unusual sight to see Jim lie on the floor
sobbing and despairing. These reactions ended when his favorite nurse
was absent even for short periods. He was then quiet and impersonal.9

Separation: The Kindertransport Arrangement

W h i l e the evacuation of British children was providing a glimpse into
the effects of separation o n early emotional development, European chil-
dren were also showing the effects of war. O n e well-defined group was
comprised of German Jewish children brought to Britain just before war
began in earnest. Immigration of these c h i l d r e n — w i t h o u t their parents—
was a result of the efforts of British refugee organizations after the homes of
 Babies, History, War, and Politics 23

German Jews were attacked in November 1938. Two thousand children

under the age of seventeen were brought to Britain by Kindertransport
trains, and then boats across the English Channel, before war broke out in
September 1939.
Observations by Anna Freud and Dorothy Burlingham showed how the
younger children coped with this abrupt, terrifying, often permanent sepa-
ration from familiar caregivers. One group of small children was described
as turning to each other for help and comfort, rather than accepting nurtur-
ing from strange adults.

John Bowlby
During and following World War II, a new and important formulation
of ideas about emotional development was offered by the theorist whose
name is invariably linked with this topic: John Bowlby.10 "Attachment the-
ory" and "Bowlby's attachment theory" are almost synonymous terms to-
day, although, as we will see in Chapter 8, another form of attachment
theory, not identical with Bowlby's work, is beginning to emerge.
Bowlby's theory of attachment involved ideas about human develop-
ment that were somewhat unusual for their time, although not completely
unique. He believed that the emotional attachments of infants and toddlers
to their caregivers were based on social interactions, not on physical grati-
fications, and that they were built into human beings as a result of adapta-
tion during early evolution. He saw a developmental timetable for
attachment emotions and behaviors that was determined by an inherited
human nature, not by learning. Most importantly—and, from the modern
viewpoint, most questionably—Bowlby considered early attachment expe-
riences to have a powerful effect on personality development. He was par-
ticularly interested in the impact of attachment history on mental health
and on criminal behavior. As a leader in the international discussion of
child development, Bowlby was in a position to stress this view, and a
world that had so recently seen thousands of children separated from their
parents was ready to listen carefully. It was within the context of Bowlby's
thoughts that most of the events about to be described occurred.

Planned Separation: The Kibbutz Movement

A movement for social change that began in the nineteenth century was
doubly connected to the study of emotional attachment in the twentieth.
The kibbutz movement began with concern about the negative impact of
close family ties on individual autonomy. The childrearing arrangements
that grew out of this concern did, and still do, provide new information
about social and emotional development.
24 Understanding Attachment

The kibbutz (a word simply meaning group) is a form of community

organization created (long before the founding of the modern state of
Israel) by some of the early European immigrants to Palestine. A form of
communal living with Utopian goals, the kibbutz is based on a philosophy
of social and economic equality in which tasks and rewards are shared.
Equality of the sexes was and is one of the goals of kibbutz life.11 Though,
in the past, most kibbutzim strongly emphasized agricultural work as a
source of income, the manual work involved could be difficult for men and
women to share equally.
Like other social groups, kibbutzim were not and are not identical, and
no single description can give an accurate picture of the two-hundred-odd
groups now in existence. However, these modern groups developed out of
a much smaller number, all of which shared similar social philosophies,
especially with respect to the functioning of families and of childrearing—
a complex question in a society trying to achieve gender equality.
A major part of the early kibbutz members' social values was a reaction
against the close family relationships of the shtetl, the small, beleaguered
Jewish community, historically typical of rural Europe. Shtetl life had been
narrow and sometimes dangerous, as its members were often forbidden to
own land and at times were threatened with attack from their non-Jewish
neighbors. (European Jews living in urban ghettoes battled these problems
too, but had more contacts with the wider world.) Intense family relation-
ships, combining love and hate with the ambivalence characteristic of im-
portant emotional ties, arose as families clung together in defense against a
hostile outer world. Poverty, danger, and some traditional Jewish beliefs
combined to reduce the opportunities open to women, and to make con-
nections with children a parent's critical emotional focus.
The early kibbutz philosophy rejected this whole system in favor of an
organization that would foster the development of the individual. Family
relations were seen as stifling rather than supportive. Intense emotional ties
between parents and children were considered a source of dependence and
weakness, rather than the autonomy valued by the immigrants.
In their desire to avoid the suffocating emotional connections they re-
called from shtetl family life, the founders of the kibbutz system did not
plan for marriages or children. They planned instead for an egalitarian,
communal life in which one's relationship to the whole group was more
important than intimate, interpersonal ties. Like most human beings, how-
ever, they did develop some interpersonal relationships and some babies
were born.
According to one original kibbutz member, Joseph Baratz,12 "[Our]
women didn't know how to look after babies." Baratz's group eventually
chose one woman to care for all the babies in a house. She was assigned
 Babies, H\stoiyr War, and Politics 25

to this purpose, while the other mothers continued to participate in

farm w o r k .
O t h e r kibbutzim adopted similar solutions, with one or m o r e separate
"children's houses," w h e r e the children eventually came to sleep as they
got older, as well as spending their days there. Mothers w h o w e r e breast-
feeding in the babies' first m o n t h s came in to nurse their infants, but they
had their o w n jobs to do and did not visit freely; the caregivers had their
child care w o r k and did not encourage mothers to participate. Except for
the period of breastfeeding, the children became the children of the k i b -
butz, rather than of their individual parents.
In the infants and toddlers' houses, the children's lives w e r e rather differ-
ent than what previous generations had experienced. Although parents felt
a special interest in their o w n children, social pressure w o r k e d against its
expression, as this w o u l d have implied that these individual parent-child
ties were m o r e important than those to the whole c o m m u n i t y . Parents'
energies were consumed by the demands of the adult group.
T h e children's house caregivers did their jobs, but they did not want or
try to create intimate relationships with the children. Caregivers changed
according to their shifts, and w h e n children reached given ages, they were
transferred to a n e w house with n e w caregivers. N i g h t t i m e care of the
younger children was a special issue because the regular caregivers were
off-duty at night. Some kibbutzim tried to have every adult take a turn
sleeping in the infants' house, but most adult members w e r e unwilling and
unskilled at dealing with a crying baby. Some slept t h r o u g h the cries and
others pretended to; as a result the chances that a child w o u l d be comforted
at night were low and unpredictable.
T h e o u t c o m e of the kibbutz pattern of child care was that babies had
only one consistent source of social experience: other babies. Adults w e r e
constantly coming and going, changing shifts, or altering their visits once
the babies were w e a n e d from the breast. O n l y the babies in nearby cribs
stayed the same, and almost all of t h e m w o u l d stay together for many years,
as the age group transferred together from one house to the next.
For the infants of the kibbutz, as for few other children, food and c o m -
fort were thus separated from predictable and familiar social and emotional
interactions. O t h e r babies could not comfort a distressed infant, but they
could provide the security of familiarity. Toddlers and older children could
actually offer comfort. As B r u n o Bettelheim described this:

At night, small children have only their mutual comforting to rely on,
since it may take quite a while for the single night worker to hear a
child who wakes up crying.... Even when the night watch is finally
summoned by the child's anxious cry, the person who comes when
26 Understanding Attachment

the child thus awakens from a nightmare, deeply shaken, is more or

less a stranger and can therefore give only small comfort. But as likely
as not, when the night watch gets there, he finds that some other child
has already soothed the anxious one.13

RECENT EVENTS A N D THEIR C O N T R I B U T I O N

TO THE UNDERSTANDING OF ATTACHMENT
Evacuation of children during World War II was not the last cruel psy-
chological experiment that has given us information about children's re-
sponses to separation and loss. More recently, events gave rise to thousands
of adoptions from foreign countries, providing an unhappy opportunity for
further research on early emotional development.

Reasons for Adoption from Abroad

Social changes in the United States, Canada, and Western Europe in the
1980s and 1990s made adoption of young infants more difficult. Concurrent
with this, couples were experiencing an increased level of fertility prob-
lems. Unmarried parenthood had become more socially acceptable, and
more girls wanted to care for their unplanned babies rather than relinquish
them for adoption. Large numbers of older children needed homes, but
couples seeking adoption most often wanted newborn infants they could
care for from the beginning—not an unrealistic wish, considering the many
problems an older child might bring to the family. Racial prejudices also
played a role, as many families of European backgrounds preferred, for a
variety of reasons, not to adopt children of color, and indeed many social
workers preferred to place minority children with the few available minor-
ity adoptive families.
Adoption of children born within an industrialized country could also be
fraught with difficulties relating to the legal rights of the birthparents, issues
about termination of parental rights, and the possibility that a birthparent
could change the decision about relinquishment or bargain for continued
contact with the child.
Foreign adoptions had considerable appeal in these circumstances. There
were young infants available; racial backgrounds were in many cases more
acceptable to parents of ethnic majority status; and there would be no con-
tact with the birthparents, thus, no question of changed decisions about re-
linquishment. In many cases, adoptive parents could feel gratified that they
were rescuing children from what were indeed vile, barbaric conditions.
In the 1980s, some questionable practices were associated with foreign
adoptions. Stories circulated about children who had been kidnapped or
whose poverty-stricken parents had been tricked or bribed into giving
 Babies, History, War, and Politics 27

them up. However, a foreign adoption industry rapidly developed to guide

Western parents in their adoptive process and to guarantee it was legal and
ethical.

Romanian Adoptees
Under the rule of the dictator Nicolae Ceausescu, the small, poor coun-
try of Romania developed a frighteningly large population of children who
existed in institutions. (Existed is the correct word; to say they were cared
for is stretching the meaning of that term.) From 1965 until 1989, Ceausescu
and his party enforced social regulations intended to double the population
and provide plenty of cheap labor. Women were not allowed access to
birth control or abortion until they had had four or five children.
Because of the country's extreme poverty, Romanian families were, in
many cases, unable to care for the children they bore. Orphanages were
established to "warehouse" children from poor families, and it became so-
cially acceptable to place children in these institutions. A large proportion
of the orphanage children came from Romany or Gypsy families, an ethnic
group considered inferior and undesirable. Children with birth defects and
problems, such as fetal alcohol syndrome, were especially likely candidates
for orphanages.
Conditions in the orphanages were almost unimaginable. Children re-
ceived little care, often being tied into cribs and given food in bottles until
a late age. There were few toys, little opportunity for free movement, and
almost no social stimulation or responsiveness from adults. Medical care
was minimal, and special needs children received no treatment for their
problems. Malnutrition was common, as were skull deformities caused by
lying in the same position for long periods. Many had hearing problems as
a result of untreated infections. There is little doubt that physical abuse, as
well as neglect, occurred at the hands of attendants and older children.
Considerable numbers of the children were HIV-positive. Because of the
institutional conditions and their pre-existing problems, many of the or-
phans showed severe developmental delays, failing to walk or talk until well
after the expected age.
In 1989, at the end of the Ceausescu regime, there were over 150,000
children in these Romanian institutions. Sadly, the changes in Romania's
social practices under that regime persisted after the fall of the dictator, and
abandonment of children by poor families has continued.
Beginning shortly after Ceausescu's fall, Western families began to adopt
children from Romanian orphanages, and they have continued to do so
until very recently. In 2004 Romania formulated a policy prohibiting many
foreign adoptions.
28 Understanding Attachment

Russian and Other Eastern European Orphans

Although the Romanian orphans accounted for a great number of ne-
glected children needing adoption or foster care and although the Romanian
situation gained the most attention due to its abrupt change in national
policy about foreign adoption, other Eastern European countries have been
major sources of adoptable children. In those countries, parents of babies
with physical handicaps are encouraged to institutionalize their children
and make them available for adoption, as are those whose poor living con-
ditions inevitably lead to developmental problems in the children. Though
adoption of these children has been encouraged, recent news stories about
the number of deaths of Russian adoptees in the United States have caused
concern in Eastern Europe, which may bring about new policies on foreign
adoption.

Adoption from China

In 1979, China instituted a "one-child" policy, limiting most families to
one completed pregnancy. The goal was to balance food production and
population size, thereby reducing poverty. This policy, of course, was
strongly at odds with the traditional Chinese desire for large families, as
well as the previous Communist policy of encouraging childbearing, and it
took some time for the rule to be enforced. There is still considerable resis-
tance to family limitation, but rewards for small families and penalties for
larger ones have been effective.
The one-child-per-family policy stresses delayed marriage and childbear-
ing, the trade-off being fewer and healthier births. When more than one
child is born to a couple, the parents may try to avoid consequences (such
as being restricted to a small apartment) by failing to register the birth,
though this ploy makes it difficult for the child to obtain an education.
There are exceptions to the one-child rule, and it seems to be the case that
the policy was intended for one generation only and may be altered in the
foreseeable future.
Because of the Chinese preference for a son to carry on the family line,
the one-child rule has a particular significance: the birth of a girl makes a
second attempt for a boy unlikely. As a result, abandonment of baby girls
began not long after the policy was initiated. This practice is in line with
the old Chinese tradition in which midwives had a box of ashes available
for the suffocation of a newborn girl, hastening the possibility of a new
pregnancy and the birth of the much wished for boy.
In one study of abandoned Chinese babies, almost all were healthy girls
and only a few were boys, all of whom were sick or handicapped.14 Most
abandonments occurred in the first six months of life, the children often
 Babies, History, War, and Politics 29

being left in places where they were likely to be found, or if they were sick,
at hospitals.
Adoption of Chinese orphans by Westerners began in about 1994 and
increased in frequency at a rapid rate. Girls are generally seen as manage-
able, quiet, and affectionate, and Westerners stereotypically think of Asians
as cooperative and intelligent. In addition, these children are babies, mak-
ing them even more desirable.

The Hague Convention

The increasing occurrence of foreign adoption in the early 1990s brought
about an agreement called the Hague Adoption Convention (also called the
Convention on Protection of Children and Co-operation in Respect of
Intercountry Adoption), which was signed in May 1993. The Hague
Convention is intended to prevent commercial trafficking in children and
other possible abuses, as well as protecting the adults and children involved
in international adoptions.
The Hague Convention is implemented in the United States by the
Intercountry Adoption Act of 2000, which sets standards and accreditation
criteria for adoption services and procedures for safeguarding children. In
the United States, the State Department serves as a central authority for
international adoptions. In addition to accrediting intercountry adoption
agencies, the central authority has the essential task of preserving records on
international adoptions.

Research on Development after Adoption from Abroad

The Hague Convention set rules for record keeping, a crucial part of
research on the developmental impact of adoption from abroad. Adoption
agencies and adoptive parents have also kept records, but information about
early foreign adoptions is not necessarily very reliable.
The results of the cruel psychological experiment inherent in large num-
bers of foreign adoptions are, of course, of enormous interest to present and
future adoptive families, and also to theorists and researchers studying at-
tachment, separation, and loss. As adoptions from Romania increased, a
group called the English and Romanian Adoptees (ERA) Study Team be-
gan a major research project.1n One goal was to examine the outcome of
this type of separation experience, something Bowlby's work predicted
would be disastrous.
The research done by the ERA Study Team was extremely difficult. The
serious developmental problems caused by the experiences of the Romanian
orphans could never be studied through experimental (randomized con-
trolled trials) techniques for ethical and practical reasons. The alternative,
30 Understanding Attachment

the quasi-experimental type of study done by the ERA Study Team, yielded
such a mass of puzzling variables that it is difficult to determine cause and
effect or to know whether a child's condition was a result of the orphanage
experience, abandonment, a combination of the two, or the orphan's treat-
ment following adoption. This work involves multiple questions and looks
at the immediate result of the children's orphanage experiences and the
children's ability to recover when placed with adoptive families, making
the resulting data exceedingly complex.
One built-in problem with this type of study is the factor of spontaneous
developmental change. However dreadful the conditions, a child who sur-
vives continues to develop, although not necessarily at a normal rate or
along a normal trajectory. Changes in adopted children are to some extent
simply the result of continuing maturation, though the nurturing provided
by the adoptive family also plays a role. Recovery from a harmful experi-
ence may largely be a matter of resilience—one's natural ability to get back
on a normal developmental pathway. All studies of child development have
to deal with this factor, as did the Romanian orphan research.
The research on the Romanian orphans had other problems. It was dif-
ficult to know what the conditions had been before the children were in-
stitutionalized. Family poverty, maternal malnutrition, maternal depression,
prenatal exposure to alcohol or drugs, problems at the time of birth, genetic
issues—these could all increase a child's vulnerability to deprivation and
lessen the child's response to better care following adoption. Although or-
phanage conditions were uniformly abominable, another difficulty had to
do with knowing an individual child's experiences. Something that clearly
made a difference was the length of time the child spent in the institution.
Children who were very appealing might have received a bit more atten-
tion, and even the experience of being with slightly older children could
have been beneficial (depending on the other children's characteristics).
The results of the ERA Study Team's work are far from simple. Some
of the children, in fact, recovered very well from their experiences after
several years of adoptive life. Others did well medically, but showed prob-
lems in their emotional and cognitive development. Importantly, there
was support for the idea that separation is only one of many important fac-
tors in development and may have a critical impact if followed up by poor
parenting.16
Do the statements in the last paragraph apply to children in U.S. foster
homes as well as to Eastern European orphans? It would be fair to say that
attachment theory has not yet integrated the ERA findings. It may be that
the conclusion above applies when a child has been exposed to many risk
factors, but not otherwise. However, most children in the United States
today do not experience separation except in the context of serious family
 Babies, History, War, and Politics 31

problems and troubled lives, and the ERA Team's conclusion may well
apply to ordinary foster care children, as well as to children dramatically
rescued from appalling institutional conditions.

CONCLUSION
The understanding of attachment has grown gradually over the centu-
ries, but theory development and research have progressed most rapidly in
the twentieth and twenty-first centuries. Early psychoanalytic thought fo-
cused on the relationship between mother and child, theorizing the con-
nection to be based largely on the child's physical needs. Later theories
considered the possibility that social needs and behaviors are innate human
tendencies rather than responses based on hunger and discomfort. Evidence
from studies of children separated from their parents during World War II
suggests that attachment and separation played essential roles in emotional
development, and John Bowlby based an entire theory of attachment on
this and later information. In the 1990s, new research was conducted when
large groups of children were adopted from foreign countries where they
had been institutionalized. This study suggests that separation alone does
not play as important a role in emotional development as Bowlby thought,
but theories of attachment have yet to respond to this work.
This page intentionally left blank
 chapter 3

The Growth of
Attachment Theory:
Connecting Ideas
through Research

T
he end of World War II allowed thinkers, who had for years been
busy with national wartime concerns, the leisure and time to consider
attachment. As we saw in the previous chapter, Britain had become
the center of psychoanalytic thought as the Nazis destroyed German intel-
lectual life. Ethological theory also found a British home with the immigra-
tion of the Dutch ethologist Nikolaas Tinbergen, winner of the Nobel Prize
for Medicine, and his colleague Konrad Lorenz. In postwar Britain, psycho-
analytic and ethological ideas came together, as did the material to which they
could be applied: observations of children separated from familiar people.

FACTORS IN THE DEVELOPMENT

OF A THEORY OF ATTACHMENT
With dreams of a better world, many postwar thinkers sought to improve
conditions for young children's development. In order to accomplish this,
however, more knowledge was needed about the foundations of social and
emotional growth. Conditions and motivation were right for an intense,
systematic exploration of early emotional life, and researchers began to focus
on the details of development. For example, a paper published by Dr. Rene
Spitz in 1945 discussed the occurrence of hospitalism, the slow development
and fragile health of abandoned infants who received little individual atten-
tion in group care. Spitz noticed particularly that these babies developed odd
reactions to strangers. Compared to the behavior of family children, for these
34 Understanding Attachment

children the "usual behavior was replaced by something that could vary from
extreme friendliness to any human partner combined with anxious avoidance
of inanimate objects, to a generalized anxiety expressed in blood-curdling
screams which could go on indefinitely."1 The implications for these chil-
dren's emotional development were worrisome, but the research itself pro-
vided one of many steps in the formulation of a valid theory of attachment.
The new studies were also shaped by an idea shared by scientists in
a variety of fields. For some years before the war, a sophisticated view
of scientific work had been emerging, and in the form of operationism,
this approach to knowledge influenced the investigation of attachment.
Operationism takes the stance that communication about scientific knowl-
edge is primarily communication about measurement. If I want to tell
someone about an observation I have made, I must recount what opera-
tions—that is, measurements—I used in order to keep track of what I saw.
Only if two observers share the same operations do they know whether
they have seen the same thing, and only then can one check his or her
observations against the other's. If observers of little boys and little girls'
attachment behavior use different operations and different definitions, they
are very likely to report that they find sex differences—even though the
children may actually be very similar, if measured with the same ruler. The
social sciences are particularly in need of operational definitions, and this
was a real issue for the early developers of attachment concepts as they
struggled to decide whether a child demonstrated attachment or not.
As theories about attachment were formulated, they were also guided by
another concept connected with operationism: the idea of falsifiability. This
term sounds like it means "fudging the data," but it actually is a way for
researchers to keep themselves honest. Falsifiability is the possibility that
expected research results will not come out the way a researcher thought
they would, and research must be designed in ways that allow this to hap-
pen. A researcher abandons falsifiability if he or she throws out all the data
that do not support the hypothesis—as sometimes happens in the study of
parapsychology, for example.
When researchers stress falsifiability, they are allowing their developing
theory to be guided by systematic empirical evidence. A theory whose
predictions are not always supported by evidence must be modified, and
modification is what modern theorists of early emotional development
have attempted to do as they study attachment.

T W O THINKERS
Out of the foundations of time and place described above came two
people whose work forms the essential basis of modern theories of
 The Growth of Attachment Theory 35

attachment: John Bowlby and Mary Salter Ainsworth. Both Bowlby and
Ainsworth had unusual qualifications in research as well as in clinical
training. Both stressed excellent observation as the foundation of knowl-
edge about child development, prefiguring the mantra of developmental-
ist Sally Provence: "Don't just do something; stand there and watch."
Bowlby's and Ainsworth's lives and thought have been beautifully de-
scribed by Jeremy Holmes and Inge Bretherton, 2 and much of the mate-
rial below has been drawn from that work.

John Bowlby: Clinician and Theorist

The British clinician John Bowlby (1907-1990) brought to the study of
attachment a variety of experiences and training. Bowlby combined an
ivory tower approach with a persistent involvement with real-world prob-
lems. His interest in child development began with volunteer work at a
school for children with problems of adjustment. He became familiar with
one distant, disengaged, aloof boy, and with another who clung anxiously
and stayed close to Bowlby. These children were Bowlby's first examples
of what would later be seen as the two faces of attachment problems.
Studying to become a child psychiatrist, Bowlby also became involved
with the British Psychoanalytic Institute at a period of rapid change in psy-
choanalytic thought about early development. In spite of Sigmund Freud's
relatively small initial interest in the emotional events of the early months,
psychoanalytic theory remained the only approach that had even a minor
focus on early development, and this focus was sharpening. The British
Psychoanalytic Institute was moving toward a formulation of object-
relations theory, a set of ideas about the development of early relation-
ships with people and the ill effects of separation and loss.
A leader in the British psychoanalytic group was Melanie Klein,3 who
proposed the possibility of direct psychoanalytic treatment of the very
young. (Sigmund Freud had attempted treatment of children's problems
only through the intervention of parents.) Klein eventually supervised
Bowlby's early analytic work, but the two disagreed profoundly on certain
issues—perhaps motivating Bowlby to demonstrate through research that
he was right on some important questions.
Melanie Klein stressed children's internal conflicts between love and hate
as the source of emotional problems. She considered the fantasy life of in-
fants to be more important than actual family events in determining mental
health. Bowlby disagreed, basing his opinion on real-world experiences,
such as training with social workers at the London Child Guidance Clinic.4
In the little time available during the war years, Bowlby had begun his
efforts to investigate the effect of children's real experiences on their
36 Understanding Attachment

emotional development. He studied the cases of forty-four delinquent

boys at the London Child Guidance Clinic, 5 an investigation that his col-
leagues dubbed "Ali Bowlby and the Forty Thieves." Bowlby showed
these boys to have experienced separation from their mothers and to have
been deprived of maternal care.
Linking his research with clinical practice, Bowlby proceeded after the
war to head the children's department at London's Tavistock Clinic. He
changed the name to the Department for Children and Parents and used
the parents' own childhood experiences as material in his work with their
children. As Bowlby continued his clinical and research work, with its em-
phasis on family relationships, a number of ideas and investigations con-
tributed to his thinking about emotional development. What began as a
general viewpoint slowly became a theory of attachment.
Films of children suffering from the effects of separation provided Bowlby
with evidence for the role of familiar caregivers in the emotional lives of
the very young. James Robertson provided filmed material, such as A Two-
Year-Old Goes to Hospital and Nine Days in a Residential Nursery,b which
documented the child's confusion and grief upon separation, and his in-
ability to respond positively to a reunion with the mother.
Experimental work on animals also provided evidence that influenced
Bowlby's theoretical approach. Harry Harlow of the University of
Wisconsin was studying rhesus monkeys in an attempt to determine how
monkey mothers became capable of effective care of their young—and
why they, like humans, were sometimes neglectful or abusive. Harlow's
work focused on the animals' early experiences with caregivers. For ex-
ample, he provided surrogate mothers in the form of monkey-sized wire
figures; some held a bottle of milk and others had soft padding to which the
babies could cling. Harlow's showed that although baby monkeys could
survive these conditions, their later behavior was much affected by their
experiences.7 The ability to court and mate was impaired, and when the
females did become pregnant, their treatment of their infants was inappro-
priate and harmful. Harlow's work substantiated Bowlby's belief that early
experiences with caregivers could set a direction for good, or poor, social
and emotional development.
As Bowlby became familiar with the ethological view, mentioned in
the previous chapter, another type of animal research provided him with
some essential ideas about attachment. The most important ethological
concept, from the perspective of Bowlby's work, was the young animal's
rapid connection with an adult. Imprinting was an intriguing phenome-
non for several reasons. It involved a social bond that might be comparable
to the emotional tie between human parents and children. It was not
linked to feeding, demonstrating that social relationships could have their
 The Growth of Attachment Theory 37

own primary motivations. It occurred very readily during a period of de-

velopment comparable to the one that most interested Bowlby. Imprinting
was also an event with a long-term effect on the individual's social and
emotional life, as Bowlby thought might be the case for children.
Most theories develop gradually, but there comes a moment when they
coalesce and are presented to the world in a nearly complete form. Bowlby's
attachment theory reached this stage in the late 1950s, and it was published
in several important papers.8 These papers drew together Bowlby's ideas
about the reasons for children's emotional attachments to their mothers and
caregivers, about the observable behaviors that indicate attachment, and
about the consequences of separation from a familiar attachment figure.
Although Bowlby's ideas became further elaborated over the years, these
papers are the true foundation of modern attachment theory.

Bowlby s Basic Theory

Any theory is a framework of ideas that connects observable events and
helps to explain the ways in which events are caused. We derive hypotheses
or predictions about the world from theories. If testing of these predictions
does not give supportive evidence, the theory needs to be reshaped. As was
mentioned earlier, a good theory must be falsifiable. It must be possible to
know when the observations do not confirm the prediction.
Bowlby's attachment theory does meet these criteria. It connects observ-
able events, such as social experience, separation, and later behavior; it al-
lows us to make predictions; and it makes it possible to know when a
prediction is wrong. (This does not mean, of course, that testing the theory
is necessarily a simple matter. Research with infants and families has many
pitfalls, and there are serious challenges to investigations where experimen-
tal manipulation is impossible, as is true about many aspects of attachment
work.)
Wltat Are Human Beings Like? Bowlby's attachment theory may be easier
to understand if we first state some of its underlying assumptions about the
nature of human beings. One important assumption is that humans share
many motivations and behaviors with nonhumans. Although the ethologi-
cal view says that some characteristics are species-specific, other character-
istics are shared across species. More closely related groups, such as humans
and monkeys, might be expected to share more motivations and behaviors
than, say, humans and fish. Information about animals' lives may thus tell
us important things about human beings.
A second assumption is that human motivations and behaviors can be
genetically controlled, as well as learned, and that when this is the case their
presence has been influenced by evolutionary factors. Such behaviors and
38 Understanding Attachment

motivations presumably contributed to survival long ago, in our environ-

ment of early adaptation. Because young human beings are relatively help-
less (compared to many other species), factors that helped them survive
their early years would be prone to natural selection. These factors would
still be present in modern humans, even though technology and social or-
ganization may have made them irrelevant to survival. For example, babies
in prehistoric times might have been more likely to survive if they stayed
close to adults; modern infants still show this tendency, even though living
in houses and other such changes have removed much of the value of this
behavior.
The Theory of Attachment. Bowlby's theory of attachment included a num-
ber of clear statements about children's emotional ties to familiar caregivers:

1. Children's attachment to their mothers, or to other familiar caregivers,

results from a basic human motivation. It is not based on gratification
of physical needs, such as hunger. The desire to be near familiar peo-
ple, a need in itself, is independent of other experiences and is a healthy
and desirable part of human relationships. Even in adults, attachment
should be seen as an important continuing factor in social life, not as
regression to a baby-like dependency.
2. The infant behaviors connected with social interactions, engagement
behaviors such as sucking, clinging, following, and signaling another
person by smiling or crying, are all related to the development of
attachment. They are also all inherited, instinctual behaviors character-
istic of human beings. These tendencies were selected in the course of
evolution because they made it more likely that babies would be at-
tended to, cared for, and helped to survive.
3. Some of the actions that will form part of attachment behavior are al-
ready present between birth and six months of age, but they are not
particularly focused on a specific one or two adults. After six months,
the baby begins to crawl and to follow adults, and soon attachment
behaviors begin to be directed toward a few familiar caregivers. From
the standpoint of evolutionary advantage, this delay in attachment is an
excellent arrangement, protecting babies from the traumatic experi-
ence that early separations might otherwise entail. If attachment al-
ready existed at birth or in the early months, the high maternal death
rates that used to occur soon after birth would potentially have caused
many emotional problems. It was a good bet that a caregiver who was
still alive when the baby was six months old would probably live for
another year or two, giving the baby time for secure emotional devel-
opment. With modern medicine, maternal death rates are quite low,
and separations do not necessarily take place in the very early months,
 The Growth of Attachment Theory 39

but this recent change does not erase the characteristics that evolved
long ago.
4. Separation anxiety is an indication that attachment has occurred and
therefore a normal and desirable event in emotional development.
Ordinarily, the child's distress is soon relieved by reunion with the
caregiver. When separation from familiar people is abrupt and long
lasting, however, the child who has achieved attachment is not only
anxious, but can be expected to go through predictable stages of griev-
ing and emotional reorganization. Although the steps of infant grief are
intense, worrisome, and demanding of adult support, they are never-
theless a normal and necessary response to serious separations. Neither
separation anxiety, nor grief folio wing serious loss, would be expected
until after there had been a focus on a familiar caregiver, and as we saw
above, this does not happen until six months at the earliest.
5. Although Bowlby did not emphasize this point, it is important to real-
ize that attachment emotions and attachment behavior are age-related,
with upper as well as lower limits. Just as the characteristics we have
discussed are not displayed before six months, at the earliest, they do
not occur in the same form after, perhaps, three years of age. The at-
tachment behaviors Bowlby initially described are typical of the late
infant period, the toddler stage, and the early preschool years. Emotional
ties alter and modulate with age, and so do related feelings and actions,
as we will see in a later chapter.
6. In later work, Bowlby suggested that attachment behaviors and emo-
tions are connected with the individual's internal working model of
the social world. This suggestion offered a way for attachment to
change developmentally as the individual passed the age period in
which instinctual factors played the strongest role. An internal working
model is a set of memories, emotions, and thoughts that determines a
person's expectations and attitudes, and that consequently shapes be-
havior. Attachment-related behaviors like preferring certain people,
maintaining near or far physical distance from those attachment fig-
ures, seeking help from people, and displaying distress upon separation
or loss are determined by this internal working model.
Of course, attachment-related behaviors are not necessarily directed
toward a consistent set of people throughout a lifetime, nor are the
behaviors themselves always the same. These facts show us that the
internal working model changes with age and experience. For example,
a baby girl may crawl rapidly to her mother and signal to be picked up
when frightened by a loud noise. Thirty years later, she may look
around to locate her husband when jostled in a crowd of people,
or may move to protect her own baby, ignoring potential threats to
40 Understanding Attachment

herself. Her internal working model of social relationships has gone

through thirty years of experience and maturation and now shapes a
different set of feelings, attitudes, and behaviors than it did in the
woman's own infancy.

Mary Dinsmore Salter Ainsworth:

Attachment Ideas, Attachment Measurement
U.S.-born, Canadian-reared psychologist Mary D. Salter Ainsworth
(1913—1999) made her own theoretical contributions to the understanding
of attachment, but her work on measurement of emotional development
gave attachment theory the self-correcting power that we demand of modern
theoretical formulations. Without the ability to measure attachment, it
would be impossible to create and test hypotheses, and, therefore, impos-
sible to know whether attachment theory was a good description of early
emotional life. It is for her measurement techniques and her organization
of research programs that Ainsworth's name is joined with Bowlby's in the
history of attachment theory.
Measurement in the Early Days of Attachment Theory. From the beginning
of serious work on attachment, accurate measurement and empirical evi-
dence were emphasized. Bowlby used case studies and narrative descrip-
tions to find evidence to test his developing theory. One of the appeals of
ethology for Bowlby was its meticulous description of animal behavior.
Bowlby regarded all relevant empirical material as useful to the understand-
ing of emotional life. Melanie Klein had forbidden him to interview the
mother of a three-year-old he was treating, a factor in Bowlby's opposition
to her thinking.
Following the war years, the films made by Robertson and by Spitz
provided a new way of collecting and displaying information about child
development. Many of us feel suspicious of evidence presented on film,
knowing, as we do today, the ease with which editing can distort reality.
Bowlby, too, was aware of this possibility, and he guided the making of
some of Robertson's films in ways that made them reliable. For example,
a child who was to be followed in one film was chosen at random from
a group, and the hospital clock was included in frames as an indication
of time-sampling.9
Ainsworth's Measurement Work. From the time of her doctoral disserta-
tion, completed in 1940, Mary Ainsworth was fascinated with ways of re-
cording memories, emotions, and behaviors. She was particularly interested
in the problem of checking one type of measurement against another in
order to achieve the highest level of accuracy. In her dissertation research,
for example, she asked students to fill out surveys about their family
 The Growth of Attachment Theory 41

experiences, but she also examined their autobiographical reports to make

sure that the simple paper-and-pencil tests gave accurate measures of their
complex life stories.
With her new husband, Leonard Ainsworth, Dr. Mary Ainsworth went
to London in 1950 and by chance answered an advertisement that led to
work with John Bowlby's research group. She became involved in data
analysis of the material filmed by Robertson under Bowlby's guidance.
Although she admired Robertson's observational and descriptive meth-
ods, Ainsworth was still not convinced that Bowlby's ethological approach
made sense. Her own interest in attachment issues led her to a study of
school-age children who had been separated from their mothers for lengthy
medical treatment. Her assessment classified these children into three
groups: those who were happy and positive about reunions with their
mothers, those who were ambivalent in their feelings, and those who were
indifferent or actually hostile toward their mothers. As we shall see shortly,
this three-way classification was later to form a crucial part of Ainsworth's
measurement of early attachment behavior.
Cross-Cultural Work. During the 1950s, Mary Ainsworth spent time in
Uganda with her husband, where she had the opportunity to observe moth-
ers and children in a different culture. If attachment behavior was really
species-specific and instinctual, as Bowlby was supposing, the observations
made in Uganda should closely resemble the findings in England.
Ainsworth arranged to visit twenty-six families with babies. She went
every other week to sit in the family living room and observe how the
growing babies began to stay closer and be more attentive to their mothers.
She gave special attention to the mothers' sensitivity to the infants' signals,
the babies' tendencies to cry or to explore, and (through an interpreter) the
mothers' reports of whether they enjoyed breastfeeding. As we will see in
a later chapter, the results of this study were much as Ainsworth and Bowlby
would have predicted; the infants of Uganda showed attachment much as
the British babies did.10
Observations in Context: The Baltimore Project. In the early 1960s—when
Bowlby had already published his three papers outlining attachment
theory—Mary Ainsworth moved to Baltimore. It was here that she set up
a complex research project for observation of mothers and infants.
Ainsworth's naturalistic observations were not only meticulous and de-
tailed; they had an additional characteristic that made them particularly
critical to the study of attachment. Like the Uganda work, Ainsworth's
Baltimore observations explored the mother's contributions to the attach-
ment relationship, not just the child's. The babies of mothers who were
sensitive and responsive had babies who showed one pattern of attachment
behavior, and those with mothers who were unresponsive and uninterested
42 Understanding Attachment

showed another. (Then, as now, it was not completely clear whether babies
influenced mothers, mothers influenced babies, or each affected the other.)
Ainsworth's Baltimore research, a model demonstration of a new ap-
proach, involved the study of connected patterns of behavior, rather than
old-fashioned frequency counts of events like crying.11 More complicated
and difficult to do than the older style of research, Ainsworth's work was
also far more capable of displaying the complexities of developmental
change. By demonstrating these complexities, Ainsworth made a critical
contribution to attachment theory as Bowlby was elaborating it.
Showing how mothers' actions guided their children's attachment was
valuable in itself, but it also demonstrated the importance of the internal
working model of social relationships. Repeated acts by the mothers trig-
gered responses in the babies, but these involved more than the primitive,
instinctual level of action suggested by the ethologists. These repeated
events let babies learn what they might expect from other human beings.
They established a pattern of expectations and attitudes that started with the
earliest relationships and would be generalized and applied to new people.
For example, the expectations learned from a mother would also be ex-
tended to a child care provider and later to a kindergarten teacher, and so
on. Importantly, the internal working model could change and develop as
the child grew and had new experiences. Early instinctual behaviors did not
change with experiences, but the internal working model of attachment
could continue to alter, eventually giving rise to the individual's attitude
toward his or her own child. Lest this description of the internal working
model sound too cool and intellectual to have anything to do with attach-
ment, let us recall that the model includes emotion and motivation based
on the powerful early feelings of the toddler seeking his or her mother's
comfort and fleeing the stranger's approach.
The Strange Situation: A Behavior Sample. Ainsworth was now interested
in two new measurement problems. Visiting families in Uganda and in
Baltimore had been enormously useful, but the collected information was
very difficult to analyze. Families can be quite different. How can we com-
pare them to each other? And, if the mother's behavior is so important,
how can we possibly tease it out of the complications of real-life situations
where mother, father, children, culture, home, and accident determine
how people act?
Ainsworth wanted to establish some way of describing infant behavior
which would allow her to compare babies, or families, or cultural groups
without being confused by background details. She was especially inter-
ested in how infants behaved in high or low stress situations, but how
could she create or reduce stress in a family, even temporarily? She needed
a standardized test, one in which the situation could be controlled, all
 The Growth of Attachment Theory 43

mothers and children had the same experiences, and trained observers
could collect the same information each time. Only then would it be pos-
sible to make comparisons or even to find norms and understand what
were frequent and what were unusual behavior patterns.
The result of Ainsworth's efforts was a test known as the Strange Situa-
tion. This measurement device has been the basis of thousands of research
studies. Indeed, we may wonder whether the topic of attachment would
ever have achieved its present eminence without the Strange Situation as a
means of exploration, and as an answer to the question: how in the world
can we compare these babies?
The Strange Situation. Like many other psychological tests, the Strange
Situation involves a narrow sample of information drawn from a wide range
of measurable factors. First, it is intended to test babies at about twelve months
of age (although some researchers have not paid much attention to this). All
the evidence shows that the great majority of babies of this age show attach-
ment behaviors, readily and intensely. This is the best time to catch behav-
ioral evidence of the child's emotional connection to a caregiver.
Second, the Strange Situation chooses a simple, common, mildly threat-
ening situation, and keeps it very brief. Though an effort is made to limit
unhappiness, the intention, of course, is to cause enough distress to the
child to elicit attachment behavior. This approach produces stress at the
level a child might feel when the mother steps away briefly when shop-
ping, but not nearly at the level of attachment behavior that would occur
when, for example, a parent and child are both terrified by a ferocious dog,
a parent runs to help another child who is hurt and screaming, or there
is domestic violence.
The Strange Situation: Wlwt Happens, What Is Measured. The Strange Situa-
tion is intended to assess the toddler's responses to a brief separation from
the mother, followed by a reunion. It takes about twenty-two minutes
altogether and requires a room with certain features and an adult who is a
stranger to the child. There are eight steps or episodes in the Strange
Situation procedure.

Episode 1: The mother and baby enter the room with an observer who
shows the mother where she can put the baby down and then where she
can sit.
Episode 2: The mother puts the baby down near some toys. The mother
does not start playing with the toys, but she may respond if the baby
starts, for example, to bring a toy, put it in her lap, or hand it to her. If
the baby has not started playing after two minutes, the mother may take
the baby to the toys. (For completeness, we should note that by this age
44 Understanding Attachment

normal babies can crawl or even walk alone; they can pick up toys, put
them down, or throw them; they understand some words and may speak
a little. Although they do not play in very complex, planned ways, they
are interested in toys; they like to pick them up, mouth them, throw
them, or bang them against things. If they are walking, they often like to
carry things, especially big things.)
Episode 3: A stranger comes into the room, greets the mother and baby,
and sits quietly near the mother for one minute. During the next minute,
the stranger talks with the mother. Then for a third minute, the stranger
gets down on the floor and tries to play with the baby. The mother then
leaves the room quietly. The baby generally notices that she is leaving.
(This introduces, in a mild form, the stress factor that interested
Ainsworth.)

Episode 4: The mother is still absent. The stranger sits on her chair, but
responds to the baby, if the baby tries to start play. The stranger offers
comfort, if the baby shows distress. The mother stays out for three min-
utes, if the baby is not upset, but if there is distress, the mother returns
and the stranger leaves.

Episode 5: The mother calls the baby's name from outside the door and
then comes in. If the baby needs comfort, the mother offers it and tries to
get play started; if not, she sits on her chair and responds, but does not start
play. After three minutes, she leaves, saying, "Bye-bye, I'll be back soon."
Episode 6: The baby stays alone for three minutes. If there is distress, the
stranger comes back.
Episode 7: The stranger offers comfort, if the baby is distressed, or else
stays seated on the chair for three minutes. If the baby is too upset, the
mother returns before the three minutes have elapsed.
Episode 8: The mother returns, the stranger leaves, and the mother
spends three minutes, as she did in the first episode.

During the eight episodes of the Strange Situation, trained observers

look at certain behaviors as they occur in each episode: The amounts of
play and exploration that occur when the baby is with the mother or with
the stranger and the amount of crying and the ease with which the mother
can comfort the distressed child. And, above all, the observers look at
reunion behaviors—the response of the child when the mother returns to
the room.
Classification of Infants' Attachment. A test like the Strange Situation is not
very useful, if every child is described separately and no comparisons or
 The Growth of Attachment Theory 45

connections are made between children. One of the points of having a

standardized test is that we can classify or group children who are similar to
each other. We can then search for factors that might explain what causes
individuals in one category to be different from those in another. Researchers
who use the Strange Situation put children into categories on the basis of
their behavior, and the categories they use are much like the ones Ainsworth
used in the studies she first did while working with Bowlby.
In one careful study, the largest group of twelve-month-olds tested in
the Strange Situation—about 65 percent—was classified as securely attached.
These toddlers played and explored when with their mothers, but played a
little less when the stranger came, in Episode 3. They began to cry shortly
after the mother left. The reunion when the mother returned was a happy
one. These babies actively sought contact with their returning mothers,
and they were easily comforted and ready to play again. (This group was
called Group B by the researchers.)12
The second group, Group A, was smaller and included about 20 percent
of the tested babies. These insecure-avoidant children were less concerned
with the stranger and did not necessarily cry when the mother left. The
reunions were not happy for the mother or the baby. These children ac-
tively avoided the returning mothers, rather than seeking them, and actu-
ally snubbed the mothers who tried to approach them.
About 15 percent of the toddlers were classified in Group C, with
insecure-ambivalent attachment. They were anxious and explored little, even
in Episode 2 when the mother was present. During reunion with the
mother, these children simultaneously went to her and pushed her away or
resisted being picked up. The mothers did not seem able to comfort the
distressed children.
A fourth classification, Group D, was applied to a small proportion of
tested children by another research team.13 The Group D children were
designated as being insecure-disorganized/disoriented, and their behavior was
quite unusual. In fact, although most of us have seen children who showed
the A, B, or C patterns, the D pattern is not often observed.
Group D babies are most different from Groups A, B, and C in their
reunion behavior. When the mothers returned to the room, these Group D
children did not behave in uniform, predictable ways. Their feelings seemed
contradictory, and they might vigorously avoid and then vigorously ap-
proach the mother. They might get into physical contact with her and then
turn their heads to gaze away. They sometimes had facial expressions of
fear, seemed confused and apprehensive, rocked themselves back and forth
in a stereotypical way, or suddenly fell to the floor looking dazed. Babies
who behave like those in Group D obviously need further investigation and
possibly treatment. These issues will be discussed in a later chapter.
46 Understanding Attachment

OTHER MEASURES O F A T T A C H M E N T
Fascinating and productive as the Strange Situation is, it is not the final
step in measuring attachment patterns. As human beings get older, they
continue to have emotional attachments to others, but the ways they show
these attachments change with age. Even preschoolers placed in a situation
like the Strange Situation do not behave in exactly the same ways as twelve-
month-olds. By adulthood, people still have emotional bonds and impor-
tant internal working models of social relationships; these are expressed in
mature ways, not like those of an infant. (We would be very worried about
an adult who cried when a loved one had been away for three minutes.)
Individuals who are past the toddler stage follow different rules, and they
cannot be tested in the same manner.
There has been work in assessment of attachment in preschool children,
using techniques such as asking children to finish incomplete stories.
Assessment of school-age children has proved more difficult, and the
eminent child psychiatrist Charles Zeanah has expressed doubt that any-
one is presently capable of managing this. One approach for preschool and
school-age children has been the use of a Q-sort technique in which
observers decide which type of attachment is characteristic of a given
child.14
More recently, the attention of researchers has turned toward the as-
sessment of adult attachment relationships, including connections with
parents, marriage and friendships, and attitudes toward children. Again, this
kind of measurement is not going to resemble the Strange Situation. Adults
do not show the same responses to separation as toddlers do, though there
are some similarities, such as behaving differently when around strangers
or feeling ambivalent when reunited with a person who has made us feel
deserted.
The study of attachment relationships in adults focuses on the internal
working model of social relations, which has presumably changed, devel-
oped, and become far more elaborate as the years have passed. We may be
able to get at that model more effectively by talking about relationships,
rather than by looking at overt behavior. The Adult Attachment Interview
is a way to do this.13
The Adult Attachment Interview uses open-ended questions about
childhood experiences to get at the adult's internal working model. The
specific experiences reported are meaningful. Stories of a family cooperating
happily suggest a very different working model than do tales of bleak years
in an orphanage or an abusive foster home. But analysis of the Adult
Attachment Interview goes beyond the stories that are told to the way the
interviewee tells them. Do the stories give an organized, coherent picture
 The Growth of Attachment Theory 47

of the individual's thoughts about family history? Are there separate stories
that do not form a consistent pattern? Does the individual speak in clear,
descriptive sentences, or hesitate, trail off, begin again? Does the inter-
viewee claim that he or she does not remember much about childhood
family relations and dismiss them as a matter of little interest?
As was the case for the Strange Situation, responses to the Adult
Attachment Interview can be placed into a few categories. Adults are con-
sidered to be in the autonomous-secure group, if they recount an organized
story of childhood memories—whether or not they find the memories
pleasing or satisfying. The term preoccupied is used to categorize adults who
have many, contradictory childhood memories, but no way of organizing
them into a meaningful whole. Finally, those who state that they have few
or no memories of early family relations are classified as showing a dismissing
pattern. Adults' responses to the AAI are quite reliable, that is, researchers
usually find similar results each time, and these results do not seem to be
related to other memory or intelligence factors.16

ATTACHMENT THEORY A N D
MEASUREMENT: LINGERING QUESTIONS
Our historical account has shown how attachment theory has grown
over the last fifty years or more, but there are some broad issues and impli-
cations that still need consideration. The very idea of emotional attach-
ment, as Bowlby described it, is attractive to some and almost repulsive to
others. Some responses seem almost parallel to the early rejection of Freud's
view of infantile sexuality. College students are often negative about the
idea of attachment, perhaps fearing that the autonomy desirable at their
present age would be impossible if attachment had already occurred. Others
who are newly introduced to the attachment concept become thoroughly
confused. They try to puzzle through Bowlby's theory and find it disturb-
ing that they cannot reduce this complex system to one or two simple
foundations. To close, we need to deal with some general questions about
attachment theory, as it exists today.

Is There a Precise Definition of Attachment?

The simple answer to the question: what is attachment? is that it refers to
a long-lasting emotional tie between a child and a familiar adult—one that
lasts even after the child is an adult. Like most simple answers, of course,
this one leaves much to be desired. Several points need to be considered in
order to achieve a more exact definition and to go beyond knowing it
when we see it.
48 Understanding Attachment

A tie may be best understood as an internal emotional state that causes

a predictable pattern of positive, attentive, interested behavior of one per-
son toward another. In the case of attachment, however, there is a com-
plication. Although we can assume that a tie lasts for a long time, the
behavior pattern is stable and predictable only for short periods of time.
The toddler's preference to stay close to a caregiver and his or her tears
upon separation are predictable for just a few months; after that, the child
will show this marked preference only when under stress. The preschooler
who turns to a parent for help and care is showing a positive behavior pat-
tern, but it is not the same pattern we see when an adult cares for a much-
loved, elderly parent. A long-lasting emotional tie gives rise to different
behavior patterns, as both the child and the caregiver change and grow
over time.
Of course, much of the emotion connected with attachment is positive
in nature. Our desire and preference for other people's company are defin-
ing characteristics of attachment. However, this long-lasting emotional
connection involves a variety of feelings that include not only happiness
but also relief after a reunion and anger, grief, longing, jealousy, ambiva-
lence, and resentment after separation. The quality, intensity, and duration
of these feelings change with age and because of other factors, such as the
behavior of the other person.
The emotional connection or tie between child and adult implies special
characteristics. The concept of attachment focuses primarily on the child's
emotions and internal working model of social relations. The attachment
tie involves the child's preference for the adult's company and the wish to
be cared for by the adult. However, the nature of the tie, the child's expec-
tations, and the child's experience of emotional interaction with the adult
depend, in part, on the adult's feelings and behavior. Unless the adult also
has some positive feelings toward a particular child, the child is not likely
to have the caregiving and social experiences that create a long-lasting tie.
The relationship is not in fact one-sided, but the study of attachment tends
to emphasize one side—that of the child.
Though a long-lasting emotional tie is determined in part by the adult,
it relies even more strongly on factors internal to the child. According to
Bowlby's attachment theory, infants bring to social relations a set of
built-in, hard-wired instinctual behaviors toward the environment and the
caregiver. These are essential for the formation of attachment, and they
include a readiness to prefer familiar rather than unfamiliar people and
demonstrate behaviors like looking at and following them. As age and the
internal working model progress, however, the instinctual behaviors be-
come less important. They are replaced by the set of emotions, memories,
and beliefs that maintain the attachment tie after early childhood—often
 The Growth of Attachment Theory 49

into adulthood—and that allow early attachment experiences to influence

other, later emotional connections.
Attachment is thus, in part, an emotional tie, but it is also the developing
of a set of motivations, behaviors, ideas, and feelings. These cause our dif-
ferent responses to familiar people and to strangers, especially when con-
cerning our expectations that others will care for us or that we will care for
them. The details of attachment feelings and behavior change drastically
from early to later life, but the internal factors seem to be consistent as an
individual matures.

Is Attachment Good?
The short-term advantages of an infant's attachment are evident. In early
childhood, the major behavior linked to attachment involves staying near a
familiar adult, especially in a strange place, or when anything unusual is
happening. The wish to stay near acts as an invisible playpen and has enor-
mous safety value for creepers and toddlers. Unexpected events or strange
people send the child scurrying to the caregiver's side. Although adults al-
ways need to monitor young children carefully, normal attachment behav-
ior means that the adult does not have to spend all of his or her time
retrieving a wandering child.
Whether attachment is desirable over the longer term is a question fre-
quently asked by young parents. Parents in the United States place a high
value on independence and are concerned with the idea that attachment in
early childhood might culminate in a weak, dependent adult personality.
The idea that parents should be sensitive and responsive to infant's signals,
and should let the child stay close, arouses in many American parents the
fear that the child will be spoiled, demanding, selfish, cranky, and undisci-
plined. American parents often prefer the idea that their children will grow
up into independent, self-sufficient, friendly people. They may be con-
cerned that long-lasting emotional ties will somehow make these character-
istics impossible to achieve. Some have even expressed the wish that their
day care center frequently change children's caregivers to prevent attach-
ment,17 or they have rejected breastfeeding because they feel it may cause a
baby to be too attached to the mother.
As we will see in later chapters, however, there are probably no grounds
for the concern that attachment will cause harm, especially in the form of
clinginess and dependency in later life. On the contrary, there is a good
deal of evidence that secure attachment provides an excellent foundation
for personality and social development. A securely attached child has a
better chance to be both independent and friendly than one who has had
poor early relationships. Adults who have had good early attachment
50 Understanding Attachment

experiences have advantages in marriage and parental relationships, al-

though it is certainly possible for people to overcome moderate early at-
tachment problems.

Why Don't Children Show the

Same Attachment Behavior All the Time?
If attachment involves an internal working model, we would surely ex-
pect individuals to show the same attachment behavior at all times, but they
do not. Likewise, we might expect children growing up in the same family
to have similar internal working models of relationships and attachment
behavior, but this does not necessarily happen. Why is this? Do these facts
indicate that the idea of an internal working model is wrong?
When we think about the way attachment changes with age, it's obvious
that a particular child will show changes in attachment behavior over
months or years. It is also true that a child can show urgent, intense attach-
ment behavior on one occasion and seem quite nonchalant about separa-
tion on the very next day.
Situational factors are important triggers of attachment behavior. Securely
attached children do not show much concern about staying near familiar
people until something happens that is construed by the child as a threat of
separation.
Here is an example of a change in attachment behavior that depended on
circumstances.18 An unusually tall, male college student observed toddlers
and preschool children walking with their parents in a shopping mall. If
allowed to walk by themselves, the children lagged as much as twenty feet
behind their parents. The children were capable of keeping up with the
parents, but if the parents stopped to let them catch up, the children also
stopped, maintaining their chosen distance. When the large student ob-
server approached the parents to ask some questions, the lagging children at
once closed the gap and stayed close to the parents while the conversation
continued.
Did these children suddenly experience an alteration in instinctual pro-
cesses or in their internal working models? No, of course not. What they
did experience was a change in the environmental situation in which they
and their parents were functioning. A stranger approached, and he was
unusually big, as well as being personally unfamiliar. The children's re-
action tells us something quite important about attachment behavior:
ordinarily, circumstances must stimulate the occurrence of the behavior.
Internal processes alone are not enough to cause observable attachment
behavior.
 The Growth of Attachment Theory 51

The circumstances that trigger attachment behavior involve fear of

strange people or unusual events, especially those that imply separation. No
actual threat need be present, and there need not be any potential for real
harm to child or parent. The triggering events are different for human be-
ings at different ages and developmental stages. For babies, at the end of the
first year, attachment behavior—crying and following—may be caused by
ordinary events an adult would find quite unthreatening. For instance, a
mother is sitting on the floor playing with her baby. About to sneeze, she
jumps up and rushes across the room to get a tissue. The baby wails and
scrambles after her, as if there is a genuine threat of abandonment. Similarly,
the garbage truck that has come twice a week all the baby's life makes a
loud noise, and the child rushes to be near a familiar person.
In the temporary absence of triggering events, these children might show
very little behavior characteristic of attachment. They may wander far in
their explorations and cry very little, but should a strange event occur, all
the behavioral evidence of attachment is displayed.
In older children and adults who are less easily threatened, we often see
that illness or injury brings back attachment behavior common to babies.
Stress or fear of any type seems to bring out the longing for familiar people.
Most of us feel that we would rather die at home with those we love than
with strangers in the hospital, and this is not just because we think we will
receive better care. One elderly lady, bedridden at home after a stroke,
wanted to have photographs of all her children and grandchildren placed by
her bed so she could spend her days gazing at them.
When children in the same family have different attachment behaviors,
there are at least two factors that may be at work. First, there is no guaran-
tee whatsoever that children in the same family will have had the same at-
tachment experiences. Each relationship is unique, so even identical twins
may experience different interactions with their caregivers. Second, each
child's emotional life is influenced by temperamental factors. Temperament
is an individual's biologically determined tendency to react to the world in
certain ways. Each human being has a unique set of reactions to life events,
and these help determine how attachment behaviors and emotions are dis-
played. For example, a toddler who is temperamentally calm, mild, and
cheerful is less likely to show intense attachment behavior than a brother or
sister who is easily distressed and negative in mood. This difference be-
tween two children does not mean that one child is less securely attached
to a parent than the other.
Differences in attachment behavior between children, as well as differ-
ences in the same child on different occasions, occur because such behavior
is not just a simple reflection of an emotional tie. The internal working
52 Understanding Attachment

model, temporary circumstances, and other personality characteristics work

together to produce a child's reaction to separation.

Is Bowlby s Theory the Only Attachment Theory?

Many careful studies have described attachment behavior, and there is
little question about the facts of attachment. The theory created by Bowlby
and Ainsworth is the most elaborate and fully realized effort, of those at-
tempted thus far, to understand why human attachment behavior exists.
However, Bowlby's theory of attachment is not the only explanation of
attachment behavior. Freud's suggestions have already been described in an
earlier section of this book. More recently, attempts were made to discuss
attachment according to the behaviorist perspective, a view quite differ-
ent from that of Freud or Bowlby. This approach suggested that mothers
and infants provided positive reinforcement for each other's attention,
thus, mutually shaping mother-child attachment behavior.19 As psychology
changed over the years, it left the behaviorist approach behind, although
the idea of mutual influences has remained important.
As the field of psychology continued to develop over the last twenty or
thirty years, the innate, unlearned characteristics of human beings became
an important focus of study. Bowlby's view that attachment began with
instinctual processes fit well into this approach. As psychologists concerned
themselves with innate human characteristics, they returned to an old idea:
that the inborn nature of humans is caused by genetic factors, and therefore
could have been shaped by evolution. The characteristics best suited to
survival in the ancient environment of early adaptation, would be the char-
acteristics selected for, and passed down to, modern humans. These could
include mental and behavioral tendencies, as well as physical structures.
Ideas about the inheritance of human characteristics have become an
organized viewpoint called evolutionary psychology. Without denying that
human beings can and do learn from experience, evolutionary psycholo-
gists concentrate on the inheritance of behaviors that would have helped
in the survival of our remote ancestors. Such inherited characteristics in-
clude emotional responses, like anger and jealousy, that may be maladaptive
or dangerous today, but were useful in the distant past. Social emotions,
social behavior, and emotional relationships are important to evolutionary
psychology because human beings ordinarily live in groups and depend on
each other for survival.
Evolutionary psychologists have not proposed an attachment theory that
is different from Bowlby's, but they have seen the Bowlby theory as
congruent with their viewpoint. (Bowlby himself was deeply interested in
evolutionary concepts and wrote a biography of Charles Darwin.) 20
 The Growth of Attachment Theory 53

Attachment theory assumes that certain important human tendencies are

inherited, and that an infant's responses to caregivers are instinctual in
nature. The existence of these inherited characteristics is connected with
the survival needs of our ancestors, as well as with the needs of modern
humans. Attachment is, of course, an important example of the human
social relationships that are essential for the survival of the individual and
the species. Bowlby's attachment theory is, thus, very much in the main-
stream of modern psychological thought, making this an important reason
for continued interest in this approach.
Present-day work is beginning to ask some questions about attachment
that differ from Bowlby's. It may be that attachment will soon be seen as
only one aspect of developing relationships with others. Chapter 8 will
summarize some of this new work.

CONCLUSION
Our growing understanding of attachment has involved the interweav-
ing of measurement techniques and theoretical changes. Modern demands
for systematic evidence encouraged the formulation of measurement pro-
cedures that made it possible to test aspects of attachment. The data pro-
duced by those techniques was fed back into theory and stimulated new
ways of thinking. Attachment theory developed in step with attachment
measurement, but at the same time was influenced by and instrumental in
the growth of evolutionary psychology.
This page intentionally left blank
 chapter 4

Attachment, Age,
and Change: Emotional
Ties from Birth to
Parenthood

O
ur struggles in the last chapter with the definition of attachment
made it clear that this term has different meanings w h e n applied
to people of different ages. Attachment is not a thing inside a
h u m a n being or a p e r m a n e n t change that is exactly like a duck's imprint-
ing, but a w h o l e pattern of behaviors, feelings, and ideas that change grad-
ually as the individual develops. T h e great importance of this pattern is that
it continues to influence our social relationships t h r o u g h o u t life.
T h e study of developmental changes in attachment gives us essential in-
formation that helps our understanding of groups (adopted children, for
example) and of individuals (such as children w h o seem unusually clingy
and concerned with their parents). R e n e Spitz, writing m o r e than half a
century ago, prefigured this statement w h e n he urged

... establishment of norms and regularities in the unfolding of the

infant's mental and emotional development. Such norms have to be
understood as representing broad generalizations of a statistical nature.
They provide chronological age zones within which the emergence of
certain behavior patterns, of certain emotional responses and of certain
qualities of emotion can be expected with regularity With the establish-
ment of such ... regularities we gain a basis for comparison between various
groups, which is indispensable for ... further scientific investigation.1

Since Spitz's day, m u c h information has b e e n collected about attach-

m e n t as it manifests at different ages. Researchers and clinicians have
56 Understanding Attachment

generally focused on age periods that correspond roughly to particular,

predictable attachment events. These rough age periods are (1) birth to
around six to eight months; (2) from about eight months to two and a half
years of age; (3) the preschool period, from about two and a half to five
years; (4) the elementary school years; (5) the adolescence and adult years;
(6) periods when adults are caring for young infants; and (7) periods in
adulthood when new attachment patterns emerge (falling in love, for ex-
ample, or becoming a caregiver to an elderly parent). This chapter will
describe attachment-related events in each of these age periods.

SOME ESSENTIAL ISSUES A B O U T

THE DEVELOPMENT OF ATTACHMENT
Some readers will notice that prenatal life has not been mentioned as a
time when the infant experiences attachment processes. As far as our cur-
rent evidence can show, unborn babies are not forming attachments, and
children are not born with a preference for their birth mothers. In fact, in
the course of evolution, it would have been a serious problem if babies had
been born attached; so many mothers died in childbirth that a great propor-
tion of babies would have been traumatized by separation. However, be-
liefs about prenatal emotional development were at one time put forward
in the scholarly world. 2 (Today, they remain a part of popular ideas about
attachment, as we will see in a later chapter.)
It seems that the development of attachment depends on a type of event
that can occur only after the baby is born: a transactionalprocess. Transactional
processes involve many interactions between individuals and cause devel-
opmental change in a complex way.
We tend to be used to thinking about simple events causing develop-
ment. (For example, a parent might do something to a child, like wipe her
chin.) We also are used to the idea that things might get a little more com-
plicated when a parent and child interact, when one does something to the
other and the other responds or reciprocates in some fashion. (Perhaps the
father wipes the toddler's chin, and the little girl playfully does the same
thing back to him.) A transactional process goes another step beyond the
interaction. A transactional process involves many interactions carried out
over a period of time, with many small changes in each partner, which oc-
cur with each event. The father may struggle only slightly to wipe a ten-
month-old's chin, but he probably does not receive much cooperation. By
the time she is twenty months old, the daughter may come when she is
called, hold up her face to be cleaned, participate by holding the cloth her-
self, or resist with play or anger. The way the father approaches her has
changed too. He would now be very surprised if she fussed in a babyish
 Attachment, Age, and Change 57

way as she used to do. Both people have changed very gradually as a result
of the many occasions when they have affected each other, and the child,
of course, has also changed as a result of maturation.
Attachment develops as a transactional process between the child and a
caregiver. Over time, caregivers change their behavior patterns and emo-
tional responses, just as children do. However, the research information we
have about attachment is primarily about the child's contribution to the
process. To look at both partners at once is exceedingly difficult. To what-
ever extent it is possible, this chapter will examine both sides of the trans-
actional process, though most often it will be necessary to concentrate on
one side or the other. As Donald Winnicott so famously declared, "There
is no such thing as a baby"—a baby alone, that is, and we need to remem-
ber this at all times.3

T H E EARLY M O N T H S
From a time soon after birth, human babies are responsive to human be-
ings. They imitate facial expressions before they are a day old, and they
watch faces and show special interest in people's eyes. It is clear they rec-
ognize human beings as something different from inanimate objects. Many
of a young infant's actions are attractive to adults, especially the social smile
that appears at six to eight weeks in response to another person. When we
adults see a baby respond to us like this, we become interested and gratified,
and we want more of that interaction. Such attractive baby responses may
be called engagement behaviors, because they help us become more and more
emotionally involved with the baby. Although the baby's engagement be-
haviors make us feel special, however, they do not actually mean that the
baby is concerned with us as individuals.

Early Fearlessness

During the first six months after birth, human beings do not show the
preference for familiar people or the fear of unfamiliarity most often associ-
ated with attachment. There is plenty of evidence that little babies can tell
the difference between familiar and unfamiliar adults, but little proof that
they actually prefer one to the other. (An adult who is familiar to the baby,
however, is also an adult to whom the baby is familiar, and it may be that
a person who knows the baby well can comfort or play with the baby more
effectively than a stranger can.)
When abruptly separated from familiar people, young babies show little
disturbance in sleep or eating. They do not seem frightened of strangers,
and this is not surprising because they are not frightened of the dark, loud
noises, big dogs, or anything else on the usual list of childhood fears. Small
58 Understanding Attachment

babies may well respond with distress to a new caregiver who is clumsy or
fails to understand the babies' signals, but this is not the same thing as being
afraid.

Dyadic Self-Regulation
The youngest babies are characteristically more concerned with their
internal processes than they are with the people around them. Infants of
a few months may be very responsive to our social overtures at times—
because of their own needs and interests and because we just happen to do
something that gets their attention—but their responses occur on their own
schedule. Most of the time, they ignore us, cry, or fall asleep. If they are
crying, they may need a great deal of help calming down, and they may not
reach a happy, quiet, alert state until they have again gone through a sleep
cycle.
One of a parent's major jobs during this period is soothing or comforting
the baby—a task as important in its own way as feeding and cleaning.
Skillful, prompt comforting helps a baby establish a first step in emotional
development, a step on which attachment probably depends.
This first step is called dyadic self-regulation, and it is the ability the baby
has to be soothed and to recover from distress with the help of a caregiver.
The baby begins to attain calm when in a dyad, or couple relationship, with
another person who is helping. The ability to calm down without this help
is not achieved until much later and is perhaps never totally mastered, even
in adulthood. Most adults still need and appreciate sympathetic comforting
when we are seriously upset.
It is critically important to understand that an infant of less than six
months does not have much ability to calm down alone from real distur-
bance, such as a painful experience. He or she may cry frantically and even-
tually fall into an exhausted sleep, but this is not the same as attaining the
quiet, alert state that permits play and learning. After three or four months,
however, most caregivers have figured out ways of offering effective com-
fort, such as holding the baby in a certain position, rocking with a particu-
lar rhythm, or offering a pacifier. The babies also have discovered some
understanding of things that make them feel better; they settle into posi-
tions of comfort fairly easily, even beginning to calm as soon as help is of-
fered. When they are older, they may imitate a caregiver's strategy by
singing to themselves or climbing into the rocking chair on their own.
Dyadic self-regulation is a major step toward attachment, because it helps
the baby experience much more quiet, alert time than it did before. These
quiet periods are a time for looking at and listening to people, and as they
get longer, new transactional processes take place. Not only is the baby
 Attachment, Age, and Change 59

quiet and interested in other people, caregivers, who are pleased that they
can make the baby happy, are more inclined than ever to be sociable and
communicative. From about the age of two months, the baby's smiles re-
ward the caregiver, and within a few more months, the baby's laughter
follows the caregiver's playful approach.
Caregivers who have learned good comforting tricks now begin a period
of courtship, working hard to get babies to show pleasure. They concen-
trate on social play. Even during feeding or diapering, the adult's energy
goes into keeping the baby attentive and happy.
In the next months, the baby's pleasure and interest become more and
more evident to sensitive caregivers, and the caregiver finds more jokes and
simple games that please the adult and child. Strangers can manage to make
the baby laugh, too, but nothing is really as good as the familiar play routine
developed by the play-partners through transactional processes.

EIGHT M O N T H S T O T W O YEARS
The toddler period is the stage of life most clearly associated with attach-
ment behavior and emotion. As children of this age become more capable of
moving and communicating, they reveal new concerns about separation.

Reorganization: Attachment Behavior, Attachment Emotion

By about the seventh month of the baby's life, a number of factors are in
place, preparing for the behavior and feeling changes that reveal the attach-
ment process. If the baby has had the opportunity to learn from interactions
with a few consistent caregivers, he or she has some ideas about how those
people act. The baby can read the adult's signals and anticipate what is go-
ing to happen: the mother's unbuttoning of her shirt is a preparation for
nursing; a certain way the father moves means being picked up for play.
The baby also comes to expect that his or her own signals will be read,
understood, and acted upon, most, if not all, of the time.
By about seven months, most babies become wary of strangers. They are
not actually frightened, but seem serious and suspicious at first, checking
out the stranger carefully. If the stranger is friendly, the seven-month-old
soon warms up and is ready to play. It is clear from the baby's behavior that
he or she knows the difference between familiar and unfamiliar people and
that the difference is important.
By about eight months, a new emotional reaction enters the picture, and
the baby's social world is transformed and reorganized. The new emotion
is fear, a response that we do not see in the younger child. Now—and
sometimes rather suddenly—the infant signals that he or she feels fright-
ened by loud noises, sudden movements, strange events, and, especially,
60 Understanding Attachment

strange people. Fearfulness becomes part of the developing internal work-

ing model of social relationships, as strangers become sources of fear, and
familiar people sources of comfort and security. The behaviors associated
with strangers and with familiar caregivers are our clues to the development
of the child's internal working model, and we may call them attachment
behaviors. The child's facial expressions and vocalizations tell us that attach-
ment behaviors are accompanied by attachment emotions of fear and discom-
fort or pleasure and security.
The eight-month-old characteristically begins to show stranger anxiety
(fearfulness around unfamiliar people) and separation anxiety (fear in situa-
tions where familiar people may or do leave). The Strange Situation, de-
scribed in the last chapter, uses both these situations in mild forms to trigger
a child's display of attachment behavior and emotion.
Avoiding strangers and avoiding separation from familiar people are the
behaviors most often considered to indicate that a child's internal working
model of relationships now involves emotional attachment to one or a few
caregivers. Many parents have vivid memories of this stage of child life—
the toddler who persistently hid his face when approached by friendly
strangers at the grocery store (or even by a grandmother who had not vis-
ited for a while) or the child who cried with fear when her mother jumped
up to answer the telephone. Some parents are unfortunately disturbed and
worried about these behaviors, recalling their friendly, outgoing baby of a
few months earlier. If the baby is in child care, outside the home, the par-
ents may begin to fear their child has been abused or harmed in some way
that has brought on this new behavior. In fact, moderate social fearfulness
at this age is an excellent indicator of good development; it shows that the
caregivers have been doing something right. As we will see in the next
chapter, though, all children are not equally expressive about their fears,
and some who are developing well will display only mild concerns about
strangers and separation.
Attachment behaviors and emotions also depend, in part, on situational
factors. The presence of a stranger or the threat of separation is less fright-
ening in the child's own home and more troublesome in a strange place.
How the stranger acts is a factor that helps guide the child's reaction; a star-
ing person with a serious or fearful face generates more distress than a smil-
ing stranger who is careful to make eye contact brief and flirtatious. If a
parent is present, his or her actions also prompt the child's response. Seeing
a stranger, the child looks at the parent's face in a pattern called social refer-
encing. If the parent looks happy, the child is less likely to act frightened; a
parent's expression of anxiety confirms and intensifies the child's fear.
In addition to fearfulness, other negative emotions are characteristic of
this period and give rise to the terrible twos. Toddlers seem torn by the
 Attachment, Age, and Change 61

conflict between attachment and the need for autonomy, and they express
their feelings through resistance and negativism. Temper tantrums, so typi-
cal of this period of development, are most often triggered in attachment-
related situations—a mother who is shopping refuses to pick up the toddler,
for example, and the child responds by throwing himself on the floor, and
thrashing and screaming. An examination of this situation shows us impor-
tant attachment issues at work: the child is in a strange place, possibly near
strange people, and he seeks to be near the mother in reaction to this expe-
rience. She, however, does not cooperate, and intent on her shopping,
does not even offer the eye contact and friendly facial expression that might
help the child tolerate this level of distress. The child's loss of emotional
control shows the intensity of his or her fear of separation in a threatening
situation.
From about eight months to two years or so, babies show a very impor-
tant form of attachment behavior, in addition to the ones just described.
This is a much more subtle behavior than stranger and separation avoid-
ance, and can be observed only in certain situations, so parents often do not
notice or remember it. Secure base behavior is an essential indication of
attachment, however.
Children show secure base behavior when in a strange place or situation
with a familiar caregiver. The child essentially uses the adult as a base from
which to make exploratory forays. He or she stays near the adult for a
while, and then moves away to explore the interesting, but worrisome new
environment. At first the baby moves only a small distance; then he comes
back to the adult, perhaps climbing into the lap briefly for what appears to
be an emotional refueling. Then another sortie is made, and another, until
the toddler has managed to explore the situation thoroughly. As the explo-
ration proceeds, the child may not even return all the way to the adult, but
may look back, make eye contact, and find the adult's happy, unfrightened
expression enough reassurance for the time being. The adult will often of-
fer emotional support over a distance, by talking to the child, and the child
may reply with pre verbal jargon or with actual words.
Secure base behavior in the presence of a familiar caregiver is much differ-
ent from the frightened, depressed stillness of a toddler left alone or with a
strange adult in a strange place. When we see the baby can use a person as a
secure base, we understand that person plays an important role in the child's
internal working model of social relationships. An adult who can serve as a
secure base is an adult to whom the child has formed an attachment.
Secure base behavior shows us how children's emotional lives support
their ability to learn from the world through exploration. Even when they
are not exploring, however, infants and toddlers show us there are connec-
tions between their emotional attachment to adults, and their interests in
62 Understanding Attachment

objects and in communication. Toward the end of the first year, babies work
hard to accomplish joint attention.4 They catch the eye of a familiar adult
and move their own gaze, signaling the adult to look at something interest-
ing; the two look back and forth at the object and then at each other, enjoy-
ing their shared experience just as two adults do. Interesting things seem
more approachable when a familiar adult is present; they are more fun too.
Children's displays of stranger anxiety and separation anxiety are often
distressing to adult caregivers, especially in modern family life, where ma-
ternal employment and non-maternal child care are frequent and econom-
ically necessary. The disadvantages of these childhood feelings and behaviors
are obvious, and their advantages are hard to see. If we try to see life from
the perspective of our remote ancestors, however, it may be more evident
that such attachment behaviors have a very positive side. In prehistoric
times, a wandering child might easily have been killed or stolen by a neigh-
boring band of adults, and those who feared the unfamiliar were more
likely to survive. Children who preferred to stay near adults would also
have had more chances to learn social rules by watching adult behavior. In
the ancient past (and possibly today as well), adults may have done a better
job of child care when they were aware that an unattended child might
begin to cry loudly, disturbing adults, and potentially attracting the atten-
tion of enemies or predators.
Secure base behavior does not cause the same obvious problems for
adults as stranger and separation anxiety may; its advantages may also be less
easy to see. But the ability to use a familiar person as a secure base makes it
possible for the child to learn and explore, and, at the same time, stay safe.
This is a more desirable outcome for human beings than safety alone. One
modern difficulty connected with secure base behavior involves the toddler's
need for familiar people, without whose presence even the best-designed
preschool education may be valueless. Young children who are cared for in
their own homes almost automatically have a secure base. Young children
in child care outside the home may have no familiar caregivers unless spe-
cial efforts are made to encourage good relationships with consistent child
care staff. The secure base issue makes excellent early childhood care a
highly labor-intensive enterprise.

TODDLERS, SEPARATION, A N D LOSS

The period from eight months to two years is thus characterized by spe-
cific attachment-related emotions and behaviors, and these need to be taken
into account as we try to understand all young children's experiences and
plan their care. In addition to these almost universal attachment-related con-
cerns, however, there are other situations that arise for some young children.
 Attachment, Age, and Change 63

These have to do with the child's short- and long-term reactions to abrupt,
lengthy separations from familiar people. These reactions, aspects of what
Rene Spitz called hospitalisnV are fortunately not part of every child's early
family life, though they do occur for far too many toddlers.
A toddler who has achieved attachment to one or a few caregivers will
respond with serious distress to a separation that lasts for more than a day or
two. Far from being too young to be affected by the loss of a parent, these
children are at their most vulnerable age. Their responses to loss are com-
pletely parallel to the grief and mourning response of a bereaved adult. (In
fact, the idea of stages of grief, so popular today, initially came out of work
with orphaned young children.) Playing, learning, and other developmen-
tal processes may be put on hold for a period of months, and physical health
too may be affected.
In the nature of things, of course, a young child may suffer an abrupt
separation lasting days or weeks for reasons that are not very serious: the
parents may have decided that their marriage needed a second honeymoon,
or a new baby may have been born. A reunion with the parents eventually
occurs, but children under two years do not understand what is happening
and cannot always accept it with grace. The child's behavior during the
separation shows distress, and when the parents return, they may be ig-
nored, greeted coolly, or even hit or pushed away by the angry toddler.
The child's angry or aloof behavior soon gives way to intense concern that
the parents will leave again, and there may be weeks or months of clingi-
ness, anxiety, and sleep problems before the child's worries resolve.
In some cases, of course, there has been a serious reason for the separa-
tion, and there will be no reunion. The child's distress is great and pro-
longed. Given an emotionally supportive caregiver, young children may
recover completely from a serious loss and are then able to form a new at-
tachment. In the course of recovery, however, the child will show deep
unhappiness, crying, eating and sleep disturbances, failure to play, and, of-
ten, a regression from recently achieved goals, such as toilet training. This
profound disturbance gradually gives way to a depressed withdrawal and a
lack of interest. If the lost person returns, the child shows no recognition.
Finally, a new attachment and a new interest in life can occur. (For some
children, such as the mutilated orphans of the Congo, there is no sympa-
thetic caregiver to offer a new life, and the experience of loss becomes the
central organizing feature of life.)

Adoption and Breastfeeding: Connections with Attachment?

In a later chapter, we will be examining in detail some popular beliefs
and misunderstandings about attachment. This may be a good point to
64 Understanding Attachment

touch briefly on two common ideas at odds with the material we are dis-
cussing in this section. First, we need to realize that the experience of adop-
tion does not necessarily cause problems of attachment, or of general
development. It is important to understand that children adopted early in
their first year proceed through the same developmental steps with their
adoptive parents as non-adopted children do with their biological parents.
Because attachment is based on social experiences that adoptive parents
usually provide, there is no reason that the early adoption of children should
have negative effects on attachment. (Later adoption, of course, may in-
volve a different set of social experiences, especially if it occurs after the
point when attachment is usually evident.)
A second common misconception has to do with breastfeeding. Breast-
feeding, or any other type of feeding, has in itself no known effect on the
child's internal working model of social relations (nor, as we will see later
in this chapter, is it known to influence the mother's feelings toward the
child). Ordinarily, of course, the person who feeds the baby most is also the
one most likely to play and interact socially in ways that connect to attach-
ment, playful communications being a real part of the feeding interac-
tion. Breastfeeding mothers usually have a bit more leisure or flexibility
of schedule, they are healthy themselves, and they have healthy babies.
These factors encourage both breastfeeding and attachment, whereas sick
mothers with sick babies and overwhelmingly demanding schedules are
more likely to have problems with nursing and attachment. We might see
better attachment in breastfed babies than in some bottle-fed babies, but the
feeding method would not be the direct cause of good or poor attachment
development.

T H E PRESCHOOL PERIOD
As we have seen, the social behavior of the four- or five-month-old
gives way to much different attachment-related behavior by the end of the
first year. It appears that this first stage of attachment forms the core of the
internal working model of social relations, which will continue to develop
for many years. The intense, anxious emotional concern the toddler shows
for the caregiver will change again and again, until in his own late adult-
hood, the grown-up, former toddler becomes the concerned caregiver for
elderly parents as well as for children or grandchildren. To prepare for this
series of changes, the individual has to stop expecting to be the one who is
nurtured. He or she has to become capable of negotiating or compromising
with respect to other people's needs and of developing close relationships
with new people. Without such developmental changes, the individual
would remain a big, self-centered baby.
 Attachment, Age, and Change 65

The preschool period features a child's first steps away from the emo-
tional egocentrism of the toddler. Preschool children in most cultures are
likely to experience more separations from their parents than babies do.
If they were breastfed, they have generally been weaned by this age.
Most are at least beginning to handle their own toilet needs. They walk
and talk well. Sometimes their own interests in play and in their peers
draw preschoolers away from their parents' sides. But the parents too seek
separation. A mother may have a new baby to care for, and her work at
home or away may require her to arrange that someone else care for her
preschooler.
Like older human beings, preschoolers do not mind an unpleasant event
like separation so much, if they can control when and how it occurs.
Preschool children may protest vigorously against separation from their
parents, but they also seek separation themselves, or accept it calmly after
some negotiation has occurred. For instance, the preschooler may agree to
be left at the day care center when she has received the number of kisses
equal to the time on the clock, when the coat and lunch have been put in
the cubby, or when the parent reads one story.
Separations and reunions are a fact of life for most preschool children,
but their effect on the internal working model depends on the child's ex-
perience of predictability and control. Ideally, the preschool child's internal
working model of social relations will involve a trust in compromise with
others, and will thus move the child past the belief that safety and security
exist only in the presence of a familiar caregiver. Experiences of caring ne-
gotiation are an important way to begin to trust compromise. Without
these developments, the later experience of school and independent play
with peers can only be frightening and difficult.
Parents have their own goals for their preschool children's development,
and these involve compromise too. Most parents want the child to develop
greater independence, but at the same time they want family rules to be
obeyed. Though a few parents want to exercise complete authority over
the child, most are committed to the development of a goal-corrected partner-
ship, a relationship in which both the parent's and the child's needs will be
considered, using mutual communication and planning as negotiating tools.
An important issue at this time is the parent's skill at communication; care-
ful communication is a major way to help the preschool child maintain a
sense of security when the parent is absent, and, as well, it is an essential
model for the child to follow.
Negotiation, bargaining, and compromise are thus aspects of the pre-
schooler's attachment relationships. They bring about a step toward matu-
rity in the internal working model. We can recognize attachment at work
at this stage of life when the child attempts to negotiate separation from
66 Understanding Attachment

familiar people. Parents or child care providers may recall negotiations of

this type, such as the lengthy proceedings at bedtime requiring more and
more extended rituals before the final good night. Preschoolers are often
said to be better or more obedient with less familiar people. Of course,
there is no need for them to negotiate separations with people of little emo-
tional importance to them.
Especially if they are sick, tired, or distressed, preschoolers can also show
some of the same attachment-related behaviors as toddlers do. The pre-
schooler still seeks a familiar person when frightened and has trouble feeling
secure when alone or with strangers in a strange place. Preschoolers go off
on their own sometimes, but like younger children they will quickly close
the distance from their parents if they feel any threat.
Abrupt, long-term separation from a familiar caregiver is still a blow to
the preschool child, but its effect is somewhat buffered by a developing part
of the internal working model: language and symbols that represent attach-
ment relations. A four-year-old whose parents have gone away for a month
can be helped to cope with verbal reminders that they will be back or even
by crossing days off on a calendar, whereas a twelve-month-old has no
equivalent way to symbolize events and reassure himself.
Preschool children in industrialized countries today are more likely to
find themselves separated from a familiar caregiver due to divorce than to
death. Family breakup is a particularly difficult type of separation to work
through with negotiation, because parents can rarely tolerate the compro-
mises the child wants: that Daddy or Mommy will live at their house half
the time, or that the divorce can happen, but no new partners may be
brought into the picture.
However, the most important issue for this age group may simply be that
some negotiation and joint planning take place, with the child's wishes and
feelings being considered, if not entirely gratified. Even if the child is only
offered the choice of a visit on Friday or one on Saturday, he or she needs
to have some input in the decision-making. It is also important for the
parents to affirm the child's choice by complying with the child's decision,
rather than changing plans unilaterally. Grief and sadness will always follow
serious losses in the preschooler's life, but the family's handling of the situ-
ation probably has a more powerful impact on development than does the
loss itself.

SCHOOL-AGE CHILDREN
By the elementary school years, children generally have a social circle
outside the immediate family. School, teachers, neighborhood friends,
clubs, and religious groups all form part of the child's internal working
 Attachment, Age, and Change 67

model of social relations. Children of this age may leave home for periods
of time, for a sleepover birthday party, a lengthy hospital treatment, or even
to go to boarding school. They may experience periods of loneliness and
homesickness, but on the whole, they are ready to include new situations
and relationships in their internal working models.
Although friendships with age-mates are an important aspect of attach-
ment for school-age children, parents remain at the core of the child's
emotional life. As was true of preschoolers, schoolchildren's behavior may
not reveal this fact unless there is a threat, or in cases of illness or injury,
when they desire the security of nearness to the parent. Talking on the
phone or even daydreaming about a reunion may be enough to help the
separated child—although we should note that having to use these expedi-
ents takes the child's energy away from work or play. The internal working
model of attachment still designates the parents as the ones who make you
feel safe—this is the case even when a parent has been unpleasant, unreli-
able, or overtly abusive.6
For most parents and children, the elementary school period is one of
lessened negative emotion. Both adults and children have improved nego-
tiating techniques, and parents are less likely than before to assume that the
child's wishes are just silly whims. Parents also have more confidence that
yielding to a child will solve a problem rather than create incessant de-
mands. Children's improved command of language enables them to com-
municate in situations where they once would have dissolved into tears and
had tantrums. The attachment relationship with the parent still involves
negative as well as positive emotion, but this is likely to be expressed with
sulking, dawdling, and forgetting rather than with direct anger.
Parent and school-age child are generally continuing to work toward a
goal-corrected partnership. In order to manage this, both must recognize
how the child's changing needs and abilities affect their relationship and
their ways of making decisions. If a boy of twelve says he wants music les-
sons and promises to practice, for example, a parent will generally assume
that he knows what he is talking about, even though the same promise made
at age six was not kept. The parent gradually encourages the child's initiative
and autonomy, thus moving toward a more mature, flexible relationship
between parent and child. As the parent systematically modulates the roles
each plays in the attachment relationship, he or she also helps the child build
flexibility into his internal working model of all social relationships.

ATTACHMENT, PEERS, AND POWER

Internal working models of social relations cannot be built on parent-
child relations alone. Parent-child relations involve considerable differences
68 Understanding Attachment

in power, whether it is in physical strength, knowledge, or financial re-

sources. This is much less true about many other relationships, especially as
the child's age advances. Whereas preschool children had large power dif-
ferences in many of their important relationships, school-age children have
more connections with siblings, friends, and classmates, in which power is
likely to be more equal than it is in the case of parent and child.
To have a complete, mature internal working model of social relations,
the individual needs to have experiences with a variety of power differ-
ences. Children's experiences with siblings usually involve some level of
power difference, whether based on differences in age, skills, or relation-
ships with parents. In some families or cultures, the child who belongs to
the more favored sex has extra power, in spite of youth or small size. Sibling
relationships often seem to share the formidable emotional power charac-
teristic of attachment to adults. The conflicts and sibling rivalry that emerge
from this emotion can be very provoking to parents, but with help, chil-
dren can use these situations to learn about negotiating and to extend their
internal working models.
School-age children's peer relationships outside the family should also
be considered an aspect of attachment. They provide an addition to the
internal working model of social relations, an addition that stresses the fine-
tuning of negotiation when power differences are very small. The simpler
internal working model of the early school years directs the child to get an
adult to solve disagreements with peers, but as the model becomes more
complex, children offer each other compromises, often deciding that one
will give up what he or she wants, rather than be separated from the friend.
Sometimes an adult can facilitate compromise simply by asking, "Well, do
you want to play together?"

T H E ADOLESCENT YEARS
For all families, the management of attachment concerns in adolescence
can have a critical impact on the teenager's later success in life. A fine bal-
ance between independence and emotional connection needs to be main-
tained, while the family serves as a secure base for the teenager's first
exploration of the real world.
Events in the teenager's modification of attachment to parents are in
many ways parallel to those of toddlers. This does not imply that adoles-
cents are babyish, or so incompetent that adults must manage their lives,
but like toddlers, they are in the process of finding ways to use their family
attachments as support for coping with new tasks. One of the tasks for
which they need family support is, paradoxically, emotional separation from
the parents. The aspects of this confusing situation include ambivalence,
 Attachment, Age, and Change 69

frustration, increased emotionality, and irrational responses to people and

events. Like toddlers, adolescents need responsive, supportive parents to
help them achieve emotional and social growth, but, again like toddlers,
they resist being guided against their wishes; to comply easily with parents'
desires brings on the fear of lost autonomy and developmental regression.
During the adolescent period, boys' and girls' attachment relationships to
their parents alter in somewhat different ways. Part of the difference is de-
termined by the timing of puberty (occurring on average two years earlier
for girls than for boys). The physical changes of puberty are a powerful
impetus for emotional separation, as they make it obvious that the old re-
lationships no longer work.
Families with a history of goal-corrected partnership have a relatively
easy time transmuting the old attachment relationships into models where
parent and teenager play nearly equal roles. Each may offer the other some
degree of secure base (adults need this too), but both have important goals
that are independent of their relationship. When this is the case, parents can
remain available as guides, consultants, and supporters throughout their
offspring's teen years and early adulthood, a period when good advice and
help can be most useful.
If a family has never thought a goal-corrected partnership desirable, it is
likely that the parents and the teenagers share a problematic internal work-
ing model of social relations—a model based on power, not on preference
for a familiar person. The family may have succeeded for many years in
managing children by threat or by force, but this strategy no longer works
when the children themselves become powerful. A power-based internal
working model does not allow for negotiation and compromise, but as-
sumes the complete capitulation of the weaker participant. Adolescence in
this power-oriented family involves a series of showdown struggles, even-
tually won by the teenager who then has no recourse but to leave home.
Premature separation of this type means that the adolescent must enter im-
mediately into adult responsibilities, without guidance or a secure base to
which he or she can return. Such teenagers are often correctly described as
"throwaway," rather than "runaway," children. In addition to the obvious
problems, the youth who has shared the power-based internal working
model will expect others to have the same model, and he or she will view
offers of negotiation as weakness rather than as a useful social strategy. His
or her capacity to develop a more mature, complex model of social rela-
tions will be limited, though not completely absent.
An important and complex developmental issue for adolescents and their
families is the intensification of sexuality at puberty. Although much early
sexual behavior is solitary, eventually sexuality needs to be incorporated
into the internal working model of social relations. Adult sexuality usually
70 Understanding Attachment

involves a social context and a relationship with a partner, even if only a

brief one. Like older people, adolescents fall in love, or at least feel that
another person is attractive. They experience sexual interest in connection
with other people, in reality or in fantasy. They do not simply experience
sexual desire as an event within the self. The desire and the object of desire
go together. Even masturbation involves the fantasy of a partner.
The process of bringing sexuality into the model, the choice of partners,
and the whole matter of gender identity are outside the scope of this book.
These events are related to attachment but involve many other complex
components as well. We can briefly compare falling in love to some early
steps in attachment, but although the degree of emotionality is similar,
there are some real differences. An obvious difference is in the erotic aspect
of adolescent or adult romantic love, something that is absent in the young
child's attachment to a familiar caregiver. Another difference is in timing;
falling in love or falling out of love can happen quite quickly, but attach-
ment usually takes some months to develop. It does not disappear sponta-
neously; instead it diminishes painfully and gradually, if there is a separation.
When threatened, a teenager or an adult does not necessarily seek the ro-
mantic love object as a child seeks the familiar caregiver. The internal
working model of social relationships probably does a great deal to help
shape ongoing romantic relationships, but this does not mean that the first
steps in romance are parallel to the first steps in attachment.

PARENTHOOD
Mothers love their children because of their natural instinct, don't they?
Or do they? The idea of maternal instinct is one way of explaining the
change in their model of social relationships that most women experience
after a child is born. Like many uses of the instinct idea, however, this ex-
planation becomes less satisfactory when it is examined carefully. Neither
mothers nor fathers know automatically how to care for a newborn baby.
Different cultures have different infant-care practices, a fact that suggests
infant care is not instinctive for humans. If it were, everyone would behave
in exactly the same way.
Because the word instinct does not seem to describe human parental
behavior very well, investigators in this field at one point turned to "bond-
ing" as a substitute. In an earlier chapter of this book, we referred to the
term bonding as a description of an emotional connection between human
beings, and we noted that this was a nontechnical use. Technically, the
word bonding is used only to describe the development of an adult's posi-
tive feelings toward a child. There is no question that the feelings we call
bonding can be extraordinarily powerful; they preoccupy and motivate
 Attachment, Age, and Change 71

parents in ways almost unimaginable to people who have not experienced

this part of life. Particularly in the early months of the baby's life, bonding
involves constant thinking of the baby, intense feelings of love, as well as
some of anger and hatred, and the motivation to care for the infant's needs
at all costs. Parents experiencing these feelings are literally more comfort-
able with their own hunger or pain than with the idea that the baby is
hungry or suffering. Yet, at the same time, most human parents feel occa-
sional flashes of anger and the temptation to attack the infant; some parents
neglect, abuse, or even kill their children. Obviously this process of bond-
ing is not a simple one. It requires explanation.
The awkward word "bonding" makes an already complex situation more
confusing than it might otherwise be. The word seems to imply somehow
that each member of a bonded pair is in some way equally attached to the
other. But the development of the adult's positive feelings (bonding) and
the child's (attachment) are on quite different schedules. The adult's sense
of connection to the baby may, and usually does, exist for months before
the baby shows any behavior that suggests attachment to the adult.
Eventually we see the child and the adult function as a pair; they watch
for and concern themselves with each other, and they each bring some
powerful feelings to their emotional connection. The period of intense
connection is short-lived, however, for the child begins to move toward
negotiating separation, and the adult may then have a new baby, a new
romantic interest, or any of the many emotional concerns with which adults
need to deal.
For a period of time in the 1970s, some authors wrote about bonding as
if it were an instinctively determined form of rapid learning, much like
imprinting. 7 Researchers working in this field thought that bonding was a
change in a parent's thoughts and emotions, most easily created shortly after
a child's birth, and triggered in part by the physical appearance and behav-
ior of the newborn baby. Early experiences with seeing and touching the
young infant were thought to cause a long-term emotional change in the
parent, much as following an adult duck causes a change in the duckling's
response to other ducks.
Because of the comparison with imprinting, some researchers investigat-
ing bonding thought that early contact between mother and newborn
might be a powerful factor influencing the mother's emotions. In modern
times, as more babies were born in hospitals rather than at home, mothers
became less and less likely to see and hold their babies right after birth. The
babies were customarily taken away, weighed, cleaned up, wrapped, and—
since they were often chilled by this time—taken off to the newborn nurs-
ery to be warmed. Hours often passed before mother and newborn were
reunited. (In hospitals whose policy required that the baby be checked by
72 Understanding Attachment

the staff pediatrician during his rounds, the delay could be as much as
twenty-four hours.) When the baby was finally brought to the mother, she
might feel timid about unwrapping her child and bothering the nurses; the
visit might be brief, and usually there was little privacy.
Researchers working on bonding were concerned that these hospital
policies might be interfering with bonding, or at least that the conditions
were less than ideal. They proposed that mothers and babies be given a
period of early contact within a short time after the birth. The father was
to be included, if possible, and he and the mother were to be given privacy
and encouragement to undress and examine their baby. When parents had
this experience, they were usually engrossed with the baby and thrilled
with the sense of doing something very important.
The evidence from the reports of the parents was not enough to prove
any real advantages from early contact. Did the parents' early contact create
a bond that helped them care for the baby, who in turn developed particu-
larly well? This was the kind of question that is essential but so difficult to
answer experimentally. How was it even possible to create a situation where
one group of parents had early contact, but another comparison group did
not? And what about ethical concerns? What if it should prove that early
contact actually did some harm to the family, rather than giving the antici-
pated benefit?
Fortunately, a situation allowing research did arise, as one hospital
changed its policy from delayed to early contact for healthy babies and
their mothers. A comparison was made between the development of the
delayed-contact babies, born in the last week of the old policy, and of the
first week's group of early-contact babies. The babies' development was
followed over several years, and significant advantages were reported for
the early-contact group. 8
The researchers proposed the following chain of events: (1) Early contact
brought mother and baby together when the mother was most ready to
bond, soon after giving birth. (2) The newborn was at that time more alert
and responsive than the baby would have been the next day, and its appear-
ance and behavior affected the mother emotionally, causing her to form a
bond to her particular baby. (3) Bonding was an internal emotional and
cognitive change that made the mother more aware of, and more con-
cerned with the baby's needs, so she became more sensitive, responsive,
and receptive to the baby's signals than she would otherwise have been.
(4) The mother's increased sensitivity helped her do a better job caring for
the baby. (5) As a result, the baby's condition was improved; bonding led
to good child health and development.
The idea that early contact could accomplish so much was eagerly
welcomed. Programs to encourage early contact were established, and
 Attachment, Age, and Change 73

hospital policies altered to allow this. The idea was generally accepted that
bonding—and therefore good development—depended on early contact.
(Even today, nurses sometimes speak about bonding when they are simply
helping mothers with early contact and encouraging mother-child interac-
tion.) Some parents were concerned and distressed about bonding, when
health problems delayed contact with their newborn.

A More Recent View

As the excitement about bonding died down, more unanswered ques-
tions surfaced. The mothers in the initial bonding study had been poor,
educationally deprived teenagers in unplanned pregnancies. Would affluent
older women respond the same way? It began to appear there might be
many factors at work in the determination of a mother's attitude and a
child's development. Of course, this was far from the first time social sci-
ence research had come to this conclusion.
We now know that it would be more useful to think of bonding as a
change in an internal working model rather than an instinctually deter-
mined process such as imprinting. This view does not diminish the power-
ful emotions and motivations that are part of bonding, any more than
attachment should be considered less emotional because of its function in
the internal working model of social relations.
It is probably correct to say that bonding, like other aspects of the inter-
nal working model, may be best understood as part of a dynamic action sys-
tem. In such a system, many factors work together to determine an
outcome—in this case, the parent's feelings and behavior and the child's
good or poor development. An important characteristic of a dynamic ac-
tion system is that a small change in one factor may bring about a large
change in the outcome, though even a large change in a different factor
may affect the outcome very little.
Bonding—both the parent's feelings and the parent's responsiveness to a
child—may be much easier to disturb than the child's attachment to the
parent. Attachment processes are robust and hard to shift, but changes in
the many factors that determine bonding are more likely to occur.
The baby itself supplies many of the factors that influence bonding. A
healthy, alert baby responds easily to parents and gives clear signals about
its needs, so the parents are able to feel satisfied that they are doing the
right things. A frail or premature baby is less responsive. Parents are deeply
gratified by the sense of communication with their baby. Serious birth
defects do not necessarily interfere with bonding, but those that affect the
eyes and facial expression make it harder for the parent to have that sense
of communication.
74 Understanding Attachment

Factors in the parent also influence his or her emotional response to the
baby. Fear and grief seem to interfere seriously with bonding. A parent who
is mourning the death, during the previous year, of a loved one may have
difficulty feeling an emotional connection to a newborn. (This applies to
grief about an earlier miscarriage or stillbirth, too, and the death of one twin
is especially problematic.) Anxiety about a baby's health seems to trigger a
sort of premature mourning that interferes with bonding. Maternal depres-
sion lasting for more than a few months causes problems with the infant's
emotional development, putting the relationship more and more at risk.
Situational factors can also distort bonding. A difficult, unwanted, or
frightening pregnancy leaves the mother less prepared than she should be
for beginning a new relationship. If a mother can see her sick newborn only
through a nursery window, she may feel that the child is not her own.
Most new parents bring to their situation some factors that could inter-
fere with bonding. Many do not experience their first contact with the
baby as a thrilling moment of falling in love. But days or weeks of interac-
tion with the baby cause an emotional reorganization which changes the
internal working model, and results in powerful positive emotions and a
deep preoccupation with the child. Although early parenthood can involve
fury and frustration, most parental feelings are strongly positive, enabling
them to provide enough sensitive, responsive care to support the child's
good development.
Do the hormonal changes of childbirth really cause bonding, or at least
make it more likely to occur? Probably hormones play only a small role, if
any. Fathers and adoptive mothers, who have no dramatic hormone
changes, show bonding to the same degree as biological mothers. We need
to remember that the hormonal changes of pregnancy and childbirth are
also accompanied by important experiences that can occur in the absence
of pregnancy. A biological mother has normally known of her pregnancy
for six months or more and has had plenty of time to focus her thoughts
and feelings on the person she expects her baby to be. This experience—
which is shared by fathers and adoptive mothers—may do as much or more
to prepare her for bonding as biological events do. A pregnant woman also
experiences much social guidance, as other people tell her of the ups and
downs of birth and motherhood, treating her as a prospective mother and
helping her prepare to be one. Fathers may receive this type of guidance
mainly from their wives, their own parents, and intimate friends. Adoptive
mothers may receive it through the adoption agency, but they do receive
it just as pregnant women do.
A persistent myth about bonding is the belief that breastfeeding causes it
to occur. (As we saw earlier, it is also common to assume, and equally inac-
curate, that breastfeeding causes attachment.) People who have this belief
 Attachment, Age, and Change 75

often offer an unlikely picture of a mother gazing into a very young baby's
eyes as the child nurses. This very old idea certainly goes back at least to
Shakespeare's time as we see from Lady Macbeth's line, "I have given suck,
and know how tender 'tis to love the baby that milks me." In the early
twentieth century, just as Freud was suggesting the gratification of needs
was a source of attachment, one writer proposed that lactation is the cause
of maternal love.9
Of course, it is not practical or ethical to try to test the role of breastfeed-
ing in an experimental fashion. In order to do so we would have to make
sure that certain mothers breastfed and that others did not, a most difficult
task, as well as one involving potential dangers for some children. What we
know about the effects of breastfeeding comes from our observations of
self-selected mothers who have chosen breast or bottle-feeding on their
own, or at least without any researcher's advice. However, observations of
breastfeeding and bottle-feeding mothers give us no reason to think that
one group loves their babies any less than the other. Adoptive mothers,
who rarely breastfeed in this country, ordinarily experience bonding, and
their children have excellent developmental outcomes if they have been
adopted early in their lives.10 If breastfeeding were necessary for bonding,
the extreme popularity of bottle-feeding in the 1940s and 1950s should
have created catastrophic developmental problems, and this did not occur.
In addition, the necessity of breastfeeding would mean that no father could
ever bond to his child.
It may well be that mothers who have bonding problems may also be less
likely to breastfeed. It is also the case that a sick newborn, especially one
with facial birth defects, may be too weak to breastfeed, and bonding prob-
lems may occur. However, what we are seeing in these cases is the effect of
other factors on both bonding and feeding, not of feeding on bonding, or
vice versa.

T H E STOCKHOLM SYNDROME
A rare and little-understood form of adult attachment is sometimes called
the Stockholm Syndrome after the city in which it was first clearly de-
scribed. This syndrome occurs in situations where an individual, usually a
woman, is threatened or held hostage in a case of genuine danger. The
hostage-taker negotiates for the woman's safety with the authorities, using
the captive's life as a bargaining chip. Rather than hating and fearing the
captor, as might seem logical, the hostage falls in love and does not want to
be rescued or separated from the captor.
The situation occurs so infrequently that it has hardly been studied in de-
tail, but the characteristic emotional changes seem to have much in common
76 Understanding Attachment

with normal forms of early attachment. Studies of concentration camp in-

mates during World War II describe some individuals as admiring and imi-
tating the guards in ways that seem parallel to both early attachment and the
Stockholm Syndrome. Sigmund Freud also referred to a similar process of
identification with the aggressor in adults. It is not known whether the
Stockholm Syndrome actually involves the same emotional processes as nor-
mal attachment, or whether this is a pathological process made possible by
some vulnerability in the hostage.

REVIEW A N D SUMMARY
Attachment and bonding both have to do with attitudes and beliefs about
specific people. Neither could develop in a social vacuum, for each requires
the participation of at least two people. We focus on the feelings of the
child when studying attachment and those of the adult when studying
bonding, but it would be ideal if we could look simultaneously at the steps
of both partners in their interpersonal dances. Perhaps we can manage a
little of this if we simply examine which events are happening at which
given time.
Shortly after a baby's birth, the parents are interested in and attracted to
the baby, and they respond happily if they can feed their child or get the
baby to look at them. The baby cannot do a great deal, but does sometimes
look at faces and eyes, and may even imitate facial expressions, giving the
parents a sense of gratification. As parents bond, they become more preoc-
cupied with the baby, and more sensitive and responsive to signals. This
helps to comfort and soothe the baby, who becomes gradually a little more
able to regulate his own emotions, enabling him to spend more time quiet,
alert, and responsive to adults. The parents are aware of this change and are
encouraged to try more and more playful communications. The parents
now strongly prefer their baby to other babies and know how to play with
and soothe her, but the baby does not yet particularly prefer the parents.
By about eight months, the child becomes capable of fearfulness and
begins to stay close to the parents, especially if there is any threatening or
unfamiliar event going on. Parents respond to this new behavior with
mixed emotions and often with some anxiety. They change their behavior
toward the child in anticipation of the fear and distress the child may ex-
press. Some parents are supportive and comforting, others punitive about
the beginning of attachment behavior.
During the later toddler and preschool period, child and parent begin to
move apart emotionally and socially. The child can tolerate more separa-
tion, but wants to negotiate it in order to keep some sense of control.
Parents at this point may have reason to encourage separation; a new baby
 Attachment, Age, and Change 77

may be on the way, or parents may be turning toward work, school, or

other adult activities. The child's negative emotions may make the parents
ambivalent, both longing to be with the child at times and repelled by tan-
trums and irritability.
School-age children and their parents ideally have arrived at a goal-
corrected partnership. While parents retain the greatest power and au-
thority, they can also respect their children's maturity and ability to make
decisions. For both parents and children, the relationship may seem casual
and low-key—until there is a threat to one person, which brings the other
in haste.
The adolescent years again involve a focus on separation, but this time
it is the parent who may be reluctant and want to negotiate, while the
teenager wants to move toward independence. However, the teen will
do better with emotional support from parents than she will if abrupt sepa-
ration occurs. In the teenage years, as well as later, close emotional connec-
tions to friends form an essential part of the individual's internal working
model of social relationships.
This page intentionally left blank
 chapter 5

Families, Experiences,
and Outcomes: What
Difference Does
Attachment History
Make?

A
startling news story in early 2004 revealed a strange family history
and raised questions about the effects of experiences on children's
emotional lives. At a children's birthday party, a young mother
named Luz Cuevas encountered a former friend with a six-year-old girl
who was identified as the friend's daughter. This was not particularly un-
usual, but Ms. Cuevas immediately wondered whether this child might pos-
sibly be her own daughter, Delimar, who had disappeared as a week-old
infant, apparently having been consumed in a blazing house fire. Unnoticed,
Ms. Cuevas obtained a clipping of the child's hair and managed to get a DNA
analysis. The result? It was Ms. Cuevas' own child, who must have been
kidnapped as a tiny baby by the woman claiming to be her mother, the fire
having been set deliberately to cover the kidnapping. When these facts
were discovered, Delimar was taken by the state child protective service
agency and eventually returned to Ms. Cuevas.
Delimar, thus, experienced two abrupt separations: the first at one week
old and the second at six years of age. What would we expect to be the
results of these experiences? Chapter 4 has already answered this question
with respect to the earlier separation. As devastating as the event was to her
mother, the infant Delimar would have had no difficulty with the change
of caregiver at one week old, as long as she was treated well.
What about the separation at age six though? This question opens up
many issues. Are there short-term or long-term consequences of relation-
ships and of losses at different ages? How do early experiences influence the
80 Understanding Attachment

individual as he or she matures and develops? Does the individual's personal-

ity show the influence of attachment experiences from any time in the past?

WHAT, W H E N , A N D WHY: ASKING

QUESTIONS A B O U T A T T A C H M E N T A N D PERSONALITY
The questions in the previous section could be answered in a wide vari-
ety of ways, for they have more than one implication. To do that, we need
to define what we mean by "influence" more carefully.

What
In asking about the influence of attachment events, we might be inquir-
ing about serious psychopathology. In Bowlby's early study of the forty-
four boys,1 attachment was thought to be a major factor determining
antisocial behavior. (Bowlby's later work did not stress this idea as much as
his earlier writings.) Some writers have speculated on whether attachment
problems might be factors in autism or even adult serial murder. This ques-
tion is a complex one and will be addressed in Chapter 6. At this point, we
can say that attachment alone is probably not the cause of major emotional
disturbances of the types mentioned above.
For our present purposes, the influence of attachment experiences may
be best defined as the extent to which attachment events shape an individ-
ual's emotional and behavioral reactions to other people—the unique char-
acteristics that we refer to as personality. Such influences could change from
one stage of life to another, but for our purposes, they would only include
feelings and behaviors within a normal range. The focus is on individuality—
normal personality differences—and on the minor emotional disturbances
that may be troublesome but do not interfere severely with education,
work, marriage, and parenthood. Because attachment involves an internal
working model of social relationships, we tend to look for the effect of at-
tachment experiences on relationships, rather than on, say, memory or ath-
letic prowess.

When

If attachment experiences have long-term effects, they could crop up at

any time during life, and they might not be equally evident at all times.
There are many points in life when these effects might occur, and we must
search throughout the life cycle to find evidence for and against long-term
effects.
There is another kind of "when" to deal with, too: if attachment expe-
riences have a long-term influence, at what point in life might the most
 Families, Experiences, and Outcomes 81

important attachment experiences occur? We need to handle this question

carefully and open-mindedly and realize that more than one answer is
possible.
Most of us assume experiences in early childhood have powerful effects
on personality development. Freudian theory and its derivatives (including
Bowlby's attachment theory) have impressed this idea on us and stressed
infant determinism as an almost unquestioned principle of development.
Though we do not have very good evidence that this principle is correct,
we just assume it is. Developmental psychologist Jerome Kagan suggests
that infant determinism is one of the seductive ideas about which we should
remain cautious—ideas that seem self-evident but that may be wrong. 2
It is not necessarily the case that the attachment experiences of early
childhood are paramount in an entire lifetime of social relationships. We
will find, though, that most existing research focuses on early childhood.
There is not much work that looks at later experiences as essential, but this
does not mean they could not be as important as early events.

Why
What leads us to ask this kind of question in the first place? Why should
attachment events have a real impact, when some other early experiences
do not? Part of the answer involves the enormous significance of emotional
relationships and social behavior in human lives. From childhood and be-
yond, our success and happiness are linked to emotional connections with
others. The preschooler who fights other children and resists his teacher will
be punished, deprived, removed from the group, and prevented from hav-
ing many satisfying experiences. The schoolchild who is sullen or quarrel-
some is not chosen to go on a special trip or be in the school play. Although
some adults may achieve professional success through intellectual abilities,
most of us need to use good social skills and cultivate personal relationships
to be successful at work and home.
Fortunately, we can pursue the question of attachment and personality
because we have the tools to do so. Chapter 3 described techniques for as-
sessment of attachment in children and in older people. No one would
claim that our tests of attachment are perfect, but they have been used for
many years by large numbers of researchers, both to collect new informa-
tion and to make comparisons with much older work.
Tests like the Strange Situation have shown us the continuity or stability
of the measurements we make of attachment behavior. This is an essential
factor in our understanding of the links between early attachment experi-
ences and any later impact on personality. To a considerable extent, we can
expect that a twelve-month-old, categorized as securely attached using the
82 Understanding Attachment

Strange Situation, will show related, similar characteristics years later; an-
other child, classified as insecure-avoidant will develop along different, but
equally predictable lines.3 When they are years older, the children will not
behave in just the same ways they did as toddlers, of course, because all
their ways of acting will have matured; however, the differences that once
existed between the one-year-olds will resemble the differences that now
exist between the five-year-olds.
This continuity or stability of attachment behaviors tells us that some
characteristics were already established at an early age; this means it is pos-
sible that the characteristics were actually caused by early attachment expe-
riences. Another fact that suggests that personality may be shaped by early
attachment experiences is that some children show different attachment
behaviors toward their mothers and than they do their fathers.4 These differ-
ences imply that attachment is not altogether in the child, but is instead in
the relationship, and thus arises, at least in part, from early experiences.

C O N F O U N D E D VARIABLES:
CONFUSED A N D C O N F U S I N G D A T A
A word of warning is in order before we discuss attachment experiences
and personality development. It is extremely difficult to sort out causes and
effects in this matter. There are some individual personality characteristics that
clearly do not come out of attachment experiences. These traits, called tem-
peramental or constitutional differences, apparently result from genetic or
other biological factors. It may be all too easy to confuse these with the child's
actual attachment emotions, behavior, or even, memories and thoughts.
Take, for example, a temperamental quality called "intensity of reac-
tion." Infants and toddlers show individual differences in their emotional
expressiveness. Some cry loudly and vigorously when unhappy, yet are
equally vehement in their expressions of pleasure, laughing and shouting
when tickled or chased. On the other hand, young children who are more
mild in their reactions may frown or look sad instead of crying, and smile
or chuckle instead of shrieking when amused.
These temperamental differences affect attachment in two ways. First,
adults may notice a child's concerns about separation when they are so
loudly and vociferously expressed, but may pay less attention to the milder
signals from the less intense child. Second, having noticed the child's unique
behaviors, the adults may provide different types of experiences to a child
who was already different. Temperament can cause children to have differ-
ent kinds of social experiences.
A parent of an intense toddler may plan to handle separations carefully in
public so unfavorable attention is not attracted by the child's cries. Some
 Families, Experiences, and Outcomes 83

parents may try to sneak out, when leaving the child with a sitter or at a day
care center, in an attempt to avoid hearing the child's loud distress. Others
may decide that care outside the home is impossible for their child, who
seems so terribly upset when left. Still others may punish the crying child
for disobedience or willfulness.
Parents of milder toddlers may feel relieved and proud of their child's
independence when little distress about separation is displayed. They may
go ahead with day care arrangements and concern themselves very little
about the moment of parting. On the other hand, some parents of mildly
reacting toddlers may be disturbed by their child's relative calm, and they
may fear that he or she has not formed an attachment.
The important point here is that children's different emotional expres-
siveness—their mild or intense reaction to separation—can cause the par-
ents to handle separations differently. The child's attachment experiences
are directly affected by the parent's actions, and indirectly affected by the
child's own temperament. Which contributes most to the child's develop-
ing personality characteristics? It is extremely difficult to tell.
A final source of confusion is connected with the question of the impor-
tance of early experience. Is the child's individuality shaped primarily by
early experience, or is it gradually molded over years of growing up? To
answer this, we would have to look at children whose infant and toddler
years were spent in one sort of emotional situation, and who then went to
quite different households. There are, of course, some children who are
adopted by new families or who live in foster families for years; with any
luck, these children go from poor emotional situations to good ones. We
do not see many who have reversed this order, so we do not have very
good material for examining the two potential formative periods. And most
children in industrialized countries live in the same families from birth
through adolescence. For the average child, chances are that the basic en-
vironment of infancy and of later years will be the same. The parents who
encouraged attachment in infancy with sensitivity and responsiveness will
continue to provide appropriately sensitive, responsive care for their older
children. If we look at the average ten-year-old, then, we have no way of
knowing whether her personality was completely determined at age two,
or whether the ensuing years did the real work of making her who she is.

SHAPING A T T A C H M E N T A N D
I N D I V I D U A L I T Y : INFANTS A N D TODDLERS
Whether we are considering early experiences or early displays of attach-
ment emotions and behaviors, much of what we know about attachment
focuses on early life. The first two years of life have received the most
84 Understanding Attachment

attention from researchers interested in the outcomes of individuals' attach-

ment experiences.

Mothers' Attitudes and Experiences

For most babies, their mothers are the major sources of social and emo-
tional experiences. The mothers' attitudes and expectations about their ba-
bies might well cause differences in their babies' emotional development.
As we discussed in an earlier chapter, the 1970s and 1980s produced a
great deal of work on mothers bonding with their babies. This research,
though difficult at best, was especially complicated because no one was able
to offer a good independent assessment of bonding. Many parent behaviors
and emotions seemed related to this concept. For instance, engrossment, the
adult's profound attentiveness to the baby, might be a step in the bonding
process and a way to measure emotional change. However, there was no
clear-cut way to say that one mother was bonded to her baby, and another
was not. It was even less possible to determine that one woman had bonded
a little, as opposed to someone else who had bonded a lot. Human behav-
iors toward young infants are enormously influenced by cultural expecta-
tions and by situational factors, such as a lack of privacy in the hospital. A
new mother may act very differently when in a room alone with her baby
than when she has a roommate whose mother, mother-in-law, and cousins
are all visiting and chatting.
Although the idea of bonding has not been abandoned completely, re-
searchers have now started to look at other, related aspects of mothers' at-
titudes. Maternal depression is one of the most important factors determining
a mother's ability to show sensitivity and responsiveness to her child and
thus to work toward secure attachment. Depressed mothers lack the energy
and attentiveness needed for sensitive detection of a baby's signals, and they
may feel too tired, preoccupied, or irritable to respond, even if they do get
the signal. Drug and alcohol use have similar effects on maternal behavior.
Grief over personal losses also has an impact on parents that interferes
with responsiveness to infants and thus potentially with attachment. The
recent death of an older baby or loved one plays a significant role here.
Early work on bonding stressed this problem, emphasizing the difficulty of
forming a new emotional connection to a baby while simultaneously hav-
ing to let go of another connection through mourning.
Grief may result from a loss of personal identity or a change in expectations
about the world, as well as from a death or other separation. For example,
couples that have had children through Assisted Reproductive Technology
(ART), after years of infertility, are sometimes happy. But they have also
been described as "unexpectedly... depressed, exhausted, and ambivalent."^
 Families, Experiences, and Outcomes 85

T h e y may fear losing the child and they may try to avoid conflict with that
child, a difficult task with respect to the negotiation of separation w e dis-
cussed in an earlier chapter.
W h e n grief and fear are overwhelming, mothers' attitudes, feelings, and
behavior are powerfully affected. A fascinating and disturbing account of
this is given in Death Without Weeping.6 N a n c y Scheper-Hughes describes
the dire poverty in Brazil and its impact on mothers' caregiving. H u n g r y
and sick themselves, the p o o r Brazilian mothers suffer many emotional and
material losses. After giving birth, they may identify a child as one w h o does
not want to live. T h e y fail to feed or care for the baby, w h o eventually dies
of malnutrition and disease. This decision w o u l d certainly mean these m o t h -
ers have failed to bond, if w e could define such a category at all. H o w e v e r ,
w h e n a baby survives and becomes stronger, the mother's interest increases.
Such mothers then describe the children as attached. O n e interviewee said,

The little one never lets me out of her sight. [She] hangs onto my skirts,
and she cries whenever I leave the house. I can't take a step without
her. What kind of children are these, so afraid to move without their
mama? So I yell at them to toughen them up. 7

This m o t h e r seems to dislike the child's signals of attachment and prefer-

ence, but it may be that she is concerned about the child b e c o m i n g i n d e -
pendent e n o u g h to survive in their very harsh environment. A mother's
o w n attachment status seems to be of major importance in determining the
o u t c o m e for her child. Mothers whose attachments are secure, as assessed
by the Adult Attachment Interview (discussed in an earlier chapter), are far
m o r e likely to have securely attached children. 8

Mothers' Behavior and Children's Attachment Status

M o t h e r s ' feelings and attitudes help to determine their behavior toward
their children, b u t it is the behavior itself that influences the child's devel-
o p m e n t of attachment. W h a t kinds of behavior are relevant here? W h a t
events shape early development so that toddlers show individual differences
in their attachment behavior and emotions?
Generally speaking, small children are strongly inclined to b e c o m e se-
curely attached to their caregivers, and most do so, despite parental ill-
nesses, marital disagreements, and less than ideal day care arrangements.
T h e y seem to have had e n o u g h sensitive, responsive adult interaction to be
secure, even t h o u g h life has not all been absolutely perfect.
Smaller numbers of toddlers behave in ways that place t h e m in the A and
C categories discussed earlier in this book, with one or another type of
86 Understanding Attachment

insecure attachment relations. They may avoid the mother when she returns
from an absence, or may go to her but not be able to be comforted by her.
Most studies of mothers' behavior and children's attachment status have
focused on the small number of toddlers whose behavior is in the D
category—disorganized or disoriented reactions to the mother's return
from a brief absence. These children have often experienced many changes
in caregiver or inconsistent relationships with their parents. Parents who
are depressed or who habitually use drugs and alcohol are particularly likely
to create inconsistent, unpredictable experiences for their children. Parents
who themselves experienced early losses, or were physically or sexually
abused, may also lack sensitivity or responsiveness to their children.
In addition to the past experiences that may cause D-category behaviors,
it is possible that a parent's behavior at the time of the reunion serves as a
direct trigger for disorganized/disoriented-child responses. When children
respond with fear or freezing on a parent's return, their actions may be
cued by the parent's own facial expressions. The parent may appear fright-
ened and thus be frightening to the child, and may also show sudden, un-
predictable behavioral changes, like abruptly moving a hand across the child's
face.9 A "natural" source of comfort and security for a toddler, the parent,
also acts in this situation as a source of fear. The child looks at the parent's
face for information about the parent's emotional state and finds there is
indeed a reason to be frightened. Freezing or collapsing may be a logical
response to a puzzle where no solution is really right.

Many Caregivers: The Effects of Change on Attachment

Human births most commonly involve a single child, and the simple ar-
rangement of one child—one mother is what most of us have in mind
when we think about attachment. For practical reasons, however, most
babies have more than one caregiver, and this has been the case throughout
human history. In traditional societies, grandmothers, neighbors, and child
nurses took turns at caring for the baby, giving mothers time for other
work. In modern industrialized groups, formal day care arrangements are
common. The experiences of some babies are additionally complicated by
a permanent change of caregiver— the mother may die, or law or custom
may take the baby, following neglect or abuse.
Few babies are cared for by a single adult at all times, and it is possible
that having one adult caregiver is not an ideal arrangement when it does
occur. Generally, two or three adults are greatly involved with a given
baby, and secure attachment can and does readily develop in this situa-
tion. But what happens to attachment when the circumstances are quite
different?
 Families, Experiences, and Outcomes 87

Boarder Babies. The boarder baby is an unfortunate modern phenomenon: an

infant born in a hospital, abandoned by its mother, and prevented, by legal or
practical considerations, from going to a family member or to a foster home.
Boarder babies are cared for in the hospital for many months, fed, rocked,
and talked to by an ever-changing roster of nurses and volunteers. Adults
who pass by are attracted by the infant; they pop in to say hello, then quickly
leave to continue their work—the baby learns a mirror-image form of social
response, smiling quickly, but failing to sustain the interaction. The boarder
baby's behavior is quite different from that of a family child, who from an
early age learns to sustain a conversation and to communicate in ways that are
probably the foundation for attachment. Whether the boarder baby's oddities
are entirely due to experience with too many caregivers is hard to say though,
because the child's situation is often connected with the mother's alcohol or
drug use, which might have also affected development.
Foster Care. Change of caregiver is a major feature of foster care for in-
fants and toddlers, with many babies in foster placement experiencing fre-
quent reassignments. Foster care may be planned as a temporary measure
while parents find appropriate housing or solve medical problems, and in-
fants may go back to their biological parents when difficulties have been
solved. However, it is not uncommon for the same or new problems to
arise. Then the baby is likely to go to a new foster (or "resource") family,
perhaps because the old one has taken another child. Visits with biological
parents may be encouraged, but they are usually infrequent, awkward, and
filled with anxiety, especially for parents who know they are being watched
for evidence of bonding or attachment. Transitions may be handled very
poorly, as in the case of a one-year-old boy who was taken from a now
familiar foster home while he was asleep, only to wake up at the house of
his prospective adoptive parents, whom he had never seen before.
It is common for children who have had many early foster care changes
to be impulsive and aggressive in personality, and it is easy to blame this
outcome on the lack of opportunity for attachment. However, in the inter-
ests of genuine understanding, we should remain aware of potential diffi-
culties that might have existed before the child ever went to foster care.
Behavior problems and other personality differences can be based on ge-
netic factors, or on health problems and drug or alcohol exposure before
birth. It is also true that some foster homes unfortunately model aggressive
behavior, either in the form of maltreatment by foster parents or the ex-
amples of older children placed in the home.
Adoption. Children may be adopted in a variety of circumstances—
straight from the delivery room, after many foster placements, after trau-
matic experiences, or in later childhood, when we would expect attachment
already to have occurred. Studies of adopted children seem to show little or
88 Understanding Attachment

no unusual emotional development when adoption takes place in the first

few months, though there are some differences in the types of problems
shown by early adopted and non-adopted children. Children who are ad-
opted between eight to twelve months and three years of age are more
likely to show general developmental problems in language, attention, and
cognitive ability, as well as trouble dealing with frustration. Children ad-
opted at age five to six, or older, after having lived in a consistent, well-
functioning family up to that point, may be sad or depressed. Though they
may continue to mourn their loss for years, they otherwise do not show
serious developmental problems. (Delimar Cuevas, whose case introduced
this chapter, would be an example of this.) The most important factor for
these children is not the fact of adoption, but whether they received con-
sistent, sensitive, responsive care during their infant and toddler years.
Separations and Losses. Babies who are less than six months old generally
show little real distress about separation, as long as they are well cared for by
a responsive caregiver. It is only when clear attachment behaviors, like stay-
ing near, occur that separation becomes a painful matter. Brief separations
from familiar caregivers result in protest, often followed by calming, and
then a renewed brief complaint, as the familiar person returns. Protest is less
intense if the child is in a familiar place (his or her own home, ideally) and
if the substitute caregiver is somewhat familiar. Divorce and separation from
one parent is responded to like any other separation from a familiar person.
Long-term separations are extremely distressing to the older infant or tod-
dler, especially if they occur abruptly or involve being in an unfamiliar place
with unfamiliar adults. A film made by one of Bowlby's colleagues, James
Robertson (Nine Days in a Residential Nursery), shows a toddler's loss of emo-
tional control, difficulty eating and sleeping, and development of physical ill-
ness.10 Continuing separation triggers a grief reaction that eventually enables
the child to forget the lost person, perhaps over a period of a year. The extent
to which a severe separation experience like this has long-term effects (such as
a tendency toward depression) probably depends on incidental factors, such as
the availability of sympathetic, comforting adults or the possibility of entering
a stable situation with a supportive new family. Infants and toddlers benefit
greatly when adults can serve as buffers against the experience of loss.
Infant and Toddler Day Care. When young children are cared for in groups
outside their homes, their social experiences are rather different than they
might be within the family setting. How does group child care affect infants
and toddlers? Do they lose their attachments to mother and father? Do they
have to try to function without a secure base? Is their emotional or com-
municative development harmed?
These questions were asked repeatedly during the so-called "day care
wars" of the 1990s, when psychologists struggled to understand whether
 Families, Experiences, and Outcomes 89

non-parental care was harmful or helpful. Insecure attachment was reported

as being associated with babies' longer hours in day care, and insecure tod-
dlers were described as playing less and crying more. 11 Other researchers
sought explanations with a more positive message about day care.
There is no question that infant and toddler day care is here to stay. Few
families can manage financially without some form of non-maternal care.
But not all day care centers are alike; it is possible to find attachment-
friendly child care. Paradoxical as it may seem, day care centers can help
toddlers develop and maintain attachments to their parents, both through
support to the parents and through encouragement of children's attach-
ments to child care providers. Children generally have several caregivers,
and good attachments to all of them are possible. Attachment to a day care
provider does not remove or diminish attachment to the parents.
To encourage the formation of secure attachments, day care centers
need to use assigned caregivers, each staff member having a small number
of children for whom she is the consistent and familiar caregiver. Because
children are apt to be at the day care center for periods of time that over-
lap shifts, a second group of providers needs to be assigned in the same
way. Attachment-friendly day care also reduces turnover in staff, allow-
ing children to go for many months without having to adjust to a new
face (staff salaries are usually at the heart of the turnover problem).
Attachment-friendly day care centers can also train their staff to help chil-
dren transition at the beginning and the end of the day. They can be
supportive of parents who have trouble saying goodbye and encourage
children to use "blankies" or a favorite stuffed animal to comfort them-
selves. Day care centers that encourage attachment in these ways are
likely to be among those accredited by the National Association for the
Education of Young Children.
One additional point about attachment and day care has to do with the
child's developmental status. Infants and toddlers who are in a period of
intense concern about separation are not likely to adapt very readily to a
new care arrangement. However attachment-friendly a day care center
may be, it cannot easily overcome the anxiety about strangers and separa-
tion that are so characteristic of the end of a child's first year. Parents should
be aware that a toddler's protests about day care are probably due to separa-
tion concerns; they are not likely to be a sign that he or she has been abused
or molested at the day care center.

Marital and Family Separations and Divorce

Infants and toddlers can experience separations and loss as a result of
their parents' occupations and relationships. The Iraq war, for example, has
90 Understanding Attachment

forced both mothers and fathers to separate from their young children,
causing distress, anxiety, and possible changes in attachment. For many
other infants and toddlers, marital disagreements, separations, and finally
divorce may disrupt attachments to parents.
Such separations certainly have short-term effects on many young chil-
dren between six months and two years of age. Far from being too young to
notice, they are strongly affected because they are too young to understand
anything that is happening. Anxiety and clinginess are common responses to
family separations. Sleep problems, depression, and temporary behavior
changes are also common in these cases, there being an obvious parallel to
the child's reaction to the loss of a familiar person through death.
Paradoxically, infants and toddlers often show what appears to be in-
creased distress when they have a visit with a parent who has left the house-
hold. For instance, an eighteen-month-old may have settled a bit (meaning
he may be sleeping better and appearing less disturbed) in the month
following his father's departure from the marital home. Then the father
reappears for an afternoon visit, and the child responds with renewed sleep-
lessness, clinginess, and restlessness. The custodial parent may read this as a
signal that visits had better be avoided, or may even believe that the visitor
treated the child abusively—a belief that may culminate in a charge of
sexual abuse so frequent today.
Handling this situation is a delicate task, especially in light of the emo-
tional distress of both parents and its negative impact on their coping skills.
It is difficult for people in this situation to see beyond the present discom-
fort to the long-term goal of maintaining a positive relationship between
the child and each of the parents. Nevertheless, it is clear that most parent-
child relationships have real value for the child's development, and that
such relationships are preserved by staying the course through the early
years of separation.
To maintain attachment to an often-absent parent, infants and toddlers
need visits that are as lengthy and as frequent as it is possible to make them.
These visits may—indeed, often will—be followed by increased expression of
distress by the child, but the distress is caused by the absence, not by the visit.
In these circumstances, tantrums or other displays of emotion should not be
punished, but accepted as what they are, a reaction to family disruption.
Ideally, the conditions for visiting should be as familiar as possible to the
child. Allowing the absent parent to come to the young child's home is the
best approach, allowing the child to deal with the situation in an environ-
ment of maximal security and familiarity. However, such home visits may
be agonizing for both parents, reminding them of matters they want to
forget. If the child shows distress, it may be extremely hard for the custodial
parent to stand back and let the other parent work to mend the relationship.
 Families, Experiences, and Outcomes 91

Of course, if other adults—grandparents, sisters, boyfriends—have come

into the picture, home visits may involve disagreements or even violence.
For reasons such as these, visiting parents often want to take infants or
toddlers onto the adults' own turf and deal with the relationship in a setting
with some degree of privacy and leisure. The problem with this approach
is that the parent may be frightened of unanticipated problems and may
decide to avoid visiting rather than risk having the child screaming in dis-
tress. Young fathers may be particularly in need of guidance here.
Take, for example, a situation in which an eighteen-month-old girl has
been sleeping in her mother's bed after a marital breakup; both mother and
child have found the physical contact comforting in a time of anxiety and
loss. Now the child is scheduled for an overnight visit with her father in his
new apartment. He does not know what the child's sleeping arrangements
have been and plans to put her in a crib in the guest bedroom. The little
girl screams with fear at this. She tries to get out of the crib, calls repeatedly
for her mother, and finally falls asleep exhausted at 4 a.m. This leaves the
father to ponder whether he is an acceptable parent and whether his former
wife has abused the child, spoiled her, or in some way turned her against
him. This father may well begin to avoid visits even though, in truth, he
loves his child and wants to maintain the attachment relationship. We should
note, by the way, there is a double whammy of consequences at work for
this father. Given our society's present concern with sexual abuse, if he
were to take the sobbing child into his bed, he would risk serious accusa-
tions, perhaps followed by a declaration there could be no contact and even
imprisonment.
The early months and years of a parent and child's visiting relationship
may be difficult to handle for all concerned. Perhaps the real key to success
is understanding the visit as a step toward a life-long attachment relation-
ship, and, therefore, toward the child's development of a positive internal
working model of social relationships. This means the visit needs to be fo-
cused on parent-child interaction that is developmentally appropriate for
the child. Constructive visits will not usually include the parent's romantic
partner; if the romantic relationship is a serious one, the partner can wait to
form a connection with the child until parent and child are doing well to-
gether. Similarly, grandparents need to wait their turn, unless they are al-
ready very familiar people who have their own secure relationship with the
child. It is especially inappropriate for a parent to arrange an adult social
activity and to leave the child in someone else's care in the course of the
visit. It need hardly be said that habitual lateness or failure to arrive for a
planned visit is not a way to establish a secure attachment.
Does the short-term distress of the divorced child also cause long-term
personality effects? This is a most difficult question to answer because the
92 Understanding Attachment

child who experiences parental separation will usually live for years in a
household shaped by that separation. The circumstances may include fi-
nancial uncertainty, the need to move to a new house, distress and depres-
sion for the custodial and perhaps the non-custodial parent, and possible
new relationships with stepparents and stepsiblings, all in addition to the
continuing difficulty of balancing a visiting relationship.
It is possible for parents and children to maintain strong, developmen-
tally appropriate attachment relationships, in spite of early marital separa-
tion, and for children in these circumstances to continue on a trajectory of
positive personality development. However, this desirable outcome re-
quires care, energy, and a certain amount of good luck, and may occur a
good deal less frequently than we would like.

SHAPING A T T A C H M E N T A N D
PERSONALITY: T H E PRESCHOOL YEARS
As we discussed in an earlier chapter, preschool children show their
emotional attachment in different ways from those typical of toddlers.
Preschoolers often initiate separation themselves, and they can accept it
well, just as they can be appropriately sociable with friendly strangers. In
cases where they are disturbed about separation, preschoolers may show
their feelings by squabbling with the mother when she returns.
The child who does this seems to be trying to re-establish important
aspects of the relationship with the returned mother rather than just try-
ing to be close to her, as a toddler would do. The internal working model
of the relationship is in need of adjustment, and the child is seeking to
readjust it.
The preschool child's existing attachment relationships seem related to
other aspects of mood and behavior. Secure attachment in the past seems to
accompany the preschooler's willingness to comply with rules and her
trusting acceptance of adults as legitimate authorities. Such children also
tend to cooperate with their peers and do not need a lot of control by adult
caregivers. Less securely attached preschool children can be deliberately
disobedient, and may behave provocatively in order to get the attention of
an unmindful parent.12 For example, in one family where the father was
physically ill and weak, the preschool children responded to his withdrawal
by acting up until he took off his belt to punish them.
Both serious loss and everyday stress have potential effects on children of
preschool age. Separation from familiar people and adoption or foster care
placement at this age can lead to long-term sadness and irritability, although
these are less likely when the new family is effective in helping the child's
adjustment. Even holidays, birthday parties, and vacation trips can trigger
 Families, Experiences, and Outcomes 93

negative responses in preschoolers, presumably because of unfamiliar events

and the withdrawal of the parent's normal attentiveness to the child. Using
time-out discipline techniques may simply worsen these situations if the
real problem is a sense of unnegotiated separation. Time-in, where the child
stays quietly with the parent until calmer, may work better.
Marital separation and divorce are still difficult for preschoolers to deal
with, although preschoolers have the advantage over toddlers of better lan-
guage skills and a more realistic concept of time. Children under the age of
four have much trouble with the long, hard task of letting go emotionally,
and, even if they will continue to see a parent on visits, they must let go of
their old picture of family life. Preschoolers are still immature, in emotional
reactions and in thought processes, so they are likely to persist in mourning
their loss rather than coming to terms with it. They have difficulty under-
standing that the loss is a permanent one, and they lack the clear awareness
of their own feelings needed for resolution.13
When divorce occurs, it is hard for preschoolers to accept the reality of
the parents' estrangement, especially if there are repeated trial separations
followed by reconciliations, or if one parent often voices the belief that
there will be a reunion. It is also quite difficult for the preschooler to dis-
engage from the parents' conflict and go on with developmentally appro-
priate tasks; preschoolers are not yet adept at separating their own feelings
from those of others. Preschoolers' reactions to divorce make it difficult for
them to put their energy into their normal developmental processes.14
The preschooler's reaction to loss is of a different nature from that of the
infant or toddler. Rather than the devastating withdrawal of security, the
preschool child experiences loss as an occurrence strongly contrary to his
need for joint planning and negotiation of separation, the foundation of his
present sense of trust and confidence in a parent. Neither death nor divorce
is negotiated with a child—and he or she may not even be informed about
what is happening until presented with a fait accompli. In addition, parental
reactions to either form of loss involve shock, anger, depression, and with-
drawal, reducing their capacities for negotiating and buffering any aspect of
the child's situation. Many non-custodial parents exacerbate the situation
by continuing to treat the child as he or she was treated when the divorce
occurred, rather than negotiating and fine-tuning the relationship.

THE SCHOOL-AGE CHILD:

A T T A C H M E N T , EXPERIENCES, A N D OUTCOMES
Teachers and Parents
By the time they reach school age, children have expectations about peo-
ple's interactions with each other. These expectations usually have evolved
94 Understanding Attachment

from the early attachment relationship between parent and child, a rela-
tionship that has its parallels in other relationships, such as those between
teacher and pupil, older and younger person, or a more powerful person
and a weaker one. First-graders are well known for their tendency to call
their teacher "Mommy" and address their mothers as "Mrs. Brown" by
accident. This does not mean that schoolchildren of any age think the
teacher is their mother, but it does show the child is considering the paral-
lel aspects of the relationships. The child is thinking about how teacher-
pupil relationships can be expected to work, given the extent to which they
resemble parent-child relationships. The internal working model of rela-
tionships, developed in the early years, may thus help determine how the
child acts toward teachers, who in turn will behave in ways that influence
the child's attitudes toward school.
For good or ill, however, parents remain at the core of a school-age
child's emotional life. How this works varies from family to family, class to
class, and culture to culture, but it is generally true that parents are at its
center, even for children who go to boarding school or who leave home
for other reasons.
The nature of the child's attachment to the adult continues to shape his
or her relationship. Of course, school-age children display much less attach-
ment behavior toward their parents than they did when they were younger,
and they may seem to take the parents for granted much of the time. When
the parents go on a trip or the child goes to sleep-away camp, however, the
child may show concern about the separation. For example, one ten-year-
old went to Girl Scout camp for the first time, when her parents were leav-
ing for a long-planned cruise; she cried every day, and finally disclosed her
terror that there would be a shipwreck and her parents would drown.
When there is an apparent threat, the child tries to feel more secure by
getting closer to the parent, talking to them on the phone, or even just day-
dreaming about a reunion. Under these circumstances, it becomes harder
for the child to do schoolwork or even play constructively. It does not
matter whether the parent is an unpleasant, unreliable, or even abusive
person; the child still turns to him or her when feeling threatened. 13 The
desire for contact with the parent can be set off by threats other than
separation, too: being bullied on the playground, getting sick at school, or
being punished in the classroom. The child's internal working model still
includes the idea that relief from anxiety can be obtained through contact
with the parent.
School-age children regularly experience separations from teachers who
have become familiar, either when the teacher leaves the school or when
the child passes to a new grade level. Because American families tend to be
geographically mobile, many children experience separation from teachers,
 Families, Experiences, and Outcomes 95

schools, and classmates as the family moves away. The practice of having
the child's home address determine which school is attended is another
common source of separation experiences, for even a move to a new house
may put the child in a new school district. Most difficult for the school-age
child is the family disruption following divorce. It often means a new house
and a new school, the transition coming at a time when the child is least
able to tolerate it.
Just as the school-age child is continuing to construct new mental catego-
ries and to learn new problem-solving skills, he or she is also constructing
new ways of thinking and acting toward parents and similar adults. In the
course of doing so, school-age children move toward behaviors and emo-
tions that allow some autonomy, as well as gratifying the wish to cooperate
and be on good terms with the parent. A school-age child may comply with
a parent's requests incompletely, but just enough so the parent is not actively
angry or punitive. (For example, the child puts her clean clothes in the dresser,
but leaves one sock hanging out or in the laundry basket.) Although such
behavior is annoying to parents, its goal is not actually to irritate them, but to
build toward a future relationship in which the child achieves greater inde-
pendence and the parent relinquishes responsibility and authority. Whether
the parent is relaxed and tolerant, or intrusive and authoritarian, about these
matters will help shape the child's understanding of later social relationships.

Sibling Relationships
As children get older, their relationships with siblings also add to the
internal working model of relationships, which is becoming more compli-
cated. Attitudes and relationships of school-age children with siblings are
determined by complex interactions among many factors. Power differ-
ences are almost always present (even between twins) because there are
differences in age, specific skills, or relationships with parents. In some fam-
ilies or cultures, the child belonging to the more favored sex has power in
spite of youth or small size. The internal working models of sibling rela-
tionships have not been studied much, but they probably have considerable
influence over other intimate relationships in later life. Especially when
children are near in age, sibling relationships seem to share the emotionality
of parent-child attachment relationships with an even greater display of
anger and frustration at times.
Although school-age children are able to do some negotiating in their
relationships with siblings and peers, they still need parents to provide some
scaffolding, or support, allowing the children to bring their social skills to
bear on emotional conflicts with other children. Because neither child in
the relationship is a very mature negotiator, they cannot do as well together
96 Understanding Attachment

as each could do with an adult. Conflicts and rivalry between siblings can be
terribly provoking to parents, but it is likely that with parental help, chil-
dren can use these conflicts to understand more about social interactions.

Models of Relationships with Babies

The relationships of school-age children with infant siblings are espe-
cially interesting for two reasons. First, they let us focus on the early stages
of an individual's development of parenting skills, and second, they let us
focus on the older child's own social abilities, there being a relatively small
contribution from the baby.
Some, but not all, school-age children show behaviors indicating their
emotional involvement with a baby brother or sister, and they may even
have been fascinated during the mother's pregnancy. Older siblings may get
close to the baby and work to get eye contact. They notice cues given by
the baby's behavior, such as pushing a bottle away, and they speak for the
baby as adults do, stating what they think are the infant's feelings and wishes.
The older child may express pleasure in the baby and practice rhythmic
turn-taking and initiation of communication. 16
Some older siblings, of course, do not do any of these things. They may
appear oblivious or actively hostile toward an infant sibling in the early
months. Affectionate interest in the baby seems to result from some family
characteristics.17 The older child has to be interested in the baby to begin
with—an older child who is unduly stressed by the birth may not feel inter-
est, and, of course, some babies are more responsive and interesting than
others. The older child also needs to have chances to hold and interact with
the baby without being directed by the parents, and not all parents feel com-
fortable with this. The development of a mutual relationship between older
child and infant is also more likely if the parents have modeled this tie in their
sensitivity and responsiveness to the older child. If the parents have included
the older child in planning and taking responsibility for the baby (feeding, for
example), but have not gone beyond the child's interest level, the relation-
ship is encouraged. Finally, it can be helpful if the parents remark on the way
the baby is responding ("I'm not sure he likes that.") rather than telling the
older child what to do. There is little research showing whether this family
activity helps the older child parent when he has grown up and had his own
family, but the experience would seem to be an effective way of elaborating
the internal working model of relationships to include infant care.

Divorce and Death during the School Years

Just as responses to separation told us about infants' emotional attach-
ment, family transitions like death and divorce may reveal important aspects
 Families, Experiences, and Outcomes 97

of school-age children's emotional development. Loss of a parent through

death or divorce creates a painful and stressful transition period, which has
a real impact on children's mood, school performance, and general adjust-
ment. Loss of a grandparent can have a similar impact, but the less close
the relationship, the less the child will be affected. In the case of Delimar
Cuevas, her abrupt separation from the woman she had thought was her
mother was no doubt exactly parallel to the death of a parent.
The death of a family member is more possible for school-age children
to understand than it is for preschoolers. Nevertheless, the process of griev-
ing for a deceased parent and returning to reorganized emotion is a matter
of at least a year. During the mourning period, children are likely to cry
more than usual and to have trouble concentrating in school. Some also
have difficulty sleeping. While these problems decrease by the end of a
year, bereaved children, at that point, are more likely than others to have
some health problems.18 Many children also dream and think about the par-
ent, often treasuring objects that belonged to the parent.
Though it has become a normative, statistically common, and socially
acceptable experience for school-age children in the United States, divorce
is nevertheless a family transition with major impact. Remarriage solves
some practical and financial problems for families, but the emotional adjust-
ment of children is about the same in divorced, single-parent, and divorced,
remarried families; in both cases there are more difficulties than in non-
divorced families.19
Both death and divorce trigger the need for change in the child's internal
working model of social relationships. It may actually be more difficult for
children to adjust to divorce than to the death of a parent. In both situa-
tions, the internal working model must alter to take in the fact that a person
to whom the child has been attached has simply gone away, but this does
not necessarily mean the adult did not love the child. The model has con-
tained some assumptions about negotiation and about the child's skills in
maintaining relationships, and now the child knows there are at least some
situations where these assumptions do not work. Human death becomes
more familiar to the child at this age, and it is possible for her to understand
that death is beyond human control. When divorce occurs, however, there
are some cognitive and emotional tasks forced upon the child; these tasks
are difficult and they are also highly relevant to the elaboration of the in-
ternal working model.
The emotional tasks of adjusting to divorce take years, and they may not
even be possible until adolescence.20 First, the child must simply accept the
fact that the parents' separation is real, not a joke or a mistake. Second, the
child needs to disengage from the parents' conflict and learn to ignore it;
the school-age child is better at this than the preschooler. School-age
98 Understanding Attachment

children have more trouble if parentalized by a custodial parent who turns

to the child for companionship and comfort. If the child does not disen-
gage, though, she will become more and more lonely for her peers and
have more difficulty in school, and the increasing stress will be problematic.
The internal working model will go through some confusing and inappro-
priate alterations, if the child becomes the parent's caregiver, reversing the
normal parent-child roles.
The child's third emotional task, in terms of adjusting to divorce, is reso-
lution of the loss—the act of giving up the wish for the lost family life, and
moving toward the future rather than longing for the past. This entails a
sense of powerlessness and humiliation in the face of the immovable ob-
stacles presented by the parents, who are both beloved and apparently in-
different to the child's feelings. Again, the internal working model of social
relationships must take in the complexity of the divorced family's relation-
ships. The same is true for a fourth task, the resolution of anger toward the
parents and the self-blame the child may experience, if he believes he has
been the cause of the divorce in some fashion. And, as a final step in dealing
with divorce, the child needs to accept the situation as permanent. There is
no going back, even though only about 25 percent of divorced parents
achieve a new, cooperative, co-parenting relationship with each other in a
way that a child can accept as an adequate permanent arrangement.21 These
steps need to become part of the child's internal working model; otherwise,
in the child's later view of relationships there will be a sense that there can
be no positive relationships in the future, that anger and blame are major
parts of important relationships, and that there can be no stability or conti-
nuity in intimate relationships.

ADOLESCENCE: EXPERIENCES
SHAPE T H E INTERNAL W O R K I N G M O D E L
Most adolescents and their parents have to work out ways of negotiating
separation after they have long shared a close relationship that evolved from
early attachment ties. Usually parents have developed some family patterns
that worked when their children were younger. Now, the dramatic changes
of physical maturation make some of the old patterns ineffective or even
destructive. Parents have to do a great deal of fine-tuning to maintain the
relationship while the teenager is changing so rapidly. The adolescent, too,
has to alter his or her internal working model of relationships, no longer
can a teenager think of the emotional connection with the parents as a
younger child does.
The way a family works together to change relationships depends on
their history together, and as well, on their present circumstances. For
 Families, Experiences, and Outcomes 99

example, in one study, girls who saw their mothers as affectionate and ac-
cepting were more capable of expressing their emotions than girls with less
accepting mothers. 22 The girls who could explain their feelings well used
their skill to help others to understand them. Thus, they could contribute
in a mature way to the reworking of family relationships, as well as using
the ability for a realistic remodeling of their internal working models.
We discussed earlier how important it is for preschoolers and parents to
participate in negotiation of separation. At that point in development, of
course, it is the parents who have reason to want to separate; the children
are the ones who generally like the parents near. During adolescence, when
negotiation of separation comes to the fore again, it is the teenager who
wants to move toward independence, while most parents cherish the child's
last years spent in the heart of the family. Ideally, negotiation can produce
a gradual separation; unfortunately some parents demand that the adoles-
cent remain a child to the point where he or she seizes independence pre-
maturely. In each situation, there is a different impact on the teenager's
internal working model of social relationships. A preferable model would
include the idea that people are able to work together, trusting in a mutu-
ally agreeable outcome. A less preferable view is that the support and care
of others comes only with complete compliance and dependence. Just as
parents of toddlers tend to use the time-out method of discipline—raising
more separation concerns—it is common for parents of teenagers to use
grounding, which increases the adolescent's anxiety about independence.
Negotiation of rules with adolescents can be a very effective parenting
technique. However, it requires time, energy, patience, and acceptance of
the idea that rules should change as a teenager matures. Parents who nego-
tiate also need to admit their side of the negotiation usually involves their
own needs, rather than some absolute standard of behavior. This admission
may be humiliating for some parents, who may feel as if they have revealed
a personal secret to the adolescent. For example, a recently divorced mother
may condemn a teenager's dating as disgraceful, rather than disclose her
anxiety about her own sexuality.
Negotiation of separation may be most difficult when the family history
has positioned the teenager as a hurried child.23 This term has been applied to
a child who has become parentalized into a parent's friend and confidante
following divorce. A divorced mother, for example, may tell her job and
relationship troubles to a preteen child, asking the child's opinion and ad-
vice. The child may experience adult-like responsibility for household tasks
and adult-like authority for decision-making about clothing, companions,
bedtime, and so on. When the hurried child arrives at adolescence, how-
ever, the parent becomes frightened by the autonomy the child has been
exercising. The possible dangers to an inexperienced adolescent become
00 Understanding Attachment

apparent to the adult, who knows how other people may interpret sophis-
ticated clothing on a thirteen-year-old girl or the freely chosen video of a
thirteen-year-old boy. The parent tries to reduce the teenager's power be-
low what it used to be, although the teenager knows that the teen years are
supposed to be a time of increased independence. Negotiation of rules will
not be possible for this family unless parent and child can explore the mis-
takes of the past, which will probably require the help of counseling.

The Adolescent Parent

Adolescent parenthood is often seen as a problem in industrialized coun-
tries. Teenage mothers are generally unmarried, uneducated, and unem-
ployed, and their lives do not give adequate support for a baby's development.
Although even schoolchildren may have begun to make a place for babies in
their internal working models of social relationships, most adolescents have
not yet achieved a model that takes infants' vulnerability into account. One
young mother, for example, said her two-week-old son was "just like his
father" and would hurt women. She complained that the baby had scratched
her face while waving his hands and declared he "knew it hurt" because she
had said "ouch." Another shouted at her baby to stop making "disrespect-
ful" noises and slapped him when he persisted with the sound. The adoles-
cent parent's immature internal working model can cause an insensitive,
unresponsive approach that fails to acknowledge how babies differ from
adults. The parent's immaturity thus fosters poor attachment in the baby.

A D U L T H O O D : T H E EFFECTS O F PAST EXPERIENCE

By the time adulthood is reached, every individual has had a wide expe-
rience of events related to attachment and has constructed the pattern of
thoughts, emotions, and behaviors that we call the internal working model
of social relationships. Although it is possible that events during adulthood—
like divorce or the death of a child—could influence a man's or a woman's
view of emotional ties, most studies in this area have concentrated on mem-
ories of the past.
An obvious question has to do with the effects of past attachment expe-
riences on marriage. Certainly, adults whose early lives featured neglect,
abuse, and abandonment would seem unlikely to trust or cooperate with
anyone, including their spouses. However, the importance of attachment
issues in marriage may be overwhelmed by other factors, such as cultural
and religious variables, economic considerations, childrearing needs, and so
on. It may be easier to see the effects of past attachment experiences on
other social relationships, and particularly on the relationship of the adult
with his or her own children.
 Families, Experiences, and Outcomes 0}

Pregnant women usually have a mental image or description of the fu-

ture nature of their yet-unborn babies. Because the mother cannot really
know much about the individual baby, it is possible that her ideas are
shaped largely by her own past attachment history and her current internal
working model of social relationships. In one recent study of pregnant
mothers' beliefs about their babies, the women's descriptions of their babies
were linked with the babies' actual attachment behavior at one year of
age.24 Mothers whose descriptions were complicated, rich with detail, and
coherent, and who expressed sensitivity toward the baby and a sense of
competence, were likely to have babies who showed secure attachments at
twelve months. These mothers were also likely to have gratifying memories
of their own childhood attachment experiences.
Though interesting, this study was not able to show exactly how the
mothers' beliefs and memories had effects on the babies' development of
secure attachment. (Certainly, there was no implication that the infants
picked up their mothers' prenatal thoughts.) The mothers' memories and
expectations somehow influenced their caregiving, and the caregiving pat-
tern in some way affected the babies' emotional development, culminating
in secure attachment. It was also true that some other factors in the moth-
ers' lives appeared to play a role here. It seems, however, that the mothers'
internal working models of social relations, derived from their own child-
hood experiences, were an important factor in shaping that most adult re-
lationship, parenting.

SUMMARY A N D CONCLUSION
Differences in early attachment experiences apparently help to shape
personality differences, especially those that have to do with social and
emotional relationships. Rather than a direct connection between infant
experiences and adult personality, however, we see a long series of gradual
steps toward development of the adult internal working model of social
relationships. These steps involve experiences with other people, but they
also depend on maturation of the individual's intellectual and emotional
abilities. Six-year-old Delimar Cuevas, whom we met at the beginning of
this chapter, would have responded very differently to her change of homes
than a twelve-month-old. Many children and adolescents experience fam-
ily situations that can distort the development of ideas about social relation-
ships, but they can benefit from positive guidance by adults who are aware
of the attachment issues characteristic of particular age groups. By adult-
hood, internal working models of social relationships should be complex
and many-faceted, allowing for appropriate social and emotional relation-
ships and behaviors with a variety of people, including infants.
This page intentionally left blank
 chapter 6

Attachment,
Mental Health,
and Psychotherapy:
Emotional Ties and
Emotional Disturbance

D
oes attachment have any connection to serious mental illness?
Many people have a vague idea that it does, but they are not very
clear on the details. One child care provider at a training confer-
ence questioned a speaker who had alluded to attachment: "Isn't there a
disease about that? Stranger anxiety or something?" Another member of
the audience, a nurse, spoke of a fifteen-month-old who was equally
friendly to all adults: "I just thought he was really nice and well-adjusted."
Talking about older children, people may describe a child as too clingy or
demanding or as a Mommy's boy. There may be general comments about
adults, concerning attachment issues or attachment problems.
This chapter will focus on connections between attachment and mental
health, focusing on the part attachment concepts play in psychotherapy and
the use of psychotherapy for attachment problems. Though there are many
unanswered questions in this area, the attachment concept gives us a useful
approach to the understanding of mental health.

ATTACHMENT AND MENTAL HEALTH

A repetitive theme in this book is that attachment plays out differently at
different ages. An infant or toddler shows characteristic attachment behav-
iors, and seems to have characteristic attachment emotions and expectations
about other people. Preschoolers, school-age children, and adults have age-
appropriate characteristics related to their attachment history. Individuals in
!04 Understanding Attachment

these different age groups also have their own forms of mental illness, which
are displayed in age-typical ways.

What Is Mental Illness?

The term mental illness refers to unusual and disturbed emotions,
thoughts, and beliefs, generally troubling the rest of us, as well as the af-
flicted individual. When these factors are only mild or intermittent, they
may interfere with the person's education, work, friendships, and love re-
lationships, causing loneliness and other difficulties. At a more severe level,
mental illness can make the individual dangerous to himself or to other
people.
For over a century, one focus of those who study mental illness has been
to construct diagnostic categories, or descriptions, of types of mental illness.
This task was formalized in the Diagnostic and Statistical Manual of the
American Psychiatric Association (DSM), a volume that has gone through
a number of revisions.1 The DSM takes into account not only the degree
of disturbance of a behavior but also its social acceptability. (For example,
homosexuality was, at one time, a category of mental illness in DSM, but
as social attitudes changed in recent years, the category was removed.)
Revised editions of the DSM discuss and list criteria for different diagnoses.
These reports are generated by working committees that must make sure
empirical evidence exists before adding a new category to the manual.
From a practical point of view, DSM provides diagnostic codes that are
used by private and public health insurers.

Mental Illness, Experience, and Genetic Factors

To begin this discussion of mental illness, we should remember there have
been many serious changes in thinking on this subject. In mid-twentieth
century, mental illnesses were commonly believed to have been caused
by the patient's past experiences. A patient's experiences as a child were
carefully scrutinized as possible causes of emotional disturbance, and the
mother's role was considered to be especially relevant, even perhaps blame-
worthy. Autism, for example, was attributed to the harm done by an overly
intellectual, so-called "refrigerator mother" who could not give her child
warmth and love. Schizophrenia was thought to result from early experi-
ence with contradictory, double bind communications, in which the child
could not safely discover exactly what an emotional message meant.
This stress on experience as the major cause of mental illness is far less
common today. Serious explanations of mental illness generally assume an
important genetic component, either one directly causing the emotional
disturbance, or one that makes the individual unusually vulnerable to
 Attachment, Mental Health, and Psychotherapy 105

certain harmful experiences. Because of this emphasis on genetic factors, it

is unlikely that modern psychologists or psychiatrists would count attach-
ment experiences, in and of themselves, as the cause of serious disturbance.
However, this does not rule out the possibility that attachment experiences
work with other factors to cause mental illness.

Infant Mental Health

The term infant mental health may sound unfamiliar or even a bit
bizarre. Can there be crazy babies? Can tiny children in some way parallel
the adult schizophrenic who hears voices, or the violent, paranoid patient?
No, of course not—and, in fact, infant mental health generally refers to
good physical, cognitive, and emotional development. These tend to go
together in infants, rather than dividing into clearly separated categories as
they do in older people.
Infants (and this term properly includes children up to about three years
of age) can show early symptoms of mental illness, however. A pervasive
developmental disorder, such as autism, may begin to affect the child very early
in life, even at birth. These disorders, which have genetic factors as their
primary cause, interfere with communication and therefore, with social
relationships.
Does attachment play a role in disorders of infant mental health? This is
a very difficult question to answer because many causes are linked with
many outcomes in early development. Sensitive, responsive parents are
likely to encourage attachment in a child, at the same time that they foster
good physical health, nutrition, growth, and language development. The
child who shows good attachment development is likely to be better off in
every way, not because he or she is attached, but because the experiences
that improve attachment also improve everything else. Factors like mater-
nal depression create insensitivity and unresponsiveness, which in turn in-
terfere with the development of attachment and of other behaviors and
abilities.
Children, who spend their lives in brutally neglectful foreign institu-
tions, like the Romanian children studied by Sir Michael Rutter, 2 have
little or no opportunity to form emotional attachments. But this is not a
complete explanation of their developmental problems. They are also
poorly nourished, receive little health care, get little exercise (or may actu-
ally be physically restrained), and experience little in the way of speech or
other communication. They may even have been placed in the institution
because of existing physical or mental problems. It would be a great mistake
to assume that attachment is the single, causative factor in these unfortunate
children's problems.
06 Understanding Attachment

An important consideration for infant mental health is the fact that the
attachment behaviors and feelings that would indicate emotional distur-
bance in an adult may be quite normal and healthy in a child. The child
care provider's question about stranger anxiety being a disease shows how
infant behavior may look when we mistakenly use an adult context. Anxiety
about unfamiliar people and places is a normal part of infant development,
as is the need for a secure-base person to make exploration possible. These
anxieties should almost never be considered indications of infant emotional
disturbance. On the contrary, as we will see later, it is the absence of anxi-
ety that might be an indication of mental illness.

Mental Health after Infancy

Pervasive developmental disorders, such as autism, are usually diagnosed
by the end of the infant period; the parents are alerted to the problem be-
cause the child fails to talk. During the preschool and early school years, the
problem most likely to emerge is that of impulsive, aggressive behavior,
which can verge on conduct disorder. Oppositional and Defiant Disorder
(ODD) is one syndrome characterized by this kind of behavior. During the
school years, some children are diagnosed with depression or with bipolar
disorder. Both O D D and bipolar disorder involve emotional disturbances that
impact social relationships and thus have a connection with attachment (al-
though they are probably not directly caused by attachment experiences).
In the late teens and early twenties, some individuals' thinking, emo-
tions, and behavior become so disturbed that a diagnosis of schizophrenia is
made. Schizophrenics' social relationships are severely distorted; these
problems are thought to be largely determined by biological factors and not
by previous attachment experiences.

DO ATTACHMENT
EXPERIENCES CAUSE M E N T A L ILLNESS?
Attachment is a robust process, and most children make a clear emotional
attachment to familiar caregivers. For a sizeable minority of the world's
children, however, attachment-related experiences do not follow a smooth
path. In industrialized countries, children may have caregivers who are de-
pressed or physically ill and who cannot provide the sensitive, responsive
care that is ideal. Domestic violence, alcohol, or drug use may lead to a
child being placed in foster care or adopted by a new, unfamiliar caregiver.
In the developing countries, the death of a mother from AIDS, or in civil
or other warfare, may leave a child who receives a minimum of care from
virtual strangers. Do these experiences cause actual mental illness? It is
 Attachment, Mental Health, and Psychotherapy 107

possible that a resulting disorganized attachment can be serious enough to be

considered a mental illness, but the answer to this question is not simple.

Disorganized Attachment Status

In a previous chapter, we looked at the Strange Situation and found that
a small proportion of toddlers responded in an unusual way to their parents'
departure and return. These children were classified as having a Disorganized/
Disoriented type of attachment, characterized by strange, contradictory be-
haviors and emotional expressiveness.
One study of Disorganized/Disoriented toddlers gave a detailed descrip-
tion of the children's unusual behavior. Some of these children showed
a sequence of contradictory behaviors, moving rapidly from one type of
emotion to another. Some children moved suddenly from what ap-
peared to be strong emotion to a much less emotional state. For instance,
"[ijmmediately following proximity seeking and a bright, full greeting with
raised arms, the infant moves to the wall or into the center of the room and
stills or freezes with a 'dazed' expression." For other children, the sudden
change was from calm to anger, as in a case where the "fi]nfant [was] calm
and undistressed during both separations from the parent, but becomes
extremely focused upon the parent, showing highly distressed and/or angry
behavior immediately upon reunion." Other Disorganized/Disoriented
toddlers managed to express simultaneously their desires to be near the par-
ent and to stay away. An example would be children for whom " [m] ovements
of approach are repeatedly accompanied by movements of avoidance such
as the following: (a) infant approaches with head sharply averted; (b) infant
approaches by backing toward parent; (c) infant reaches arms up for parent
with head sharply averted or with head down." 3
Another characteristic of Disorganized/Disoriented toddlers is their use
of undirected or incomplete movements. The authors described situations
where the child moves her hand toward the parent, then stops and pulls it
away without touching. There are also very "slow or limp movements of
approach to [the] parent, as though the infant is resisting the movements
even while making them (underwater' approach movements)." 4 These
underwater movements, which are less obvious than freezing, are defined
as "the holding of movements, gestures, or positions in a posture that in-
volves active resistance to gravity,"5 such as standing with the arms held out
waist-high. These odd movements may be accompanied by a "dazed or
trance-like facial expression ... [a] tense, smooth closing of the lids or ...
[a] lifeless stare."6
The researchers also saw these children acting frightened of the parent,
and acting confused when reunited with the parent. For example, the
108 Understanding Attachment

children sometimes looked frightened when picked up by their parents, or

a child would be seen "flinging hands about, over, or in front of face, or
over mouth, with fearful expression."7 These toddlers sometimes turned
suddenly and approached the stranger (Strange Situation) with their arms
raised. This happened when the parent reentered the room after the stan-
dard separation. As a final indicator of confusion, the children would some-
times fall to the floor when they approached the parent or the parent
reached toward them.
Do Parents Influence Disorganized Attachment Behavior? Do these descrip-
tions of odd behavior mean that inside these children there is a strange set
of ideas and feelings about people—a bizarre internal working model of
social relations? That is one of the possibilities, but it must be remembered
that any social interaction takes place between two people. Both parent and
child had roles to play in the Strange Situation, and further research showed
that the parents of these children were not behaving normally either.
Parents of Disorganized/Disoriented toddlers may be frightening to their
children, and they may be frightened themselves.8
For example, the parents showed unusual voice patterns in talking to
their children. They used a "breathy, extended, falling intonation" that
sounded "ominous" or "haunted," especially when greeting or touching
the child.9 The parents also used voices that dropped suddenly to a deep
pitch, which was described as "startling, especially when the speaker is a
woman whose pitch and intonation suddenly seem to belong to a male." 10
The parents of Disorganized/Disoriented children in this study also made
strange and unusual remarks to the toddlers. Sometimes their comments
suggested that real harm could come as a result of the child's actions. One
said, "You'll kill that little [stuffed] bear if you do that!" Another responded
to the toddler pushing a toy car across the floor, "Gonna have an accident!
Everybody's gonna get killed!"11 The parents also would suddenly begin
"chase" games, using the common "I'm gonna get you" pattern, but with
a frightening voice and movement style. Sometimes they showed directly
that they were afraid of the infant; one backed "away from the infant while
directing the infant not to follow in a stammering, apprehensive voice—
'Don't follow, d-don't.'" 12
Why would parents or toddlers behave in these very unusual ways?
The authors of the descriptions above felt there were several possible
reasons. Among them was the chance that there might be neurological
problems in some Disorganized/Disoriented toddlers. More importantly,
however, they felt these parents had had traumatic experiences, which were
attachment-related—for example, the death of one of their own parents.
Why would the loss of a beloved person make the mother's behavior so
strange and confusing to her own child? For one thing, the child may remind
 Attachment, Mental Health, and Psychotherapy 1 09

the mother of her own, early emotional ties and losses. She may then actu-
ally be frightened when she looks at the child, and she may want to get away
from, what is in her mind, a frightening situation. The child looks at the
mother's face and then looks around, trying to figure out what is so fright-
ening to her—he sees that he must be a source of her fear. This is in itself
disturbing to the child, but in addition, the mother cannot offer effective
comfort, because her fear is increased when she is close to the child. The
child has frequent experiences of alarm, but, in the end, he must cope with
them alone, even though his impulse is to rush to the mother for security.
Is Disorganized /Disoriented Attachment a Mental Illness? The Disorganized/
Disoriented classification does not seem to match our definition of a mental
illness (although it is unusual behavior and does involve emotional distur-
bance in the form of frightened behavior). It does not involve problems of
development as serious as some seen in twelve-month-olds, such as autism.
The Disorganized/Disoriented child still manages to communicate with
others, to develop language, and to play. If the child has attachment rela-
tionships other than the one with the mother, they may be more normal, as
long as the behavior of the other adults is more normal than the mother's.
Children with Disorganized/Disoriented attachment do have the poten-
tial to develop later disturbances of social interactions, both with peers and
adults, especially if they have few opportunities for early interactions that
are emotionally healthy. No one would deem this status a sign of good
mental health or positive future development. However, as so often hap-
pens, it is hard to know whether later problems result directly from the
early attachment status itself, from the fact that these children often have
been maltreated, or from the experience of having to live with a trauma-
tized and anxious parent.

Reactive Attachment Disorder and Alternative Diagnoses

Historically, one of the earliest concerns about attachment and mental
illness dates back to John Bowlby's reports about the forty delinquent boys
whose past attachment experiences he studied. Bowlby and others initially
felt that children who failed to form attachments had a strong potential for
criminal behavior. However, after further work, Bowlby and his colleagues
then reported they did not think that children who have been institutional-
ized or emotionally deprived would commonly develop psychopathic
characters (a conscienceless personality, possibly associated with attacks on
others who seem to be interfering with the person's wishes).13
Reactive Attachment Disorder: The DSM Category. The 1994 edition of the
Diagnostic and Statistical Manual of the American Psychiatric Association (DSM-
IV) included a category called Reactive Attachment Disorder. This disorder, often
1 10 Understanding Attachment

abbreviated to BJ\D, was considered to be a result of inappropriate early so-

cial experiences. DSM categories present a set of criteria to help clinicians
make a diagnostic decision, and in this case the criteria included both behavior
symptoms and the child's past history. Here are the diagnostic criteria for di-
agnostic category 313.89, Reactive Attachment Disorder of Infancy or Early
Childhood.14

A. Markedly disturbed and developmentally inappropriate social related-

ness in most contexts, as evidenced by either (1) or (2):
1. Persistent failure to initiate or respond in a developmentally appro-
priate fashion to most social interactions, as manifest by excessively
inhibited, hypervigilant, or highly ambivalent and contradictory re-
sponses (e.g., the child may respond to caregivers with a mixture of
approach, avoidance, and resistance to comforting, or may exhibit
frozen watchfulness).
2. Diffuse attachments as manifest by indiscriminate sociability with
marked inability to exhibit appropriate selective attachments (e.g.,
excessive familiarity with relative strangers or lack of selectivity in
choice of attachment figures).
B. The disturbance in Criterion A is not accounted for solely by develop-
mental delay ... and does not meet criteria for a Pervasive Developmental
Disorder.
C. Pathogenic care as evidenced by at least one of the following:
1. Persistent disregard of the child's basic emotional needs for comfort,
stimulation, and affection.
2. Persistent disregard of the child's basic physical needs.
3. Repeated changes of primary caregiver that prevent formation of
stable attachments (e.g., frequent changes in foster care).
D. There is a presumption that the care in Criterion C is responsible for
the disturbed behavior in Criterion A (e.g., the disturbances in Criterion
A began following the pathogenic care in Criterion C).

Reactive Attachment Disorder is referred to as the Inhibited Type, if the

child shows more of Criterion A l , and the Disinhibited Type, if more of
Criterion A2.
The RAD diagnosis continues to be included in DSM and has been used
in courtrooms and other public discussions. However, the category has
been severely criticized, and alternative suggestions have been, and con-
tinue to be, made. RAD has been described as "not a particularly clear or
coherent diagnostic entity." 13 The same authors note that not all children
who receive poor care develop symptoms of RAD, and some develop such
symptoms even though their care has been good.
 Attachment, Mental Health, and Psychotherapy 111

The DSM criteria for R A D do not make it clear how serious the
symptoms are, although many of them seem related to the disturbed be-
havior categorized as Disorganized/Disoriented attachment. It is hard to
know whether excessive familiarity means conversing eagerly with, or
climbing into the lap, of a friendly stranger. And how unfamiliar is a
relative stranger—is it the child's mother's friend who comes unexpect-
edly to pick him up at the day care center, a new child care provider, or
the mailman?
The RAD criteria also assume knowledge of the child's early social ex-
periences, which may be difficult or impossible to know accurately. The
child in question may have been abandoned, without any identifying infor-
mation, or, equally problematically, the story told by a noncustodial parent
or child care provider might be seriously biased.
Finally, the R A D criteria are confusing because they do not define de-
velopmentally appropriate behaviors, consider cross-cultural issues, or deal
with the effects of context, such as the behavior of the unfamiliar person or
characteristics of the environment. Because the criteria require more than a
brief observation of the child's behavior, decisions may be subject to dis-
torted reports by adult caregivers. It is possible that these factors will pro-
duce a diagnosis of RAD when some other diagnostic category would be
more appropriate.
According to one commentator, empirical studies of Reactive Attach-
ment Disorder are quite rare, and there have been difficulties with the
definition of the syndrome.16 The DSM description is unusual in that it
includes the background causes and attachment history of the child, as well
as his present symptoms. Most DSM categories focus on criteria related to
the individual's present situation.
ICD-10. Attachment disorders are also listed in the International
Classification of Diseases and Related Health Problems, commonly called
ICD-10. 17 ICD-10 is the European version of DSM. ICD-10 lists two
separate disorders: Reactive Attachment Disorder of Childhood and
Disinhibited Attachment Disorder of Childhood, and these are parallel to
the two types of syndrome combined in DSM. One difference between the
classifications is that ICD-10 does not include the background and history
as criteria for the disorders.
Alternative Ideas: Five Types of Attachment Disorders. The well-known child
psychiatrist Charles Zeanah and his colleagues have suggested five possible
outcomes of poor early attachment experiences.18 These categories can be
used with children from age one to four or five. The criteria are based on
behavior alone, with no attempt to depend on the child's early life history.
In addition, the category can be assigned even if the child shows the behav-
ior with only one adult caregiver—an important point, because as Zeanah
12 Understanding Attachment

has shown, children can have very different relationships and behaviors
with different familiar adults, as well as with strangers.
Zeanah's list of possible attachment disorders begins with Type I:
Nonattached Attachment Disorder. In this condition, the child develops no
preference for a familiar person and appears detached and uninterested in
people because he or she has had no opportunity to form attachments.
As we have seen elsewhere in this book, infants do not need a great deal
of help forming an attachment; they do need the chance to experience
the responsiveness of a small number of attentive people and to have
time to become familiar with them. Children with a Type I problem
may have been in institutions or had many changes of caregiver, and
perhaps were also badly neglected. They are able to form attachments
but have not had the chance. Their internal working models of social
relations do not link comfort or security with the nearness of familiar
people.
Boarder babies, who are sometimes cared for in hospitals for months,
their mothers having abandoned them, typically have lives that can lead to
a Type I disorder. They spend a good deal of time alone, with staff mem-
bers occasionally popping in, getting the baby's attention, and leaving again.
With this kind of social experience, the baby develops a quick, superficial
social reaction; he or she does not persist in interactions or seem upset
when the interactions stop abruptly. (Hospitals sometimes post notices in
these babies' rooms, requesting people to come in for fifteen minutes or
not at all.)
Zeanah and his colleague Alicia Lieberman gave a case description of a
child who fit the Type I category. He was twenty months old at the time
and had been with the same foster family for nine months. He was aggres-
sive and he bit himself and others, but he would stay for hours in the place
where he was put down. He did not reach for toys, preferring to stare at
spinning objects—often considered a symptom of autism. When moved
to a new living situation where he was favored, nurtured, and given
much attention, however, he improved greatly and began to show a prefer-
ence for one of the staff members. Apparently he had needed the much-
increased, responsive and interested interaction.
Zeanah and his colleagues have also proposed the diagnostic category
Type II: Indiscriminate Attachment Disorder, which resembles the disinhibited
type of Reactive Attachment Disorder described in DSM-IV. This group
of children will go to anyone, are overly friendly, and do not show the
usual secure base behavior, or run to a preferred caregiver when fright-
ened. Any adult seems to be equally comforting. (We should note the
potential dangers here, as these children are easily approached by molest-
ers.) Children who show these behaviors are likely to have been in
 Attachment, Mental Health, and Psychotherapy 1 13

institutions early in life, but by no means do all institutionalized children

fall into this group.
Zeanah has suggested that children in Type II may not simply be indis-
criminately friendly; they may combine that characteristic with accident
proneness and excessive risk-taking. Zeanah described one thirty-four-
month-old, risk-taking child, who was living with his mother and who was
hard to manage because of tantrums, aggressiveness, and other behaviors,
like running into the street. Put into the hospital for assessment, this little
boy did not behave recklessly. He played quietly and was unusual only in
his tendency to ask total strangers for hugs and kisses. When his mother
visited, however, he again took risks and went out of control. The risk-
taking behavior was part of the disturbed relationship with the mother, not
a characteristic of the child.
Zeanah's proposed Type III: Inhibited Attachment Disorder involves a very
different kind of behavior than described above. These children want to
stay near their familiar attachment figures, even at times when they might
be expected to play and explore.
Zeanah described two subcategories of a Type III disturbance. In the
Excessive Clinging category, the child shows an extreme and continuing
desire to stay near the caregiver (beyond the usual initial shyness that wears
off with time), as well as a tendency to show an anxious mood when around
unfamiliar people. One little girl described by Zeanah had apparently de-
veloped this behavior pattern as a result of her mother's ongoing fears about
the child's congenital heart ailment. The mother had a very different rela-
tionship with another daughter.
A second Type II subcategory is Compulsive Compliance, in which the
child obeys the caregiver immediately, a pattern he or she has developed
in order to avoid physical abuse. The caregiver may practice serious mal-
treatment or simply use excessive punishment. The symptoms of this
disorder are most obvious when the child is with the caregiver; the child
is likely to be subdued, wary, and unspontaneous. (Although normal
caregivers do try to establish some habits of quick obedience for safety
reasons—"Hot! Don't touch!"—an excessive amount of this interferes
with negotiation, which is a healthy part of the developing internal work-
ing model of social relations.)
Zeanah described one little boy named George, as he played in the foster
home and was then taken to visit his natural mother. Thirty months old,
George had been with the foster family for ten months. In the foster home, he
played freely and loudly and was somewhat resistant to the foster mother.
When a caseworker appeared to take him on a visit, "George screamed in a
terrified manner and attempted to hide," having to be carried bodily to the car,
where he cried throughout the trip. He stopped crying at the mother's door.
1 14 Understanding Attachment

He looked soberly at his mother when she opened the door, but he
did not greet her [He] walked in silently and stood in her living
room. The natural mother told George to sit down, and he did so
immediately George did not move from where he sat down
initially unless his mother directed him otherwise He watched
her constantly, maintaining a vigilant and anxious expression
He cringed slightly when she spoke to him, but he always complied
immediately with her directives.19

W h i l e this was going on, George's brother, w h o was visiting from another
foster h o m e , ran around and vocalized.
T h e difficulty George was having clearly emerged from the relationship
he had with his m o t h e r — h e did n o t behave this way with other adults—
nor had his m o t h e r developed the same relationship with the other child.
T y p e IV: Aggressive Attachment Disorder is another possible o u t c o m e of
p o o r attachment experiences proposed by Zeanah. This category involves
a persistent, pervasive level of anger and aggressive behavior toward the
parent, but can also include symptoms of anxiety, such as sleep disturbance.
T h e child may hurt himself (by head-banging, for example), may have
lengthy tantrums, and may hit, bite, or shout, "I hate you!" These children
have often been involved in, or seen, domestic violence. T h e y may have
received inconsistent care and unpredictable physical punishment.
Zeanah and Lieberman suggested one m o r e type of attachment disorder
called T y p e V: Role-Reversed Attachment Disorder. In this pattern, the child
behaves parentally toward the caregiver, b o t h nurturing and acting bossy or
punitive.
As an example, Zeanah described t w e n t y - f o u r - m o n t h - o l d Beth, w h o
was brought in for an examination because of sleep problems. H e r m o t h e r
was depressed and cried during the interview.

Beth played with [toys] only in the first 20 minutes, ... when her
mother seemed more animated. Her mother looked sadder and began
to cry as she described her frustration and hopelessness Beth walked
over and signaled to be picked up. She remained in her mother's
lap Beth began to stroke her mother's hair gently. She continued
caressing her mother's hair and back intermittently for about 20
minutes. At times, however, she pulled her mother's hair vigorously
and provocatively.20

Is Aggression Part of Attachment Disorder? Although statistics do not show

an increase in child and adolescent violence, Americans are very c o n -
cerned about aggression a m o n g children. T h e mass media have emphasized
killings by the y o u n g , and a topic of recent public discussion has b e e n
w h e t h e r very y o u n g killers should be tried and executed as adults. Are
 Attachment, Mental Health, and Psychotherapy 1 15

there connections between attachment experiences and youth violence?

As we will see, clinical psychologists and psychiatrists say one thing, and
popular opinion says something else.
We do have some empirical evidence about the connection of aggression
and attachment. Some degree of aggression is a part of normal develop-
ment. How much is acceptable, and what kind, depends on family and
cultural standards. But excessive violence is also part of a number of early
emotional disturbances. As for its connection with attachment, Zeanah's
Type IV attachment disorder involves aggression, but the DSM category of
Reactive Attachment Disorder does not. Aggressive behavior is described
as a problem in several early childhood emotional disorders. Oppositional
and Defiant Disorder (ODD) is characterized by aggressive responses to
adults. Autistic children may react with severe tantrums when an adult at-
tempts to guide or manage their behavior.
Serious psychotic illnesses, such as early onset schizophrenia, can also
involve intensely aggressive behavior. In a case that received much public-
ity a few years ago, Malcolm Shabazz, the then twelve-year-old grandson
of Malcolm X, set the fire that killed his grandmother, Betty Shabazz. The
boy had shown serious emotional disturbance since early childhood: he had
set fires and spoken of an imaginary friend called Sinister Torch. He and his
mother had had physical altercations that left them both bloody. Although
this case and others undoubtedly involved disturbed relationships, attach-
ment issues were by no means the direct cause of the violent behavior.
Children who have had poor attachment experiences are also likely to
have experienced violence and to have modeled their behavior on others'
aggression. As Zeanah noted in his analysis of Aggressive Attachment
Disorder, these children were likely to have been involved in domestic
violence. When children are institutionalized, too, chances are all too
great that they will undergo or see some violence perpetrated by their
peers or staff members. Children in the foster care system are often there
because of domestic violence, and they may encounter still more violence
within the system. However, this source of aggressive behavior is parallel
to the experiences that cause disorders of attachment, not a direct result of
those experiences.
Some beliefs about attachment and violence are without an evidence
basis. A section in Chapter 7 will discuss a popular theory of attachment
disorders that is advocated by some practitioners and parent groups. We
can refer to this belief system as the Attachment Therapy (AT) philosophy.
Adherents to this set of ideas stress a connection between rage and disor-
ders of attachment, but they do so without any evidence basis. This group
agrees with the view so succinctly stated in West Side Story: "He's depraved
on accounta he's deprived." The depravity in question is considered to
1 16 Understanding Attachment

begin with disobedience and cruelty to animals and then develop into serial
killing, a characteristic expected of all adopted children who are seen as
suffering from attachment disorders.
The AT belief system emphasizes the idea that complete control over
children's actions is the only way to eradicate the intense aggressiveness
expected to result from a poor attachment history. All adopted children are
considered to be full of rage and to direct their anger first at younger chil-
dren and pets, and later at their adoptive parents. A treatment intended to
neutralize rage is advocated by this group and will be discussed later in this
chapter.
Again, there is no systematic evidence to support this belief system. It is
clear from both the DSM criteria for Reactive Attachment Disorder and
from Zeanah's categories that aggressiveness is not a hallmark of disorders
of attachment, although it may be present in some cases.
Assessing Attachment Disorders in Older Children. The descriptions of at-
tachment disorders given above show a good deal of variation. There does
not seem to be a single, invariable form of attachment disorder even among
preschool children. Although poor attachment experiences have a real po-
tential for causing poor emotional outcomes, their consequences can be
different for children in different situations, and can be influenced by indi-
viduals' basic temperamental characteristics. By the time children reach
school age, they have developed along individual lines to such an extent
that early attachment experiences are only one factor among many that
determine emotion and behavior. Assessments of Reactive Attachment
Disorder at that point may not be possible, as leaders in this area of research
have stated.
Teachers and parents need to be cautious about claims that Reactive
Attachment Disorder can be diagnosed by means of checklists or through
observations by mental health counselors. One group of counselors, for
example, has suggested that they can diagnose R A D in three to seven
controlled five-minute observations, but the evidence presented is quite
questionable.21

ATTACHMENT, DISORDERS OF
ATTACHMENT, AND PSYCHOTHERAPY
Serious disorders of attachment need treatment. Not only do they cause
unnecessary stress and fear in early life, but as development progresses, they
distort the child's experiences and beliefs about other people. This section
will examine therapeutic interventions that seem to help compensate for
a poor early attachment history. In addition, this section will consider the
fact that even people with mild attachment problems may benefit from
 Attachment, Mental Health, and Psychotherapy 1 17

having these considered during psychotherapy in adult life. And therapists

themselves need to take their own attachment experiences into consider-
ation as they work with patients.
In the discussion of therapy below, the reader may see what seems to be
peculiar language. Interventions that are intended to deal with disorders of
attachment or that take attachment issues into account are often called
attachment-based therapies. Why not simply attachment therapy? As we will see
below and in Chapter 7, the term attachment therapy was preempted some
years ago by a group of quasi-professional practitioners, who use it to mean
a treatment of their own design. This treatment is diametrically opposed to
the ones we will stress; it is based on a completely unorthodox set of beliefs
about attachment. In order to clarify the differences between the unvalidated
attachment therapy and evidence-based treatments, it has become important
to use terminology that shows what type of intervention is being discussed,
and the term "attachment-based" indicates treatments generally accepted by
clinical psychologists, clinical social workers, and psychiatrists.

Therapists and Other Intervenes

Therapeutic interventions for childhood attachment disorders are far
from the simple talking therapy model used in adult counseling and similar
clinical work. Instead, many types of professionals employ a wide range of
interventions. These can include parent education, foster parenting tech-
niques, residential treatment, medication, and individual or family psycho-
therapy. The people performing the interventions may be teachers, social
workers, psychologists, psychiatrists, and foster, biological, or adoptive
parents.
What Do Intervenes Bring to the Table? We saw earlier in this chapter that
a child can show disturbed attachment behavior with one adult caregiver,
but behave quite normally with another. The adult brings something to the
situation, and the child responds to it. Similarly, we see that adult interven-
e s can influence a child's attachment-related behaviors through their own
emotions and expectations.
It is by no means a new idea that both patients and therapists bring their
emotional histories to the treatment room. The patient's emotional back-
ground has long been thought to provide a unique responsiveness to and
relationship with the therapist. The feelings that result from past emotional
experience are termed transference, and are considered a refocus on the ther-
apist of some beliefs that stem from early relationships with parents. For
example, the patient might feel angry with the therapist, not because of the
therapist's own actions, but because of a pattern of feelings toward au-
thoritative, caregiving people, a pattern that developed in interactions with
1 18 Understanding Attachment

the patient's parents. This transference of patterns is like the application of

an internal working model of social relations, a model developed early in
family life.
Far from being utterly rational, completely neutral observers, therapists
also bring their own emotional backgrounds to the therapy room in the
form of countertransf erence. The therapist's responses to the patient are shaped
in part by the therapist's internal working model. Countertransference has
sometimes been a matter of concern, because it can interfere with the ther-
apist's objectivity. Recently, however, some therapists have begun to think
about countertransference as an authentic part of the interpersonal relation-
ship between therapist and patient, and one that gives important feedback
to the patient about other people's feelings. This may be especially impor-
tant when the problem to be resolved involves separations, loss, and
trauma.22 Therapists who work with traumatized patients may feel much
anxiety and find the intensity of the relationship overwhelming. The ther-
apist wants the patient's attachment to be transferred or refocused on him,
but this involves accepting both the patient's desire for closeness and its as-
sociated danger. Many therapists find the situation awkward, difficult, and
a producer of many contradictory emotions. A great puzzle for therapists is
the extent to which they should allow the patient to see these feelings, and
the extent to which they should suppress their emotional responses.
As with the relationship between therapist and patient, children and
caregivers are affected by their interpersonal interactions. Children affect
adults, who in turn affect the children, who in their turn influence the
adults again—or, if you want to begin the sequence with the adults, that
statement would be equally true. Children have immature internal working
models of social relations that help guide their interactions, and adults have
much more advanced models that make their own contributions.
As we saw in an earlier chapter, adults' internal working models can be
measured by means of the Adult Attachment Interview (AAI). Some re-
search indicates that the most effective foster parents, the ones who best
establish a secure attachment in their foster children, are those who are
judged autonomous on the AAI. These parents have confidence in emo-
tional attachment and are able to think coherently about their own early
attachment experiences. Children are likely to show high rates of disorga-
nized attachment behavior when cared for by foster parents who do not
have an autonomous state of mind with respect to emotional ties.23
People who foster or adopt young infants are in an excellent position to
establish a secure relationship and guide the child to a positive internal
working model of social relations. For those who care for older children,
however, the relationship may be rockier. This is not surprising when we
recall that the older child brings along an internal working model in which
 Attachment, Mental Health, and Psychotherapy 1 19

adults may be expected to be untrustworthy or even cruel. W h a t is m o r e

surprising is that foster or adoptive parents may inadvertently behave in
ways that bring out the worst in the children and in the family system.
In one clinical study, a n u m b e r of problems were cited as characteristic
of adoptive parents of older children. 2 4 All the parents in this study were
worried and distressed by the children's behavior, even to the extent of
regretting the adoption, feeling anger at the child, and experiencing c o n -
cern that the child might have inherited the biological parents' problems.
These reactions are difficult to hide from children, and are b o u n d to influ-
ence the affection, spontaneity, and playfulness that w o u l d ideally be part
of the children's lives.
T h e adoptive parents w e r e also unexpectedly oblivious to the child's
communication of anxiety or wishes for care. It was as if the parents were
so convinced the adopted child could not prefer or care for them, they
missed the child's signals. For example:

A 15-month-old girl looked momentarily worried when the father

left the room, and smiled imperceptibly on his return. He remarked:
"she did not notice I was gone, did she?" ... Parents also misinterpreted
temper tantrums, defiance and noncompliance as signs that the child
did not care for them, instead of seeing them as age-appropriate
expressions of anxiety and fear of loss.25

T h e adoptive parents observed in this study also tended to discipline

children for behavior that was actually an indication of the child's attach-
ment. T h e y often used t i m e - o u t techniques, w h i c h unfortunately fed into
the child's fear of rejection, separation, and abandonment. These parents
did not usually choose firm but comforting responses such as time-in, an
approach in w h i c h the child stays close to the parent until calm. T h e au-
thor, Alicia Lieberman, noted that few of the adoptive parents had been
well prepared by their adoption agency, especially with respect to the e m o -
tional needs of later-adopted children.

Types of Intervention
Treatment of emotional problems today most c o m m o n l y emphasizes the
family system and assumes that every family m e m b e r influences each of
the others. Therapeutic interventions with children are especially family-
oriented. Children are not sent to be fixed, nor are children the sole focus
of intervention, even w h e n the child's behavior and feelings are the reasons
the family has sought treatment. W h e n the child's difficulties seem to be
relationship-connected, all the family's relationships need to be considered,
and m o r e than one may need serious w o r k .
20 Understanding Attachment

Unfortunately, this is sometimes unacceptable to parents; some cannot

tolerate it well enough to participate in therapy. When the therapist says
that every family member has an influence on the child, parents sometimes
seem to hear this statement as an accusation of blame for the child's behav-
ior or emotional state. Treatment must involve the parents, as well as the
child. It can be quite difficult to convey this message without implying that
the parents are responsible for the problems. Families often prefer to bring
their child to therapy so that the child can be "worked on," rather than do
the complex, cognitive, and emotional work of understanding how the
family functions. Indeed, some therapists prefer this approach, too.
When attachment is an issue, however, the therapeutic work must in-
volve relationships, and each person in a relationship contributes to the
outcome. It is difficult to imagine how a child could be treated for a disor-
der of attachment without the participation of a major attachment figure
from the child's everyday life. To do otherwise would be like doing mar-
riage counseling with a husband or a wife alone—this is better than noth-
ing, but not nearly as effective as having the couple work together.
Whether or not a child has been diagnosed with a specific attachment
disorder, there are a number of different circumstances in which therapy
may be sought for problems associated with attachment. Families who have
adopted an older child and feel that child is unreachable may want interven-
tion. Adopted or non-adopted children who take risks or disobey frequently
may be thought to need attachment-based interventions. Adults who have
difficulty with relationships may also seek attachment-related therapy.
Autistic children are no longer considered candidates for attachment-
based therapy, although psychologists once thought their pervasive prob-
lems stemmed from the absence of attachment. Some parents, unfortunately,
are still convinced that the serious developmental problems of the autistic
child are relationship-based, and that genetic problems like Tourette
Syndrome can yield to relationship-focused treatment.
As we think about attachment-based therapy, we need to consider the
treatment of poor children who are diagnosed with serious emotional prob-
lems. In many states, unhappily, families cannot receive state aid for a child's
needed mental health intervention unless the parents relinquish custody.
The parents' rights are legally terminated, the child no longer lives with
them, and the parents have no right to visit with the child or even to obtain
information about her condition or whereabouts. Whether or not the
child's emotional disturbance began on a relationship basis, the effects of
separation and loss now become part of the problem. At the same time,
intervention with the biological family becomes impossible.
Parent Education. The clinician Alicia Lieberman, whose observational
research was discussed earlier, has suggested that parent education is an
 Attachment, Mental Health, and Psychotherapy 1 21

essential intervention, especially for adoptive families. Adoptive parents

need to know what special problems to expect from an adopted child, es-
pecially one who has had little opportunity to form attachment relation-
ships, having lived in multiple foster homes or spent months, or even years
in a foreign institution. In addition, these parents need to receive "emo-
tionally supportive, child-focused developmental guidance, which allows
the parent to become more receptive to the child's often overlooked or
misunderstood efforts to obtain reassurance."26 For children who use and
understand language, this reassurance can come in the form of storytelling,
having a discussion about what has happened to the child, and naming the
child's fears, with an assurance that they are groundless. For children who
do not talk, games of peek-a-boo or hide-and-seek may convey some of
the same messages—and these games are fascinating to children who are at
the point in development when attachment concerns are most intense.
Adoptive parents who have raised their own children are not necessarily
ready to deal effectively with the adopted child. They may be sensitive
enough to ensure a good outcome in a relatively easy situation, but they
have to work harder and think more about picking up the signals of ad-
opted children, who need a particularly strong sense of continuity and
communication. These children's communication cues may be easy to
miss; the children may miss the parents' cues as well. For example, Lieberman
suggests, "Parents need to over-emphasize their response to the children's
muted signs of need, marking separations with unambiguous demonstra-
tions of sadness and assurances of a prompt return, and underlining reunions
with clear demonstrations of joy." 27 An analogy may be drawn to what we
do when speaking to a hearing- and speech-impaired adult; we speak loudly
and clearly enough, we give the person a view of our face, and we use
gestures and repetitions. To understand the adult, we pay careful attention
and try to become familiar with the person's unique habits.
Guidance for Foster Parents. Foster homes are sometimes thought of simply
as holding environments—somewhere for children to be fed, sheltered, and
protected, before a permanent arrangement is made. In reality, foster care
provides a variety of scenarios for children. Some children remain in foster
care for years without permanent arrangements being made for them. They
may be in placement because the biological parents cannot provide for
them, because of parental illness, or for other reasons in which parental
rights would not be legally terminated. It may be that the biological parents
try again and again to care for the child, but on each occasion are forced to
give the child up because of longstanding or new problems. Other children
may spend a short time in foster care, but are there because of traumatic
circumstances, such as the death of a parent in a car accident, or, worse,
domestic violence.
122 Understanding Attachment

A relatively recent foster care problem is the shortage of foster parents

w h o can be full-time caregivers for the children. In a n u m b e r of cases, foster
mothers w o r k outside the h o m e , and children in their charge are often
placed in day care, further complicating their experience with relationships.
C a n foster care b e c o m e an effective intervention and not just a way to
keep y o u n g children off the streets? T h e r e is n o doubt that it can and
should. W h e t h e r this will ever happen is questionable, however, because
the steps necessary to make this an effective solution will not be cheap.
Foster care needs to b e c o m e an experience that can guide children to the
development of positive internal w o r k i n g models of social relationships. As
w e k n o w , the first step toward this accomplishment comes from experi-
ences with a small n u m b e r of consistent, sensitive, and responsive caregiv-
ers. Consistency of caregiving is difficult to achieve if children attend day
care and live in a foster h o m e , w h e r e there are apt to be too many caregiv-
ers. Increased wages for foster parents w o u l d be needed to enable one or
b o t h foster parents to stay at h o m e with the foster child.
Some specific suggestions have been made about improved training for
foster parents. 28 O n e idea is that foster parents learn h o w to gently chal-
lenge y o u n g children to change their expectations about other people and
to alter their internal w o r k i n g model, o n e that predicts rejection and indif-
ference from adults. Foster parents behaving w i t h normal sensitivity are
likely to respond by backing away w h e n foster children seem not to w a n t
them. H o w e v e r , the children do need care and contact, and foster parents
need to be helped to give this care even t h o u g h from the child the adult
gets only mixed or negative signals. H e r e is o n e such intervention:

[C]aregivers [are encouraged] to provide nurturance to these children

even though the children are behaving in ways that suggest they do not
want or need nurturance flfT] a child turns away from the caregiver
when hurt, the caregiver is encouraged to respond in a nurturing way;
for example, at first the foster mother may just pat the child's back
gently while the child sits on the floor turned away from her. Over
time, the child usually allows the foster mother to come closer, and
offer more support and nurturance. 29

A second point in the same intervention program involves the unusual

physiological reactions to distress many y o u n g children have w h e n they
have been traumatized by separation. Foster parents need to learn to buffer
children's stress levels by creating a "very controllable interpersonal w o r l d "
in w h i c h the child knows w h o will be present and h o w they will consis-
tently respond to social communications. 3 0
Floor Time (DIR). W h a t about direct individual treatment of children w i t h
attachment-related problems? Are there therapeutic techniques that seem
 Attachment, Mental Health, and Psychotherapy 1 23

to work? One important approach has been reported to be effective when

used by parents, teachers, and therapists, either individually or together.
This technique, developed by the leading child psychiatrist Stanley
Greenspan, is called floor time, or Developmental, Individual-difference,
Relationship-based therapy (DIR). 31
DIR is based on the assumption that cognitive, social, emotional, sen-
sory, and physical development are interlinked. Communicative experi-
ence with other human beings creates emotional reactions and a chance to
learn what to expect in the world. Floor time involves communicative
experiences. It is intended to support the development of capacities much
like the internal working model of social relations posited by Bowlby's at-
tachment theory. These capacities involve a set of interwoven feelings,
expectations, beliefs, and memories that serve as guides for relationships.
Floor time is, in practice, exactly what it sounds like: an adult gets down
on the floor with the child and follows whatever leads the child gives. The
adult—whether parent or therapist—does not use floor time to tell or show
the child what to do or not to do, but to help support the child in his own
ideas and plans. If the child can and will talk, talking can be part of floor
time, but nonverbal communication may be all that is used.
Adults who are doing floor time need to be patient, relaxed, and atten-
tive to the child's cues. Otherwise, the rule of following the child's lead
becomes impossible to obey. An anxious, impatient adult may also make
the mistake of thinking that a child is doing something intentionally, when
this may not be true.
Floor time needs to begin with engagement between the adult and the
child; the adult becomes attuned to the child's thoughts and feelings by try-
ing to respond to the cues the child gives, even when these are hard to read,
as may be the case with foster or late-adopted children. Floor time proceeds
with two-way communication, allowing the therapist to understand which
responses draw a returned response from the child. In a video recording of
Greenspan working with a withdrawn four-year-old, the therapist shows
the boy a toy lion. "Can he shake your hand?" The boy indicates refusal
with a brief movement of his head. "Can he shake your foot?" Again, no.
"Can he shake my hand?" A quick nod indicates the child's agreement, and
a very simple form of two-way communication is underway. The child does
not speak, but the rudiments of a relationship with the therapist are there.
Floor time can be an effective way for caregivers and therapists to work
with a child who seems aloof, withdrawn, indifferent, or out of control in
ways that are common among children with disturbed attachment histories.
An unusual characteristic of the D I R approach is its stress on children's in-
dividual differences, including their abilities for self- regulation and any spe-
cial sensory processing tendencies that can interfere with communication.
124 Understanding Attachment

Attachment Therapy and Related Treatments. Earlier in this chapter, we

noted that the term Attachment Therapy—the one we would intuitively
use in talking about treatments for attachment disorders—was preempted
some years ago by a group of practitioners whose work differs vastly from
anything we have so far discussed. As we will see in a later chapter,
Attachment Therapy (AT) is an unvalidated treatment; it is unrelated to
Bowlby's attachment theory and is directed primarily toward adopted chil-
dren, whether or not they show symptoms of attachment disorders. AT
uses physical restraint and emotional intimidation as its primary tools. There
is little or no systematic evidence about its effects, and the treatment may
accurately be described as "a cure without a disease."32
Parents adopting children from foreign countries are especially likely to
be advised by their adoption agencies to seek AT practitioners or persons
who use similar techniques, but who object to having their work classified
as AT. Psychologist Ronald Federici, who specializes in post-institutionalized
foreign-adopted children, advises parents to use a beltloop technique with
newly adopted children.33 This technique permits the child to go no more
than an arm's length from the adoptive parents, except when sleeping, for
weeks at a time. In addition, Federici suggests parents use a prone restraint
technique when the children are angry or disobedient, a practice that has a
higher probability of asphyxiation than other restraint modes. He also ad-
vises restraining the child in a cradling position to create an emotional tie
betv/een child and adult. None of these practices has an evidence basis, and
the techniques Federici recommends seem rather similar to the inappropri-
ate responses of adoptive parents as recorded in Alicia Lieberman's research,
which was discussed earlier in this chapter.
Attachment Issues and Therapy in Adults. There are a great many situations
in which attachment concerns may affect adult relationships, but attach-
ment is probably most critical when parent-child relations are in trouble.
The space limitations of this book will confine us to a few situations of that
type and the interventions they require. One fascinating topic, far too com-
plex to be considered here, involves the emotions connected with Assisted
Reproductive Technologies (ART), such as artificial and in vitro insemi-
nation, or the use of gestational surrogates. A R T is increasing as people
delay childbearing to an age of decreased fertility, and as more same-sex
couples seek parenthood. Couples' emotional responses to a successful
pregnancy following A R T are by no means simple, and therapists working
with these adults and their children need to keep many attachment issues
in mind.34
More common problems between parents and children have to do with
the parent's history of traumatic experiences, such as rape. These parents
may have a diminished capacity for sensitive responsiveness to the child.35
 Attachment, Mental Health, and Psychotherapy 125

Instead of responding to the real baby in the here-and-now, they react in

terms of an internal working model distorted by trauma. Like the woman
suffering from major depression, a mother with a history of trauma has a
compromised ability to respond to her baby. Unfortunately, the baby's cry-
ing, or her other signals for help, may reduce the mother's responsiveness
even more. "When a baby's distress or another feature of the mother's cur-
rent environment—such as domestic violence involving the baby's father—
stimulates negative affect in a mother ... [she] may confuse past traumatic
memory with her perception of her baby, and herself, in the present."36
In one case, when pregnant seventeen-year-old Marta began to feel her
baby move, she started to have memories and nightmares of having been
sodomized by her father at age five and raped by her brother about a year
before she became pregnant. She said these memories and dreams began
when her doctor told her the baby was a boy. When the little boy was
born, Marta thought he was "angry" and "manipulative." She found herself
so exhausted that she did not wake up if the baby cried at night. Clearly,
Marta's perception of her baby was distorted and unrealistic. If she re-
sponded to the baby's cry as manipulative, she could not be successful at
establishing a good relationship and providing good attachment experi-
ences for the child. Marta had little Reflective Functioning—the ability to
think about other people's minds and her own—so she was not likely to
solve this problem for herself.37
What could be done for Marta and her child? She was certainly a person
who needed to be shown rather than told, but how to manage this? Her
problem does not seem to be the right kind of situation for a talking kind
of therapy.
What worked for Marta was a series of video feedback visits, which
helped her see the interaction she had with the baby. This procedure began
with the videotaping of mother and baby as they played and interacted.
The videotapes were reviewed, and four 30-second segments were chosen,
including moments of separation and reunion.
Several weeks later, the mother was asked to come back and watch the
videotaped segments, together with a clinician. After each video clip, the
mother was asked to give her responses to some questions: "What hap-
pened? What is going on in your/your child's/the interviewer's mind?
Does it remind you of anything?"38
This simple intervention is effective in helping mothers reinterpret their
children's signals in a more accurate way. For example, one of the mothers
who received this treatment, Ms. V., had a three-year-old son, whom she saw
as angry and controlling. At the time of the videotaping, he held on to her to
keep her from leaving the room. She described this at the time as "trying to
get away with something ... trying to have things his way ... and ... feeling
126 Understanding Attachment

angry." After watching the video with the clinician, Ms. V. responded very
differently and far more accurately: "I think he was scared Oh, my God!
That was a scary moment for him. He probably thought that I was leaving
and wouldn't come back."39 Though the video feedback intervention was an
experimental one, and the families all received more conventional psycho-
therapy, this treatment may prove to be an effective way to help traumatized
mothers become appropriately sensitive to their children's communications.

CONCLUSION
When attachment experiences are poor, individuals' internal working
models of relationships seem to develop along distorted lines. The early
models are based on a few primary relationships, and if those are all diffi-
cult, the child's beliefs and expectations about relationships will also be
problematic. Because relationship quality depends on both child and adult,
even a child with one very poor relationship may benefit from other, better
relationships. A very few children have no consistent relationships, but
even these children have models of how human beings interact, because
they have experienced some minimum care and companionship.
Children with poor experiences, involving interactions with inconsis-
tent, unresponsive adults, begin on a developmental trajectory or pathway
in which they have less and less opportunity to develop healthy expecta-
tions and good social skills. An adopted child, for instance, may turn away
from the parent when hurt; if the parent accepts this as meaning the child
does not like, or want, care, the child's belief that people are not a source
of nurturance will be further confirmed. When the child's development has
been affected repeatedly by such experiences, he or she may become so
negative in his or her emotional responses and behavior toward others that
the question of mental illness will arise.
The term attachment disorder describes an emotional disturbance that is
the outcome of poor early social and emotional experiences. However,
complete agreement among psychologists and psychiatrists does not exist
on the behaviors and feelings associated with disorders of attachment, and
a number of possible types have been described. Aggression does not seem
to be a major feature of attachment disorders.
Because attachment disorders arise from experience, treatment focuses
on experiences that can change the individual's understanding of the self
and of relationships with other people. The most effective treatments in-
volve improvement of communication between adult and child; they re-
quire the adult to increase his or her sensitivity and ability to respond to the
child's communications, especially those about fear of separation.
 chapter 7

What Everybody
Knows: Popular
Views of Attachment

A
s the first chapters of this book have shown, the concept of at-
tachment is complex, and it remains the subject of much discus-
sion in professional child development and psychotherapy circles.
This fact does not prevent simplified views of emotional development from
becoming popularized. Indeed, the idea of attachment is broadly accepted
in the United States, although not necessarily with the definition we have
used in this book. Decisions and practices that claim to be based on attach-
ment are readily accepted by the media and the public, even though attach-
ment has a variety of definitions. In our Age of Attachment, statements
about emotional ties may go unquestioned.
The present chapter will examine three major popular ways of thinking
that use the term attachment, but fall outside the realm of professional psy-
chology and psychotherapy. Two of these belief systems involve parenting
practices and treatment of children. The third involves ideas that have been
developed in the courts and have strongly affected child custody and visita-
tion decisions.

ATTACHMENT PARENTING
New parents seeking advice about childrearing may come across books
and Internet sites that counsel the use of attachment parenting, an approach
developed by a physician named Dr. William Sears and his wife.] Attachment
parenting stresses the need for parental sensitivity and responsiveness.
28 Understanding Attachment

Although m u c h of the attachment parenting material is directed toward

parents of very y o u n g babies, there is also advice about school-age children
and adolescents.

Sources, Definitions, Evidence

Attachment parenting advisors refer occasionally to J o h n B o w l b y and
Donald W i n n i c o t t , another British theorist of the 1940s and 1950s. T h e y
also allude to some research on prenatal reactions of babies, and to b e h a v -
iors soon after birth. Although the basic concepts of attachment parenting
are rooted in theory and research, the actual, practical recommendations to
parents do not necessarily share this foundation.
Attachment parenting sources do not define attachment the way w e have
in this book. R a t h e r than applying the term to the child's early emotional
development, it appears to apply to the mother's feelings for her baby. For
example, one attachment parenting W e b site states:

Attachment is a special bond between parent and child; a feeling that

draws you magnet-like to your baby; a relationship that when felt to
its deepest degree causes the mother to feel that the baby is a part of
her Attachment means that a mother and baby are in harmony
with each other. Being in harmony with your baby is one of the
most fulfilling feelings a mother can ever hope to have. (http://www.
askdrsears.com)

H a r m o n y appears to be equated with the mother's sensitive reading of the

baby's cries and signals, and her effective comforting. (Fathers do n o t get
m u c h m e n t i o n in discussions of attachment parenting.)
Attachment parenting advocates also seem to consider attachment to be
apparent in the mother's m o o d and feelings. " Y o u will k n o w w h e n you get
that attached feeling for y o u r baby W h e n y o u r baby cries and y o u r e -
spond with a feeling of rightness about your response, you are well on y o u r
way to b e c o m i n g an attached parent." Mothers are advised to check their
"sensitivity index." T h e y are considered to be " b e c o m i n g attached" if they
b e c o m e increasingly sensitive to the baby. For example, increasing sensi-
tivity may be revealed by the mother's distress w h e n the baby cries, by her
anticipation of crying w h e n the baby's facial expression changes, or by a
sense of the intuitive rightness of responses to the baby.
Attachment parenting advocates thus see attachment as a change in the
parent—particularly the m o t h e r — r a t h e r than in the child. H o w e v e r , the
ultimate goal and desired l o n g - t e r m o u t c o m e is related to the child's devel-
o p m e n t . It is claimed that families w h o use attachment parenting t e c h -
niques are able to shape positive emotional development in the child,
 What Everybody Knows 129

enabling the child to develop trust in others and positive self-esteem. The
child will learn to "bond to persons, not things," have "meaningful attach-
ments with peers," and, in adulthood, achieve a "deep intimacy with a
mate." Other positive outcomes are said to be better discipline, a healthy
conscience, and less need for punishment.
Not surprisingly, there is no systematic evidence basis for attachment-
parenting claims. The assumption that more sensitive mothers guide babies
to better development is well substantiated, but the other statements are
really impossible to test in any controlled way. The attachment parenting
recommendations described below are particularly difficult to confirm
through systematic investigation.

Attachment Parenting Recommendations

Attachment parenting sources emphasize that their recommendations are
tools, rather than absolute requirements, and different choices may be
needed for babies or families with different characteristics. A strict set of
guidelines is not advised; instead attachment parenting counsels individual
choices based on what works.
For parents of young babies, however, there is a program recommended
as an effective way to begin parenthood. Attachment parenting advocates
advise a general approach as well as some specific activities they feel will
initiate a positive family life. (It is notable that some attachment parenting
Web sites include a variety of commercial offerings such as baby carriers,
although these are not presented as essential to good parenting.)

Attachment Parenting and Bonding

Attachment parenting proponents discuss bonding or "birth bonding" as
an important issue for new parents. They define bonding as "the close emo-
tional tie that develops between parent and baby at birth ... really a continu-
ation of the relationship that began during pregnancy." In spite of these
definitions, however, the emphasis is on what we might call bonding activi-
ties rather than demonstrable emotional responses. These bonding activities
are rather specific and focus primarily on the first hour or so after birth.
One bonding suggestion is that parents request a delay in routine proce-
dures, such as the eye drops or ointment used to prevent eye infection. A
second is that the baby should be kept connected by placing it on the
mother's abdomen as soon as possible after birth, pausing only if there is a
need to suction, cut the cord, or do more serious medical interventions.
Putting the baby to the breast at once is also advised, with the mother's
hormonal response to sucking considered an important guide for her
130 Understanding Attachment

mothering behavior. Similarly, touching, stroking, caressing, and skin-

to-skin contact are all recommended as pleasurable and supportive of the
baby's transition to life outside the womb. Attachment parenting advocates
also suggest that talking to and gazing at the baby are bonding activities.
Advocates of attachment parenting believe the parent-child relationship
needs more than a few hours to become established. Rooming-in, the ar-
rangement in which the newborn shares the mother's hospital room, is
suggested as a way to continue the bonding activities that have been de-
scribed, and of course these can go on during the early weeks and months
of the newborn's life.
Wlio Is Bonding? The attachment parenting view of bonding is confused—
and confusing to the reader. Most attachment parenting material stresses
the parent's emotional response to the child, for example, "staring into
your baby's eyes may trigger a rush of beautiful mothering feelings," and
one would expect this emphasis to apply most to the baby's first few days
after birth. However, attachment-parenting materials imply that something
emotional is happening to the baby as a result of bonding activities, even
right after birth. One reason given for delaying prophylactic eye treatment
is that the baby "needs a clear first impression" of the mother (who is also
said to need to see the baby's eyes).
Similarly, it is suggested that the baby needs a person to bond to (or
with) in the first hours after birth, and that the father or someone else can
be bonding while the mother recovers. The attachment parenting Internet
advice says, "One of the saddest sights we see is a newly born, one-hour-
old baby parked all alone in the nursery, busily bonding (with wide-
open, hungry eyes) with [the] plastic sides of her bassinet." The
implication being that there is indeed something the baby needs and does
shortly after birth, and that it does not depend on human beings or on
specific bonding activities. It is difficult to see what emotional reaction to
the bassinet side could occur. But what could the baby be doing, and
how would the outcome be different than if the baby were looking at its
mother? The difficulty in the logic of this explanation seems to stem from
the confusion of parent or baby actions with changes in parent or baby
emotions.

Attachment Parenting and Attachment Tools

Attachment parenting recommendations stress the idea that bonding or
bonding activities early in life are only the beginning of a developing
parent-child relationship. Proponents of attachment parenting recommend
a set of child care techniques that have a long human history, but that are
far from universal in industrialized societies today.
 What Everybody Knows 13

Breastfeeding, not surprisingly, is first among the attachment parenting

recommendations for fostering attachment (and we need to remember that
this group is defining attachment as the mother's sensitivity and responsive-
ness to the baby). Breast milk is described as a diet that promotes brain
development, but the major emphasis is on breastfeeding as an exercise in
babyreading and, therefore, as a cause of increased maternal sensitivity.
As we have noted in earlier chapters, breastfeeding has not been shown
to be a factor in the child's development of attachment (that is, attachment
as it is usually defined). Whether breastfeeding experience causes an in-
crease in the mother's sensitivity, or whether sensitive mothers more often
elect to breastfeed, is a very difficult question to answer. It is certainly true,
however, that breastfeeding mothers must pay attention to their babies'
signals—for example, so they know when to switch from the left to the
right breast. Bottle-feeding mothers may pay attention, but can feed suc-
cessfully without much communicative exchange.
Babywearing is a second technique or tool said to increase a parent's sen-
sitivity to the baby's needs and signals (http://www.parentingweb.com/
ap/babywearing.htm). Although many parents of infants and toddlers carry
their children in back or front packs, attachment parenting advocates advise
babywearing mothers to use a sling, which is a small hammock with a broad
strap that goes around the mother's back and shoulder. A breastfeeding
mother can easily arrange the sling to give the infant access to the nipple,
while the mother stays discreetly covered.
As well as an increase in the mother's sensitivity, other benefits of baby-
wearing are claimed. Although there is good evidence that carrying babies
helps reduce colicky crying, some of the other claims about babywearing
are not only untested, but also quite difficult to investigate. For example,
the practice is said to produce a secure maternal attachment, but it is not
clear what definition of attachment is being used here. Babywearing is sup-
posed to simulate the sensations of pressure, motion, warmth, and security
of the womb. This claim would appear to have some accuracy; however, it
is not at all clear whether this result is desirable or even developmentally
beneficial.
Some other claims about babywearing are questionable and are based
solely on anecdotal evidence or generalization. For example, it is claimed
that neurological development is fostered by the movement stimulation
experienced in a sling. Again, attachment parenting proponents suggest that
toddlers who are carried in slings "initiate separation sooner, and become
more self-reliant"—a difficult statement to substantiate, and one for which
there is no systematic evidence. Co-sleeping is probably the single attach-
ment parenting recommendation most likely to be criticized in the United
States. This may simply involve mother and baby sharing a bed, or a larger,
132 Understanding Attachment

family bed with enough space for both parents and several children.
Attachment parenting advocates do not insist on co-sleeping, but they do
approve it as one of many sleeping arrangements. Concerns that a baby may
be suffocated or lain upon by a parent are rejected in light of evidence that
child deaths of this type are usually associated with a parent's alcohol or
drug intoxication. In addition, some research has shown that young babies'
breathing is regulated by physical contact with adults. Attachment parent-
ing advocates see co-sleeping as advantageous both to the parents, who can
reconnect emotionally with the child at night, and to toddlers, who may
otherwise experience separation anxiety.

What Is the Connection between Attachment

Parenting and a Child's Emotional Development?
Proponents of attachment parenting use terms like bonding, attachment,
and even separation anxiety—all familiar vocabulary from earlier chapters'
discussion of attachment theory. The definitions of the terms are not the
same, however. Are there connections between attachment parenting prac-
tices and the development of attachment, as the term is used in formal at-
tachment theory and in systematic research on emotional development?
Attachment-parenting materials do not do much analysis with respect to
this question. They emphasize the effect their childrearing techniques have
on a variety of desirable developmental outcomes: happier families, better
sleep, less frustration, less need for punishment, and more compassionate
caring behavior among teenagers and young adults.
Some of these outcomes have certainly been related to secure attachment
(as the term is used in most of this book). Attachment, in this sense, in-
cludes goal-corrected partnerships between older children and their par-
ents, with less use of punishment as a natural consequence of the child's
maturation. Secure attachment in childhood is associated with desirable
child care behaviors in adult life.
Parents' sensitivity and responsiveness toward infants are the most im-
portant factors known to produce secure attachment as the child reaches
the toddler stage. If attachment-parenting tools do successfully increase par-
ents' sensitivity, it seems likely that they should be associated with secure
attachment as a characteristic of the child's emotional life. However, people
who write about attachment parenting do not clarify what they think is
happening within the child as emotional development proceeds.
It would be wise to remember that attachment is a robust, developmen-
tal phenomenon. Children are likely to form secure attachments in less than
ideal circumstances. While rough, insensitive, indifferent treatment may
interfere with the growth of attachment, it is doubtful that most families
 What Everybody Knows 133

need to use attachment tools in order to ensure their children's emotional

development. On the other hand, if parents agree that they want to follow
attachment-parenting guidelines, no harm will be done.

Cultural and Religious Issues as

They Relate to Attachment Parenting
There seems to be little evidence to support the use of attachment par-
enting techniques, but a number of parents are very attracted to this ap-
proach. The reason for their attraction may be a matter of values rather than
systematic evidence.
Attachment Parenting and Values in the United States. Attachment parent-
ing employs childrearing techniques that come from ancient tradition.
Breastfeeding and co-sleeping have been used throughout human history as
the most effective ways to care for children in the absence of modern tech-
nology. They are physically and emotionally intimate actions. As such, they
are in conflict with certain values characteristic of the United States since
frontier times, though perhaps in line with modern needs.
Family relationships in the United States have emphasized independence
over dependence, and autonomy over cooperative social skills. Children
have generally received approval for learning to take care of themselves,
refraining from crying or expressing distress, entertaining themselves, and
minding their own business. Early weaning, early sleeping through the
night, and early toilet training have been parental goals. Children who did
not cry for their mothers or avoid strangers, and who were seen as friendly,
outgoing, and self-sufficient, were especially valued.
These values, important for the success of pioneer and immigrant families,
are possibly less important in the modern world. Modern parents want to
connect children with families in order to protect them from the crime,
sexual molestation, and drug abuse that may threaten them outside the
home, as well as from domestic violence within the home. Keeping the
child in the bosom of the family can be a way to protect and guide a child
to adulthood. Attachment parenting connects with this family protectiveness
by stressing the development of mutual affection and cooperation, and the
importance of the parent's emotional engagement with the child and that of
the child with the parent. In this perspective, crying and requests for help are
seen as communications that should be responded to and encouraged.
Religious Issues. The development of the attachment parenting approach
is more understandable if we recognize its connection with Christian
thought—primarily the ideas characteristic of Protestant groups that are
fundamentalist in nature. (Although Roman Catholics and the Protestant
liturgical churches, such as the Lutherans, are certainly interested in
134 Understanding Attachment

childrearing practices, they have not been particularly involved with the
matters we are discussing here.) Readers who look at attachment parenting
web sites will see links to groups described with names such as "Christian
attachment parenting" or "gentle Christian mothers." These groups have
enthusiastically accepted attachment parenting principles as the positive,
child-centered guidelines that lead to the development of close family ties
and the rejection of domestic violence in any form.
The fundamentalist Christian interest in attachment parenting has undoubt-
edly been intensified by the adoption of an alternative-parenting program by
some church groups in the United States. This approach, one that strongly re-
jects attachment-parenting principles, is sometimes referred to as Baby wise, or
Ezzo, parenting. Its guidelines are intensely adult-centered and its techniques
are parent-directed. For example, breastfeeding at the baby's signal (a cardinal
rule of attachment parenting) is replaced with a parent-directed feeding sched-
ule for breastfed, as well as bottle-fed infants. Mothers are instructed to ignore
their emotional impulses to respond to their babies' cues, and instead follow a
recommended program. Rather than babywearing, the Ezzo approach coun-
sels planned playpen time as an important way to develop independence. This
philosophy's primary goal is the child's obedience to parental authority.2
It is possible that attachment parenting would not have grown as it has
without the obvious contrast presented by the Babywise group and the ap-
peals the two programs have made to people who share many, but evi-
dently not all, of its values.

A T T A C H M E N T THERAPY:
A TREATMENT A N D A CHILDREARING PHILOSOPHY
In spite of some differences in vocabulary, the connection between attach-
ment parenting and Bowlby's attachment theory is incompletely realized, but
it is genuine. In this section, however, we will discuss another popular belief
system. Though it claims connection with Bowlby's theory and shares some
definitions and terms, it is almost diametrically opposed in its basic assump-
tions and goals. This system involves a complete system of parenting princi-
ples as well as an intervention intended to be therapeutic. It is commonly
referred to as Attachment Therapy (AT), but attachment parenting and AT
have no connection whatsoever other than the use of the word attachment.

Attachment Therapy: The Intervention

Attachment Therapy, the intervention itself, is a physically intrusive
technique used principally with adopted and foster children. The stated
goal of AT is the transformation of a child's emotional life and the creation
of emotional attachment to the adoptive or foster parent.
 What Everybody Knows 135

Although some aspects of AT go back centuries, the technique, as re-

cently practiced, dates from the 1960s, when it was formulated by Robert
W. Zaslow, a California psychologist whose license was revoked after he
injured a patient. Zaslow developed an intervention he called Z-therapy or
rage-reduction therapy. The basis of Zaslow's technique was physical restraint,
or holding of the patient by a group of five or six adults, while the therapist
in charge prodded the patient in the ribs and under the arms, grabbed the
face, shouted, and tried to force prolonged eye contact. Sessions lasted for
hours and were intended to provoke the patient to wild anger, shouting,
cursing, and tears. Later developments in AT included having the therapist
lie with his full weight on the child patient, as well as covering his or her
mouth with a hand. 3
Zaslow himself believed that this treatment caused emotional attachment
and related it to Bowlby's work. When the Colorado physician Foster
Cline met Zaslow, he worked to bring together Zaslow's approach and
some related ideas in a form he called holding therapy or Attachment Therapy.
Cline's version of the treatment, directed at children and young adoles-
cents, was practiced for ten or fifteen years, but it remained an underground
approach, only rarely mentioned in professional child psychotherapy
groups. Attachment Therapists essentially trained each other; they were
most likely to practice in Western states where there was minimal regula-
tion of mental health practitioners.
Public awareness of AT practices increased sharply with the death of ten-
year-old Candace Newmaker in 2000 during an AT session with two un-
licensed practitioners and staff members. Candace, who had been adopted
at age six in North Carolina, was brought to Colorado for treatment of
Reactive Attachment Disorder (see Chapter 6) by her adoptive mother.
Candace received holding therapy, but this technique was not the direct
cause of her death. On the day when Candace died, head therapist Connell
Watkins (who now uses the name C. J. Cooil) decided to use a different
intervention called rebirthing. Candace was wrapped in a blanket and in-
structed to wiggle out in order that she could be reborn as the true daugh-
ter of her adoptive mother. During her wiggling, the therapists and helpers
pressed down on her rhythmically in simulation of birth contractions.
Although Watkins and her group usually limited rebirthing to a period of a
few minutes, in Candace's case, they kept her wrapped for seventy minutes,
overlooking her cries of distress, sounds of vomiting, and eventual com-
plete silence. When finally unwrapped, the child was dead of asphyxiation.
The two therapists were convicted of child abuse leading to death, and
sentenced to sixteen years in prison.4
Advocates of AT responded to this tragedy in a variety of ways. Some
insisted that Candace must have had an unknown medical condition, and
136 Understanding Attachment

that the death was an unavoidable accident. Others spoke of AT as a gentle,

nurturing process akin to rocking a baby. Many asserted that the children
being treated were saved from lives of violence, and that conventional psy-
chotherapy would make their conditions worse.
A parent-professional organization dedicated to AT advocacy, ATTACh
(Association for Treatment and Training of Attachment in Children), de-
veloped a set of ethical standards intended to prevent injury, and hired a
public relations firm to handle problems related to AT. ATTACh and other
groups presently have over eighty Internet sites that either advocate or offer
AT in some form that involves physical contact; some groups advise that
physical restraint should be carried out by parents rather than by therapists.
It should be pointed out, however, that national organizations of mental
health professionals continue to disapprove of AT. The professional journal
Attachment and Human Development dedicated most of an issue in 2003 to
the rejection of holding approaches, and support of AT among universities
and hospitals is rare indeed.

Attachment Therapy: Parenting Principles

The AT philosophy goes far beyond the use of restraint as an interven-
tion. When a child is in AT treatment, the home and family life are struc-
tured according to related principles. AT-structured experiences may be
provided in the child's own home, or the child may be placed for a time
with AT-trained therapeutic foster parents. Children who are brought for
two-week intensive treatments at AT centers stay in therapeutic foster
homes when not actually in a therapy session.
The leader of AT parenting training has been Nancy Thomas, a self-
designated parent educator and "lay therapist" who is apparently without
relevant formal education. Thomas has presented her approach in a number
of self-published books, as well as in lecture tours, workshops, and videos.5
Thomas' goals for children in AT treatment are that they become "re-
spectful, responsible, and fun to be around." Her recommended first step
toward these goals is German Shepherd training. This does not involve actual
dogs, but works toward having the child automatically and unquestioningly
obey a number of simple commands, just as a well-trained pet would do.
Obedience is achieved by practice and by insistence on correctness in every
detail. For example, the child, when called by name, should not reply
"What?" but must go to the adult and say "Yes, Mom Nancy" (or whoever
is giving the command).
Obedience and control are also established by the practice of strong sitting,
or power sitting, in which the child sits tailor-fashion, hands in lap, silent, un-
moving, looking straight ahead. Power sitting is to be done three times a day,
 What Everybody Knows 137

each time for a n u m b e r of minutes equal to the child's age in years. This does
not seem like a difficult demand until one realizes that the clock does not start
until the child is completely compliant, which may be a matter of hours.
O b e d i e n c e is also exacted t h r o u g h physical means, such as locks and
alarms on doors. Children may not have night-lights in their rooms, even
w h e n extremely anxious. Food is a major form of manipulation, and a child
w h o does not comply or does not show gratitude for a meal may go for
weeks on a diet of cold oatmeal or peanut butter sandwiches. It is notable
that school attendance is t h o u g h t of as a privilege rather than a right, so
children in A T treatment may have little or no break from these conditions
for periods of m o n t h s .
T h o m a s and other A T advocates are particularly concerned that children
must not have information about what is happening to them. T h e y do not
k n o w w h e n , or even if, they will see their parents again, or w h e t h e r they
will be allowed to eat with the family, go to school, or go o n a trip. Adults
must be in complete authority, according to the A T view, and possession
of knowledge is a part of authority.
O n e mental health practitioner, in an area w h e r e A T parenting was m o r e
c o m m o n than A T therapy, gave the following description of what she had
observed:

[A] child in foster care refers to the man living with her mother as 'dad'
and her biological father as 'father' ... and [is] given extra chores for telling
'falsehoods' about who she is related to When the child continues to
refer to the men in her mother's life as dad or father, plus slips up and calls
the foster mother 'mom', drinking 8 ounces ofvinegar while 'strong sitting'
is added to encourage connection between the 'lower attachment centers'
of the brain and 'higher reasoning,' thus eliminating 'false attachments.' 6

O u r description of A T parenting principles w o u l d not be complete

w i t h o u t reference to reparenting activities. T h o m a s and other A T advocates
believe that positive emotional change takes place w h e n caregivers carry
out babying or nurturing activities with children of school age or older.
These activities include cuddling, rocking, bottle-feeding, and feeding of
sweet foods that are usually denied to the child. These nurturing activities
occur only o n the parent's schedule or w h i m , however. T h e child w h o asks
for a h u g will not get one, but must accept a caress w h e n the caregiver
decides to give o n e .

Attachment Therapy: Assumptions and Belief System

Although descriptions of A T intervention and parenting may sound like
r a n d o m cruelty, these practices have a logical foundation in an organized
138 Understanding Attachment

belief system—which, however, is unsubstantiated by empirical evidence.

Few underlying AT assumptions have either an evidence basis or a connec-
tion with any established psychological theory.
Prenatal Emotion. AT proponents assume that emotional connections be-
tween child and mother begin before birth. Not only does the mother have
positive or negative feelings toward the coming child, as we know occurs,
but AT beliefs hold that the unborn child feels an emotional tie to the
mother. This baby's prenatal attachment can be harmed by the mother's
negative feelings, such as those involved in considering abortion. If the
child is given up for adoption, even on the day of birth, he or she experi-
ences grief and rage, and these emotions continue to be active until relieved
by AT intervention. Meanwhile, mourning and other distressing feelings
interfere with the possibility of attachment to loving adoptive parents.
Eye Contact and Catharsis. AT proponents assign an almost magical power
to eye contact between children and adults. Sustained mutual gaze, of the
kind generally seen only between parents and babies or between adult lov-
ers, is considered to be an essential step that causes the child to form an
attachment; it also serves as evidence of an existing attachment.
It is also assumed that negative emotions, once engendered, must be
expressed in order to be reduced in strength; that is, they must undergo
catharsis. Prenatal anger about the mother's rejection, or grief and rage over
separation and adoption, cannot soften with time, but must receive vigor-
ous expression, which will eventually have a neutralizing effect.
The Attachment Cycle. If the child does not experience separation and
adoption, AT proponents think he or she will continue to form a strong
attachment to the parents during the first two years of life. In the first year,
the development of attachment is thought to depend on a repeated cycle in
which the child experiences need, frustration, and rage, and the parents step
in to satisfy the need, causing the child to become more attached to them.
Children who are neglected or who have difficult, uncomfortable medical
problems do not have good experiences of this kind and do not form good
attachments. In the second year, parents strengthen the child's attachment
by setting limits and establishing themselves as authorities. Failure to estab-
lish control in this way interferes with attachment. Attachment is evidenced
by the child's cheerful obedience, according to the AT way of thinking.
Attachment Disorder. When a problem causes difficulties with attachment,
the child develops the symptoms of an Attachment Disorder (AD), consid-
ered by AT proponents to be a more severe form of the Reactive Attachment
Disorder discussed in Chapter 6. AD involves a high level of rage, and has
anger and aggressiveness as its most important components.
Just as attachment is indicated by the child's compliance and obedience,
AD begins with disobedience and lack of affection toward the parents.
 What Everybody Knows 139

If left untreated, according to AT proponents, the disorder will lead to

serious behavior problems, culminating in serial killing by boys and prosti-
tution on the part of girls.
AT proponents offer a questionnaire, the Randolph Attachment Disorder
Questionnaire (RADQ), which is intended to diagnose AD. The R A D Q
is based on parental opinion rather than professional observation, following
the AT principle that any family problems have their source in the child
alone.7
Therapy Principles. The physical restraint treatment described earlier is
intended to evoke rage and cause catharsis through the expression of nega-
tive emotions. AT proponents consider this step essential because they be-
lieve no new attachment can occur until the old rage is gone. Having
disposed of the child's rage, the parent can go on to create attachment by
repeating nurturing actions normally experienced by young infants, such
as bottle-feeding. The new attachment comes about by recapitulating the
child's old experiences, this time in a positive, gratifying form. In addition,
the child needs to experience the adult's complete authority. Some German
authors, who support holding treatment, have proposed the child cannot
feel secure unless it has been made clear that it is the parent who has com-
plete control. 8

An Evidence Basis for AT?

AT is not based on systematic research evidence, but, curiously, its propo-
nents have claimed that it is. AT materials often depend on the implication
that if something sounds like biology or mathematics, it must be true. Some
web sites offer links entitled "Attachment Therapy Works!" or the like.
There are no experimental studies of AT—studies using the randomized
controlled trial design that is considered the gold standard for research.
There are one or two studies using the much less rigorous controlled clini-
cal trial design, comparing children who received AT treatment to those
who did not, but unfortunately these pieces of work are full of confused
variables. It is impossible to know whether emotional changes were due to
the treatment or to other factors. Another complicating factor is the poorly
designed Randolph Attachment Disorder Questionnaire often used for di-
agnostic purposes by AT proponents.

Cultural and Religious Issues

Advocates of AT assume that emotional attachment to the biological
mother is universal among humans and possibly among mammals. However,
there are many cultural differences in relationships between adults and chil-
dren. In Islamic societies, fostering of children is much approved, but
40 Understanding Attachment

adoption, in the Western sense, is not. In traditional Hawaiian culture,

children are assumed to develop a strong emotional attachment to familiar
caregivers, and to have little concern with the biological parent, if he or she
has become unfamiliar.
In the United States, there seems to be a strong sense that adopted people
need to know their roots and need to be able to contact their biological
parents in adulthood or earlier. The unfounded belief that newborns im-
mediately recognize and want their mothers is connected to this idea. Both
of these beliefs lend support to the AT view that adopted children are emo-
tionally distressed and need to undergo a transformation that will make
them the "real" children of their adoptive parents.
Some aspects of AT thought have their parallels in religious and quasi-
religious beliefs. An example is the idea that the unborn baby is aware of
events and particularly, the mother's feelings. A number of religious sys-
tems hold this belief, as do Scientologists. Similarly, the stress on a child's
obedience began with the Old Testament, which (among many other
things) expresses approval of the killing of a child who strikes or curses a
parent. The Puritans of New England shared this attitude; a child's obedi-
ence was considered an essential step toward adult compliance with God's
will, and, therefore, it was critical for parents to tame a willful child.

A T T A C H M E N T CONCEPTS A N D T H E LAW
Legal and judicial thinking today takes into account some ideas about
attachment. Because it takes a long time for new psychological or psychiat-
ric information to be incorporated into law, the ideas that tend to be used
are actually a matter of popular or traditional belief, rather than cutting-
edge research or clinical work. In this section, we will look at some com-
mon legal applications of attachment concepts, and we will see how
congruent these are with current work on early emotional development.
Laws govern many aspects of human relationships, especially those deal-
ing with property, such as marriage, inheritance, and parental rights and
obligations. Until late in the twentieth century, however, neither American
nor British law concerned itself much with the emotional aspects of rela-
tionships. Family law was focused on possession of property, titles, or status,
and the ways these could pass from one person to another. Illegitimate
children had no rights of inheritance, unless the father specifically said so in
his will, and they could not use the family name or inherit a title.
Parental rights followed the same pattern. Until the later part of the
nineteenth century, divorced fathers had a complete right to custody of
their children, just as all family property belonged to men. When formal
adoption laws were established, they too were based on property law.
 What Everybody Knows 141

If one set of parents had a right to the custody of a child, another person
could not also have that right, no matter what might be the complexity of
emotional connections.
Family law began to recognize some non-property issues in the late
nineteenth and early twentieth centuries, with the advent of the "tender
years doctrine." This guideline suggested that infants and young children
would be best in their mothers' custody, because their mothers naturally
knew more about caring for them. The tender years doctrine was not con-
cerned with the child's and the mother's mutual affection, or with the pos-
sible effect of separation on the child.
The door opened to legal consideration of relationships with the first
discussion of the principle of the "best interest of the child."9 This principle
suggested that decisions about child custody should consider a variety of
factors that could potentially affect the child, not just the parents' property
rights to custody. These factors included the speed and method of decision-
making, as well as other concerns, but from the point of view of this book
the relationship between the child and one or more adults was the most
important factor. The principle of the best interest of the child brought the
concept of attachment into the courtroom, and attachment soon became a
focus of family law.

Psychological versus Legal Terminology

Although courts today take a serious interest in children's emotions and
relationships, they discuss these issues in a language different from that of
mental health and child development professionals. For example, the term
"bonding" is often used instead of "attachment."
Another term common in courtroom usage is psychological parent. The
concept of the psychological parent has been used in custody decisions as a
way of differentiating between biological kinship and an emotional rela-
tionship between adult and child, with stress on the importance of the latter
factor. The words psychological parent describe the status of an adult who
has been in a familiar caregiving relationship with a child, and who has car-
ried out the obligations traditionally ascribed to parents. The use of the
term seems to imply a mutually affectionate relationship that includes bond-
ing (parental love) and attachment of the child to the adult.
In recent years, courtroom discussion has used the diagnosis of Reactive
Attachment Disorder, a concept that is defined in the Diagnostic and Statistical
Manual of the American Psychiatric Association (and that was discussed in
Chapter 6).1() However, there have also been cases in which Reactive
Attachment Disorder was either confused or deliberately conflated with
Attachment Disorder, the diagnosis proposed by advocates of Attachment
142 Understanding Attachment

Therapy. As was noted earlier in this chapter, the definition of AD includes

serious tendencies toward violence. This distinction is a subtle one, but a
misunderstanding here may create a real distortion of the facts about a
child. Paradoxically, while psychologists and psychiatrists are denying the
assumption that violence is a necessary hallmark of disorders of attachment,
the courts appear to accept this belief.
The seriousness of this confusion of terms and definitions is evident when
we look at cases of child deaths and injuries where attachment disorders are
claimed to have been a factor in treatment of the child. For example, in
2001, an adopted Russian child, Viktor Matthey, was brought to a New
Jersey hospital in a state of hypothermia and he died the next day. The
adoptive parents, Robert and Brenda Matthey, were brought to trial for the
abusive treatment assumed to have caused Viktor's death. The defense's ar-
guments included the idea that Viktor had behaved unacceptably, had been
out of control, and had harmed himself in a pattern of behavior that indi-
cated an attachment disorder. It was argued that the adoptive parents had
been forced to restrict Viktor's food, tape his mouth, restrain him physically,
and leave him in a cold room, as ways of treating his attachment problems.

Assessments and Custody Recommendations

Because judges and attorneys are rarely experts on child development
issues, they generally seek help when decisions are to be made about child
custody or about termination of parental rights, the legal action that frees a
child for adoption. Expert testimony in these cases is usually given by a
psychologist or psychiatrist who has done a bonding assessment. (In spite of
the fact that the term bonding is unusual in professional discussions of child
development, legal discussions do not often speak of attachment.) As we
will see, the expert's testimony is actually a custody recommendation and
does not necessarily deal directly with attachment.
Assessment of attachment for custody recommendation purposes is not
as easy as is sometimes implied. In earlier chapters, we saw that measure-
ments of attachment must be different for children of different ages, and
that available standardized routines, like the Strange Situation, are inappro-
priate for this purpose, having been developed for research on groups rather
than evaluation of an individual child.
So-called bonding/attachment assessments may be used to develop very
general custody recommendations. These recommendations may also take
into account past family history, particularly whether a given adult was the
child's primary caregiver in the first years of life. Assessments may investi-
gate a parent's knowledge about childrearing techniques, availability to care
for the child, and attitudes about education and health care.
 What Everybody Knows 143

Bonding/attachment assessments may also involve observations of parent

and child together, evaluating the appropriateness of their interaction, and
whether they seem happy to be together. Although such observations could
be valuable (does the parent seem to behave abusively or is the child fright-
ened of or unresponsive to the parent?), in reality the conditions of the
observation may be so artificial that the information gathered becomes al-
most useless. For example, a two-year-old in foster care may be scheduled
for an assessment of attachment to a biological parent. The child may be
picked up by an unfamiliar social services worker in an unfamiliar car, taken
to an unfamiliar building, handed over to another unfamiliar person, and
placed with the parent in an unfamiliar room where they are watched by
another stranger. The child can hardly be expected to behave normally in
these circumstances—and neither can the parent. But it is in such situations
that many observations are made.
A recent suggestion is that bonding/attachment assessments need to look
at reciprocal connectedness.11 Reciprocal connectedness is evidence of the feel-
ings the parent and child have for each other, feelings that are expressed in
different ways toward and by children of different ages. Expressions of feel-
ing may range from eye contact and affectionate touching (parent and
young child), to cooperation in teaching and play (parent and school-age
child), to cooperation in setting rules and limits (parent and adolescent).
The idea of reciprocal connectedness seems to be based on Bowlby's prin-
ciple that goal-corrected partnership is an aspect of attachment.

Attachment Issues and Legal Decisions

Public attention was brought to the role of relationships in custody deci-
sions by two cases: those of "Baby M " and of "Baby Jessica." In each case,
parents' wishes took precedence over considerations of the child's emo-
tional attachment, but the attachment question was clearly present.
In the 1986 Baby M case, the child's biological mother, Mary Beth
Whitehead, had entered into a contract with a childless couple. She agreed
to be inseminated with the husband's sperm, to act as a gestational surro-
gate or surrogate mother, and to surrender the child for adoption by the
couple. After the little girl's birth, however, Mrs. Whitehead was reluctant
to give up the child; she delayed for some months, even disappearing with
the baby at one point. The childless couple brought a lawsuit asking for
compliance with the original agreement and eventually won custody of
Baby M. The court's decision was based on the competing rights of Mrs.
Whitehead as the biological mother, and the rights of the couple under
the contractual agreement. Baby M was approaching an age where attach-
ment to a familiar caregiver was an issue, and although child development
144 Understanding Attachment

professionals pointed this out, concern for this issue seems to have been
absent in court.
In the 1993 case of Baby Jessica, the public's attention was captured by
the distress of a toddler and her adoptive parents as they became caught up
in a net of contradictory state laws and human errors. Jessica's unmarried
biological mother had named the wrong man as the baby's birthfather. The
actual biological father was unaware of the baby's birth and was never asked
to voluntarily give up his parental rights—an essential step for a clear adop-
tion process. Jessica was adopted and reared for two years by a couple in a
state that neighbored the one where she was born, the two states having
different laws about paternal rights in adoption.
When Jessica's birthfather became aware of the child's existence, he un-
dertook legal proceedings to assert his parental rights and take custody of
the child, whom he had actually never seen. The legal battle again con-
cerned the opposing rights of the birthfather and the adoptive parents, with
little or no concern about the impact of a decision on Jessica. The birth-
father won, and the adoption was disrupted. Jessica was taken screaming
from her adoptive mother's arms and given to the birthfather.
Cases like these, together with the progress of the best-interest principle,
have encouraged the consideration of the child's emotional needs as one of
the factors in child custody decisions. But, as we shall see, legal decisions
have made a rather unsystematic use of this factor.

Child Custody Decisions and Attachment Issues

Looking at changes in consideration of relationship concerns in child
custody decisions, we can see a gradual evolution of thinking, but there
have been many inconsistencies and changes of direction. A review of cases
in the state of New Jersey shows some changes that have occurred.
Sorrentino v. Family and Children's Society of Elizabeth, 1976. In this case,
an unmarried sixteen-year-old gave birth, but had no home to take her
baby to, so she placed the child for thirty days' temporary foster care. She
decided to keep the baby, but was talked into agreeing to adoption. Over
the next six months, she twice asked for the child to be returned. She mar-
ried the baby's father and began a suit for custody of the child when the
baby was fourteen months old. The court ordered a hearing:

fit askedj whether transferring the custody of the child to plaintiffs

at that time will raise the probability of serious harm to the child.
Plaintiffs, who seek to change the status quo and to dislodge the child
from the only real home she has ever known, will have the burden
of proving by a preponderance of the credible evidence that the
 What Everybody Knows 145

potentiality for serious psychological harm accompanying or resulting

from such a move will not become a reality.

This decision obviously took into account the emotional development of

the toddler and gave it as much importance as the parents' rights.
Sees v. Baber, 1977. In this case, an unmarried mother gave her child up for
adoption when the baby was three days old. She changed her mind two days
later, and after about a month, initiated legal proceedings to regain custody.
The case arrived at the New Jersey Supreme Court when the child was about
a year old, a time when she could be expected to be at a peak in her concern
about separation. The court refused to apply any principles about attach-
ment, and gave two reasons for this: one, the court considered one year too
short a time to create a psychological bond, and two, the child was too young
for psychological evaluation. This decision completely ignored concerns
about emotional development and about the impact of separation on a tod-
dler, as well as failing to take into account the importance of time at this age,
a point particularly stressed by the principle of the best interest of the child.
Hoy v. Willis, 1978. In this case eighteen-month-old twins, a boy and a
girl, had been separated from each other at the time of the parents' separa-
tion in 1974, and the two had been placed for several weeks at the homes
of two paternal aunts. A little later, one was sent to the home of one pater-
nal uncle, the other to the home of a paternal aunt. The mother visited the
boy three times in two years. In 1976 she took the girl to live with her.
The paternal grandmother asked for custody of the boy, and this was
ordered, but there was also an order for examination and testing to see
whether removal from the aunt's home would harm the child's psycho-
logical wellbeing. By this time the child was about six years old.
A psychiatrist testified that the aunt had become the boy's psychological
parent and that the trauma of separation would create regression, depres-
sion, anxiety, and anger, conditions that would take at least seven years to
resolve and might not resolve at all. Asked by the trial judge whether he
would give the same opinion if the child had been kidnapped, the psychia-
trist said he would—whereupon the judge declared that he could not ac-
cept that line of reasoning and ordered the child returned to his mother.
The Appeals Court, however, reversed the decision, saying, "It is clear to
us that the trial judge failed to recognize and apply present-day concepts of
psychological parentage in resolving the custody issue before him." This
case is an excellent example of the inconsistent use of attachment principles
by courts at that time.
Zack v. Fiebert, 1989. The ages of the children in this important case are
not clear, but they were apparently in early adolescence at the time of the
decision. The mother of the two children had divorced and was remarried
146 Understanding Attachment

to a second husband; he had adopted the children. She became seriously ill
and died in 1987; the children spent m u c h of her illness with their maternal
grandparents, but because of a quarrel saw little of t h e m after the funeral.
T h e adoptive father remarried and the children lived with him, his n e w
wife, and her children. T h e maternal grandparents complained that the fa-
ther was unfit, and they petitioned for m o r e visitation on the grounds of
their relationship with the children. T h e court refused and defined a psy-
chological parent—in this case the adoptive father—as one whose removal
from the child's life w o u l d cause the child severe psychological harm. T h e
decision c o m m e n t e d , " [ N ] o t even the most liberal reading of the [grand-
parents'] pleadings w o u l d support the conclusion that they stand in a
parent-child relationship with their grandchildren."
T h e Zack v. Fiebert decision emphasized the importance of the psycho-
logical parent and treated biological kinship as a relatively trivial factor, thus
affirming the serious importance of attachment issues. H o w e v e r , the logic of
the decision assumed that attachment concerns w o u l d be the same for y o u n g
teenagers as they are for infants and toddlers. Whereas Sees v. Baber assumed
that a toddler was too y o u n g to be badly affected by separation, Zack v.
Fiebert assumed that teenagers were still y o u n g e n o u g h — i n each case, a c o n -
clusion contrary to what is k n o w n about attachment and separation.
B.F. v. DYFS, 1997. T h e m o t h e r in this case placed her y o u n g child (age
unstated) in temporary foster care and eighteen m o n t h s later contacted the
N e w Jersey Division of Y o u t h and Family Services to request reunion. T h e
child was in a pre-adoptive h o m e , and the Division decided to bring an
action for her guardianship, preparatory to adoption by the family. H o w e v e r ,
it did not do so for another nine months, at w h i c h time the child had b e e n
separated from the biological m o t h e r for twenty-seven months. T h e m o t h e r
was not allowed to visit during this time.
T h e court returned the little girl to her biological mother's custody, o n
the basis of the following reasoning:

[B]ecause the psychological bonding between [the child] and her pre-
adoptive family during the period following the [Division of Youth
and Family Services'] refusal to allow visitation resulted from the
improper denial of visitation rights, consideration of that bonding by
the trial court would similarly have been improper.

Thus, in spite of precedents for consideration of attachment issues, the

court apparently decided that due process t r u m p e d the best interest of the
child. T h e B. F. v. DYFS decision contributed to the record of inconsistent
judicial attention to parent-child relationships.
V. C. v. M.J. B., 1999. In an example of a n e w type of case, three-year-
old twins were the subjects of a second-parent adoption suit. V. C . and
 What Everybody Knows 147

M. J. B., a same-sex couple, had shared parenting from the time the chil-
dren had been born to one of them until the women separated when the
twins were two years old. The biological mother had custody of the chil-
dren, and her former partner requested visitation rights on the grounds of
having bonded with the children. This was refused by the trial judge, who
said: "While there is evidence ... of the plaintiff enjoying a bonded rela-
tionship with the children, the plaintiff has failed to establish the relation-
ship to have risen to the level of in toco parentis." This judge apparently
considered the possibility of degrees of attachment or psychological parent-
hood, an idea that had not appeared in decisions before.
The attorney representing the biological mother argued that the three-
year-old children were too young to be affected by separation. "The fact that
the plaintiff is a parental figure to the children and has 'bonded' with them is
not a sufficient basis [T]hese are very young children ... and children
form bonds with people such as grandparents, babysitters, [and] nannies."
The Appeals Court did permit visitation by the former partner and re-
ferred back to the Zack decision, focusing on the history of the adult-child
relationship rather than on an assessment of the children's attachment be-
haviors. "Psychological parenthood is a finding based upon the role the
person historically played in the child's life. Neither optimistic not pessi-
mistic predictions of future harm that would result from ending that role
can logically define the role itself."
The discussion of V. C. v. M.J. B., thus, included the beliefs that attach-
ments can be rank-ordered, that three-year-olds are too young to have real
attachments, and that the nature of an adult-child relationship is understood
on the basis of past events only. Relationship concepts and developmental
changes played confused roles in this decision.
Watkins v. Nelson, 1999. In this case, a seventeen-year-old, unmarried
mother was killed in a car accident twelve days after the birth of her child. The
maternal grandparents cared for the baby, asked to be appointed her guardians,
and contested the paternity claims of the biological father, who visited regu-
larly. When the baby was nineteen months old, a trial judge gave custody to
the grandparents, citing the principle of the best interest of the child.
The child, who had some physical problems, was said to be "develop-
mentally unable" to show attachment. The grandparents were given cus-
tody not on the basis of psychological parenthood but because of their
advocacy and involvement with her treatment and their experience with
learning disabilities. On appeal, a dissenting opinion argued:

I perceive no great potential for serious harm to [the child] in making

the transition [to the father's custody] Plaintiff has maintained a
relationship with his daughter since her birth The change in
48 Understanding Attachment

custody may, for a while, disturb [the child's] peace of mind ... [but]
that will soon pass, considering her extreme youth.

This set of arguments again revealed a very mixed bag of beliefs about
attachment and separation. Psychological parenthood was indeed seen as
one of many important factors, and the long-term relationship with the
father was stressed. But once again there was a failure to recognize that the
extreme youth of the child made her more, rather than less, vulnerable to
changes in relationships.

CONCLUSION
The concept of emotional attachment is not limited to professionals in
child development and mental health work, but is part of popular thinking.
Unfortunately, nonprofessional use of attachment concepts has not neces-
sarily followed research findings or advances in theory. Popular advice
about early emotional development often distorts important principles, or
it asserts recommendations that have no basis in evidence. Such recom-
mendations given to adoptive parents have resulted in problems, including
serious injuries and some deaths.
Judges and attorneys who deal with child abuse cases generally hear argu-
ments related to popular views of attachment rather than to research or
established theory. These views may involve unsubstantiated statements
about Reactive Attachment Disorder and necessary treatment.
Over the last thirty years, judges and attorneys dealing with adoption and
child custody cases have begun to consider the role of attachment and
separation issues in determining the well-being of the child. Although these
factors are receiving far more consideration than in the distant past, they are
still applied in inconsistent and contradictory ways. The developmental
course of attachment seems to be imperfectly understood in the legal
world.
 chapter 8

New Directions:
Parents, Children, and
Attachment Concepts

T
his book has examined a wealth of observations and research di-
rected toward understanding attachment. It may be hard to imagine
that there is still much work to be done in this area. Looking at
recent publications, however, it is plain that researchers are beginning to
take some very new approaches to the study of emotional development and
family relationships.
This final chapter will begin by summarizing the facts that have emerged
from John Bowlby's original work. We will then consider the possibility
that Bowlby's views may no longer fit the attachment experiences of the
average child the way we used to think they did. If children's experiences
in their families have altered since Bowlby's time, it may be necessary to
think of attachment in a different way, ideally a way that would work for
many different family patterns. The chapter will conclude with an exami-
nation of some new ideas that show promise for an increased understanding
of emotional and social life.

A N S W E R I N G BASIC QUESTIONS A B O U T A T T A C H M E N T
With fifty years of research and clinical work to draw on, we can sum-
marize what is presently known about attachment. A good way to do this
is to answer a set of questions posed by the well-known attachment re-
searcher Everett Waters. The questions are a useful way to share the basic
facts about attachment and other aspects of development. 1
150 Understanding Attachment

What Develops?
What characteristics of a person's emotional life change with age? Emo-
tions, behaviors, and thoughts about other people all change as attachment
develops. The newborn baby has engagement behaviors that attract and in-
terest caregivers, but it is not until about eight months of age that we see
children trying to stay near familiar people, expressing distress about separa-
tion, or showing pleasure and relief when reunited with a familiar person. In
the next year or so, immediate, emotional responses to separation become
less intense, and behavior begins to include negotiation of separation. During
the school years, children's attachment to their parents presents as an affec-
tionate partnership. During adolescence, other people, especially romantic
partners, begin to be the focus of emotion and behavior. In adulthood, indi-
viduals show intense emotion and characteristic behavior toward their own
children. The whole process seems to be related to the development of an
internal working model of social relationships—a set of feelings, behaviors,
beliefs, and expectations about interactions with other human beings.

What Are the Speed and Pattern of Development?

Do attachment behaviors and feelings change quickly or slowly, predict-
ably or unpredictably? Some aspects of attachment must be in progress
before the age of eight months, but the changes that occur at about that age
appear to emerge rapidly and abruptly. From that point on, however, most
changes in attachment behaviors and emotions are slow and gradual, and
these alterations continue through adulthood and even into later life.
Teenagers and adults do show some sudden changes, when they fall in love,
or develop strong feelings soon after a baby is born or adopted.

What Are the Mechanisms of Development?

What factors make attachment develop? There are really only two cate-
gories of mechanisms, or causes, of developmental change. One of the pos-
sible causes is the genetic make-up of the group or of the individual.
Because human beings—as well as our primate relatives—almost always
show attachment behaviors and emotions if they have experienced a rea-
sonably consistent and responsive social environment, we assume that the
human genome carries information that makes it easy for attachment to
occur. However, there is some research that suggests individual differences
in children's attachment behavior are not caused by genetic factors.2
The occurrence of attachment, and the type of attachment emotions
and behavior that emerge, seem to depend on social experiences, especially
the repeated experiences that make others familiar. Consistent, repeated
 New Directions 151

experiences with a sensitive, responsive adult are the foundation for secure
attachment. The continued development of the internal working model of
social relationships occurs because of an increasing variety and complexity
of social interactions, helped along by language communication and by
maturing thought patterns.
Experience is such an important mechanism for attachment that psycho-
therapy or a new family situation can help to alter the internal working
model years after the initial attachment experiences have occurred. Some
adults manage to pull themselves up by their own bootstraps and to alter
their own thinking toward a more positive and secure view of social rela-
tions. Some, however, are so strongly affected by their families' past histo-
ries of attachment and separation, it is difficult for them to change.

Are There Normal Individual

Differences in the Development of Attachment?
Should we expect social and emotional development to be about the same
in most normal individuals? There are certainly normal individual differ-
ences in young children's attachment behavior and emotional expression.
These differences seem to be connected with the child's temperament, or
a biologically determined individuality of response to the environment.
However, whereas most children are considered securely attached, there
are other variations that may not be exactly normal or desirable. There are
significant numbers of insecurely attached toddlers, and a small number
whose attachment is considered disorganized. Only the latter group is
thought of as definitely outside the normal range. Older children and adults
have individual variations in their internal working models of social rela-
tions because of their ongoing social experiences.

Are There Population Differences in Attachment?

Are there groups of people who become strongly attached to familiar
people, or other groups who do not really care who their companions are?
Groups of people who have shared traumatic experiences and separations
show different attachment emotions and behaviors than those with a good
attachment history. For example, families of Holocaust survivors seem to
pass separation concerns down the generations. 3 Toddlers who survived
the African rebel wars of the last decade were often stunned and unrespon-
sive; older children who were kidnapped and forced to kill as child soldiers
also have had unusual emotional responses to others. Even when brought
up in peaceful situations, groups of children can show emotional differ-
ences derived from their cultural backgrounds. Boys, however, are not
different from girls, and there do not seem to be population differences
152 Understanding Attachment

based on differences in a group's genetic make-up (as opposed to cultural

differences).

A C H A N G I N G W O R L D : CULTURAL A N D SOCIAL
ALTERATIONS T H A T C A N INFLUENCE A T T A C H M E N T
A wealth of information about attachment seems to exist today. But will
it be true in a few years? Can we expect changes in our understanding of
attachment? Can we expect events that might actually change the average
child's emotional development?
Of course, changes in our thinking and new research are always possible—
and to be hoped for. But it is also possible that the way people live may
alter, and with it, children's emotional experiences and development.

Mixing Cultures
Attachment theory and research have up until now been based largely on
the habits and beliefs of Western European and North American people.
Although there has been serious cross-cultural research, this research was
formulated on the basis of Western beliefs about the importance of attach-
ment experiences—as indeed any research must be formulated on some
foundation of beliefs and values.
As movement from one cultural setting to another becomes more fre-
quent, and as non-Westerners more often live in the United States and
European countries, the cultural contrasts among beliefs about attachment
become more evident. In a recent case, a Chinese couple, Shaoqiang and
Quin Luo He, who lived in Memphis, learned that a judge had terminated
their parental rights to their five-year-old daughter, Anna Mae. Mr. and
Mrs. He had initially asked to place Anna Mae in temporary foster care at
her birth, a time when they were beset by a variety of legal and financial
troubles. When Anna Mae was three months old, they agreed to temporary
custody by the foster parents. The Hes visited Anna Mae almost every
week for the next two years. They then asked to have her returned. They
got into an altercation with the foster parents, were ordered to stay away,
and did so for several months. The couple's compliance, far from creating
the good will they desired, was interpreted as abandonment.
The American and European view of this situation assumes that legal
parenthood of a child can be decided on the basis of the parents' actions,
and that placement of a child for care by another person or family is on the
verge of abandonment. For Chinese families, a more likely assumption is
that young children can be cared for others, sometimes for years, without
breaking the parent-child connection. For Chinese children in families that
 New Directions 153

have immigrated to the United States, being sent away may be a normative
experience, an experience which many other people have and which the
individual regards as an acceptable, though uncomfortable part of life. As
the separated child grows up, he or she adds this understanding to the in-
ternal working model of social relations; non-parental care does not mean
rejection or lack of parental love. This internal working model is, of course,
different from what might be found in a highly assimilated Chinese family
in the United States.
People of American culture tend to believe in some prenatal or genetic
connection between mother and child, but they simultaneously accept the
idea that parents earn their rights by showing that they care about the child.
This genetic assumption stresses that knowledge of the real father and the
real mother are important, even when these people may be thought to have
forfeited their right to the child. The belief that an adopted child longs for
the chance for reunion has produced laws that allow for open adoption and
for the availability of information to adoptees when they reach adulthood.
Another type of legal change has been the gradual alteration in the cultural
view of pregnancy outside marriage. Whereas women once feared un-
wanted pregnancy, birth, and adoption as potential sources of great social
disapproval, precluding later marriage, there is far less secrecy today. There
seems to be a social movement to revise our ideas about the parent-child
relationship, including the belief that the early relationship is permanent
and the assumption that more than one kind of attachment can exist.
It is difficult to know whether changing or differing beliefs about attach-
ment lead to different practices. If they do, however, the eventual outcome
may be a change in common attachment patterns and internal working
models of social relationships.

Changing Family Patterns

Whether or not altered beliefs are involved, there is much excellent re-
search to show that social and economic changes result in changed family
patterns and in the changed experiences of children in those families.4 Some
social factors that have transformed over the last several decades have had
the potential for altering both children's attachment experiences and adults'
beliefs about attachment.
Divorce, remarriage, and the creation of blended families form a con-
stellation of factors with implications for attachment. The issue is not so
much the experience of emotional loss, as the development of an internal
working model of emotional equivalency, where one father, mother, brother,
or sister is expected to be interchangeable with another. A father departs
from the family or changes his role dramatically; a stepfather is slotted into
154 Understanding Attachment

the empty spot, and life is supposed to go on much as usual. Stepbrothers

and stepsisters enter the picture, and the children are expected to come to
terms with each other rather rapidly. The pattern associated with these fam-
ily relationships becomes more like that of dating than of traditional mar-
riage, especially if the child experiences several divorces and remarriages, as
many do.
Another relevant family change involves the rapidly increasing frequency
of births to, or adoptions by, same-sex couples. The resulting families have
a variety of biological connections, as opposed to the single pattern for a
heterosexual couple with non-adopted children; whether the biological
relationships make any difference to the children probably depends entirely
on whether the parents think they do (and, as we saw earlier, they may
think so).
Lesbian couples have the option for one, the other, or both to arrange to
become pregnant. (One such couple each gave birth, after being insemi-
nated with sperm from the same donor, so that, as one of the mothers put
it, "Everybody's related to somebody.") The degrees of biological relation-
ship may be different for other same-sex couples, and obviously male cou-
ples can donate sperm, but they must have a gestational surrogate to carry
the baby. (These couples may use a mixture of semen for insemination, so
the real father is not known.) All these families can develop attachment
through their social interactions, but it is not known whether their beliefs
about genetic connectedness make any difference.
One possible disturbing factor for same-sex couples and their children is
the fear of separation. Laws about parental rights have been based on a bio-
logical relationship or on adoption by a person or couple following termi-
nation of the parental rights of another person or couple. Marriage has been
the only circumstance where one parent's rights could be terminated and
transferred to the new spouse of the other parent. These laws do not work
well with the same-sex couple's family. For women partners, only one
gives birth to a particular child; since the women are not married under
most current state laws, the other partner cannot adopt and share parent-
hood of the child.
In many states, this situation has cast a shadow on same-sex couples'
family relationships, although it is not known whether attachment pro-
cesses have been affected. The partner who is not the birthmother or father
knows that a disruption in the adult relationship might lead to a complete
rupture of the relationships with the children. This partner would not have
the rights to visitation or joint custody enjoyed by heterosexual couples.
Death of the birthmother or father could also have a disastrous effect, po-
tentially giving grandparents custody of the children, whom they may have
never seen.
 New Directions 155

Changes in laws about same-sex marriage can ameliorate this problem

and its potential for disturbance of developing parent-child relationships.
Meanwhile, a number of states have permitted second-parent adoption, a
legal process that allows the non-birthmother or father to establish rights to
contact with the children whom he or she may care for and love as much
as his or her partner does.

Changes in Fertility and Adoption Patterns

Attachment processes take place within a broad social and economic
context. Changes throughout the society can influence family life and thus
have a potential for changing attachment patterns.
Delayed Childbearing. As the status of women and their options for pro-
fessional education have advanced in the United States, the tendency to
delay childbearing has also increased. Women, who would have had chil-
dren in their twenties in the past, are now putting off pregnancy until the
late thirties or even forties. But, heard by their owners or not, biological
clocks do continue to tick, and reproductive cycles happen less regularly as
the years pass. Not only are these older, potential mothers more likely to
conceive a child with serious disabilities, such as Down Syndrome, they are
a good deal less likely than younger women to conceive.
Although we do not know exactly what happens, it seems possible that
women who experience fertility problems will also have some changes in
their internal working models of social relationships, especially the aspects
that have to do with the woman becoming a mother or with the place of a
child in her life. As we saw earlier in this book, the mother's memories of
her own attachment experiences appear to affect her child; it is possible that
the grief and anxiety of infertility can also do so."'
Of course, there have always been infertile couples, and history re-
cords many a queen being "put away" for her failure to produce an heir.
The modern situation is a bit different, however. We seem to expect to
be able to control our fertility. We turn it off; we ought to be able to
turn it on again. Our preference is for a small number of children, with
whom we expect to have intense emotional relationships. The vagaries
of Assisted Reproductive Technology (ART) involve the practice of
fertilizing a number of ova, implanting some, and then possibly elimi-
nating some of the implants to assure the success of others; the couple
that experiences this must decide to destroy some potential children for
the sake of others. Finally, we see women today choosing to become
pregnant by A R T while unmarried, a step that involves a different
model of relationships than in the past. All of these factors can create a
perspective on attachment that is different from the historical view, and
156 Understanding Attachment

it may be that the different perspective can itself influence how attach-
ment develops.
Adoptive Families. At the time when the Supreme Court legalized abor-
tion with the Roe v. Wade decision, it was thought by some that there
might soon be fewer children to adopt. There has been some truth to this
prediction, but, as so often happens, the story has not played out exactly as
expected.
In the United States, it is not unusual for a married woman to be un-
happy when she finds out she is pregnant. Some give birth and some abort,
but virtually none surrender the children for adoption. In the past, an un-
planned pregnancy in an unmarried woman, especially a teenager, was of-
ten terminated by abortion, but adoption was also an option. About half
the girls aborted, when the procedure was illegal; 45 percent had the baby
adopted; only 5 percent kept custody of the child, usually with the help of
their parents. Today, about half of pregnant, unmarried girls abort; 45
percent retain custody of the baby, and 5 percent agree to adoption.
Does this mean that there are now many fewer children in need of
adoptive homes? No, unfortunately, this is not exactly the case. But it does
mean there are few young infants available for adoption in the United
States. The attempts of young, poorly educated mothers to care for their
children too often end in neglect and abuse, followed by later foster care
placement. Time passes, the mother makes more efforts, there is more
trouble, and eventually termination of parental rights occurs. Now the chil-
dren are free for adoption—but they are three, four, five years, or older,
bear the scars of their chaotic emotional lives, and often have special medi-
cal and educational needs. To create good emotional relationships in these
later adoptions is much more difficult than to deal with a young baby, who
often has charming engagement behaviors and readily responds to affection.
It is no wonder that many late-adopting parents are preoccupied with at-
tachment issues.
In the wake of these social changes, a foreign adoption industry has
grown up. There are many expenses and problems associated with foreign
adoption, certainly, but foreign-adopting parents have no worries about
whether a mother's parental rights have been properly terminated or
whether a birthfather will suddenly appear, as happened in the Baby Jessica
case described in Chapter 7. Nevertheless, parents who adopt foreign chil-
dren may be struck by the artificiality of the situation, and they may won-
der how they can possibly become a real family.
The social changes outlined in this section have powerful implica-
tions for people's ideas about attachment. These changes may well encour-
age the view that attachment is a job to be done, rather than the natural
consequence of growing social interactions and experience between parent
 New Directions 157

and child. Parents adopting under difficult circumstances may even fear
that attachment will not take place and that the consequences may be
terrible.
Infant Day Care and the Day Care Wars. Concerns about attachment have
been raised by the fact that many children enter day care while still young
infants. It is reasonable to ask whether spending many hours each week
away from their parents can negatively affect children's attachment. To put
this issue in context, however, we should note that non-parental care is a
real part of the history of our species. Human parents love their babies, but
they have always wanted some respite from taking care of them. In any
traditional culture, it makes perfect sense that a young, healthy, capable
woman should spend her time in the fields, at the market, or doing weav-
ing or whatever she can to care for her family's needs; baby-minding may
be done by little girls or the elderly aunt who can't walk far anyway.
Babies have thriven in that sort of non-parental care and have done so
for most of human history. Why, then, are we concerned about day care
arrangements? The fact is, of course, they are very different from that tra-
ditional care pattern.
Babies cared for by little girls or by elderly aunts or grandmothers had
intimate contacts with their mothers many times a day. They had to, be-
cause in most traditional cultures, breastfeeding went on until the age of
eighteen months or two years. The point here is not the breastfeeding itself,
but the occurrence of real mutual attention between mother and child at
frequent intervals. Nursing babies in such societies would also have slept
with their mothers and had frequent interactions during the night.
Traditional non-parental care also gave the baby the same familiar care-
giver every day. A particular child nurse or older relative had the job, and
she was expected to know the baby well and to be able to respond to all its
needs, except breastfeeding. Child nurses were admonished not to let the
baby cry, and they developed ways of soothing, comforting, and engaging
a fretful baby. An important aspect of traditional child care, generally, was
the existence of one caregiver per infant or toddler.
Modern infant day care stands in some contrast to traditional non-parental
care. Although many day care centers would be willing to cooperate with
breastfeeding mothers, few mothers have time to make the journey from
their workplace, nurse the baby in an unhurried and interactive way, and
return to work. Neither, of course, do bottle-feeding mothers have this
chance for frequent workday interactions with their babies. Frequent
mother-baby contact during the working day is just not an option in the
modern infant day care pattern.
Whether or not an infant in day care has a consistent caregiver de-
pends on circumstances. When an infant is in family day care—a more or
158 Understanding Attachment

less informal arrangement in the caregiver's home—there is one caregiving

adult, and she may continue as the consistent caregiver for years. (Or, of
course, she may for her own reasons go out of business, forcing the baby's
family to search for a new caregiver.) In center-based day care—the form
familiar to most people—where hired caregivers work with a group of
children, a baby may or may not have a consistent caregiver. In a small
number of high quality day care centers, an assigned caregiver takes respon-
sibility for no more than three young children and is alone in a room with
them. In more common situations, the ratio of children to caregivers is
larger; there are floating caregivers who attend to whichever baby needs
them at a given time, and many children and caregivers may be together in
a large room.
Infants in day care are exposed to two additional, problematic factors.
First, day care providers work shifts of eight hours or less, meaning most
babies experience a daily changeover of caregivers because they are at the
center when the shift changes. Second, infants in day care are likely to ex-
perience abrupt changes of caregiver when familiar people leave the job.
Staff turnover is a major problem in center-based day care, largely because
pay and status are low even though the work is very hard work.
If we consider what we know about attachment, we can see that
traditional non-parental care might be expected to have a different effect
on emotional development than modern infant day care. The ideal situation
for attachment, as far as we know, involves a small number of consistent
caregivers who play and socialize with the baby and who are sensitive and
responsive to the baby's needs (food, comfort, or social fun). Consistent
and responsive care allows the baby to learn to expect good things from
familiar people, building trust and the beginnings of a secure attachment
and positive internal working model of social relationships. From the adult's
viewpoint, consistent involvement with one or two infants allows for a real
understanding of each child's needs and ways of communicating. This un-
derstanding allows the adult to be more sensitive to the baby's cues and
therefore to respond more effectively. As a result, the adult feels gratified by
his or her success, takes pride and pleasure in the child, and becomes more
deeply engaged in the relationship. The ideal attachment situation creates a
benign circle in which adult and child are progressively more involved with
each other through the first year or two of life. Traditional nonparental care
made this benign circle easy to enter.
Modern infant day care, on the other hand, may or may not make the
benign circle possible. A family day care provider is very likely to have
several other children to care for, although she may have time to get to
know and be sensitive to an individual baby. A high-quality day care center
will have providers caring for several babies, but may have an emphasis on
 New Directions 159

relationship building, a point that will be discussed later in this chapter. At

the other end of the spectrum, relationships between a baby and caregivers
may be tenuous when caregivers have too many children to work with or
they are not assigned to particular babies, and when staff turnover is too
rapid to allow babies and adults to learn about each other. When parents are
also overwhelmed and time at home is rushed so that everything gets done,
the day care infant may experience very little in the way of the sensitive,
responsive social interaction that is linked to attachment.
It is no wonder, then, that the role of infant day care in early emotional
development has been the subject of much discussion among students of
child development and mental health. Disagreements on this subject at
one point reached a pitch sometimes referred to as the day care wars.
Research in the 1980s and 1990s suggested that children with more expe-
rience of day care in infancy were more likely to show insecure attach-
ment, and that the combination of insecure attachment and more time
spent in day care was associated with more negative emotion and less ma-
ture play with toys.6 These results, however, were difficult to interpret. 7
For example, mothers of insecurely attached toddlers, who expressed neg-
ative emotion and did not play well, might be more inclined to put their
children in day care for extended periods. Mothers who felt incompetent
and ungratified by interactions with their children might well feel that
someone else could care for the child as well as they could—perhaps even
better.
Despite the day care wars, economic and other social pressures have
meant infant day care is here to stay. There is, however, a very positive
outcome of the battle, that being a much-increased emphasis on the quality
of day care, particularly in the attention paid to social and emotional needs.
Many initiatives have worked toward improvement of young children's
attachment experiences in day care, and although low-quality day care is
still common, the average experience of children in day care today may be
different than it was at the height of the day care wars. This is a most desir-
able outcome, but one that adds confusion to our ability to understand the
effect of day care on attachment.

CHANGING PRACTICE, CHANGING THEORY:

NEW DIRECTIONS IN CHILD DEVELOPMENT AND
MENTAL HEALTH PERSPECTIVES ON ATTACHMENT
As more is understood about early emotional development, both prac-
tice and theory tend to change. These changes may be providing children
with a somewhat different set of experiences than were common either
during the day care wars or in Bowlby's day.
160 Understanding Attachment

Relationship-Based Practice
Current work with parents and children emphasizes the need to form
and support good emotional relationships. This includes relationships be-
tween parents and children, those between day care providers and children,
and the relationships of parent educators and mental health professionals to
parents, children, and caregivers.
Cultural Issues in Working with Relationships. Awareness of cultural fac-
tors and the role they play in internal working models of relationships is a
major factor in working with families. A simple example involves the use
of eye contact—mutual gaze—as a method of emotional communication.
Americans of European background often seek or even demand eye con-
tact with a child, as a means of play and affection, but also as a way to exert
authority and demand obedience. "Look at me, Jimmy" is a way to over-
come the child's resistance and get compliance, and is frequently used by
white mothers, teachers, and day care providers. In non-European cultures,
however, the child who maintains eye contact, when spoken to, is showing
resistance and disrespect for the adult; a lowered gaze is considered respect-
ful and compliant. Culturally determined ways to send emotional messages
are gradually built into an internal working model of social relationships.
The way mental health professionals work with parents may be defined
in terms of a cultural view of relationships. Among Latino families, for ex-
ample, there is an expectation ofpersonalismo* This term refers to the value
of warm and individualized attention to others, with some informality and
openness, but without being overly familiar. Personalismo requires helping
professionals to give respect to parents' beliefs and assumptions, rather than
pressing upon the family the messages garnered from child development
research. One such belief among Latino families is that family members
should be interdependent, an idea that contrasts with the mainstream or
standard American goal of making babies independent. 9
Child Care and Relationships. The improvement of relationships for in-
fants and toddlers in day care has been a major thrust of child development
advocates in recent years. A national initiative, the Better Baby Care
Campaign,10 has focused on a number of improvements in the child care
standards as they relate to attachment. These include specialized training for
infant and toddler caregivers, with a stress on child development and family
relationships.
Other groups, such as the National Association for the Education of
Young Children (NAEYC), Zero to Three, and the World Association for
Infant Mental Health, are trying to bring about changes in child care and
produce a more attachment-friendly setting. They are working against the
common assumption that infant day care is the same thing as preschool,
 New Directions 161

which in turn is popularly assumed to be essentially the same as elementary

school. Such assumptions are shown in the tendency to name day care cen-
ters "Kid Academy" or "Learning Center" and to refer to infant caregivers
as teachers.
NAEYC emphasizes the use of developmentally appropriate practice in all
child care settings as well as in the home. This term refers to ways of work-
ing with children that support the normal processes of the child's stage of
development. For infants and toddlers, the development of secure attach-
ment relationships appears to be the most important of all developmental
processes and to be part of the foundation for other developmental events,
such as the use of language. Developmentally appropriate practice for in-
fants and toddlers must emphasize the growth of relationships. For three-,
four-, and five-year-olds, developmentally appropriate child care needs to
take into account the child's growing abilities to negotiate and deal with
conflicts with peers and adults. Developmentally appropriate day care is
based on relationships, not on instructional techniques.11
Attachment-Friendly Day Care Practices. To allow for the development of
good relationships, day care centers need to pay close attention to staff-to-
child ratios, group size, and staff turnover. For maximum encouragement
of attachment, however, they also need to deal with some other specific
concerns.
Feeding practices in day care settings are important ways to foster social
relationships and emotional development. The issue here is not about
breastfeeding versus bottle-feeding, but about social interactions that are
part of feeding by bottle, spoon, or otherwise. A caregiver feeding an infant
needs to be with one child at a time, holding the baby to bottle-feed or
sitting face-to-face to spoon-feed. For children old enough to self-feed, the
caregiver needs to be present and attentive, but several toddlers may be
helped at the same time; she may steady a cup, hold a bowl still, or talk and
smile. The feeding situation is not just for nutrition; it also provides an op-
portunity for intense, pleasurable social interaction, full of meaningful com-
munication. In order to use feeding in this way, the caregiver must be able
to take enough time, she must know the child and be known by him or
her, and she must be able and willing to pick up the child's communica-
tions about hunger and satisfaction. The attachment-friendly caregiver does
not try to make the baby finish a bottle, trick the child into opening the
mouth for a spoonful of food, or react to play with food as if it is an act of
insubordination.
A day care center's policy about attachment objects is also an indication
of attachment-friendliness. For infants and toddlers dealing with separation
and anxiety, the availability of a beloved blanket, pacifier, or soft toy may
be the secret to comfort. (These objects are not a substitute for the mother,
162 Understanding Attachment

for the child wants them even when in mother's lap.) A common day care
practice in the 1970s and 1980s was to prohibit such objects in the day care
center, or to insist that the child share them—an idea about as acceptable to
toddlers as the proposal to adults that they share a spouse with a friend.
Attachment-friendly care encourages the use of attachment objects, recog-
nizing that these provide a way for children to handle separation and also,
the exciting stimulation of the group.
Recognition of secure base behavior in the day care center is a way to
support both emotional and cognitive development. The toddler's explora-
tion of new people and events works much better when a familiar person
is available as a secure base to return to for emotional refueling. The secure-
base person needs not just to be available, but to cooperate in the task. She
needs to be responsive when the exploring toddler comes back, but also
attentive enough to return eye contact from a distance, or to speak to a
child who is looking back anxiously. A good secure-base person does not
regard the child as clingy or whiny, but recognizes the emotional signifi-
cance of checking-in with a familiar adult.
The handling of transitions is another factor relevant to attachment. The
arrival of a child at the day care center, going to sleep and awakening, and
being picked up by a parent at day's end are all transitions which relate to
separation and reunion concerns. These times need to be handled carefully
and with attention to the child's communications. Arrival and departure are
especially problematic times because parent and child affect each other with
their emotional communications about their situation. The parent may fear
a scene, if the child does not readily accept separation, and may be equally
concerned with the child's distress, worries about other parents' disap-
proval, and anxiety about being late to work. Or, paradoxically, the parent
may be concerned that a child who separates quietly is not sufficiently at-
tached. In either case, the child is liable to be affected by the parent's invol-
untary communication of fear and sadness.
In the late afternoon, many children will express anger about separation,
often by ignoring the returning parent and resisting what the parent tries to
do. Parents in turn feel anxiety, weariness, and irritation at the complexity
of what should be a simple process. Day care centers can help by providing
plenty of time for transitions and giving emotional support for parents and
children who have trouble with separation and reunion. At the time of ar-
rival, especially, a caregiver needs to greet the baby and help to make the
entry a gradual one.
Finally, attachment-friendly day care centers need policies that support
families' emotional lives and their connections with their children.
Exhausted and emotionally drained, young working parents often turn to
their child's caregiver for advice and nurturing, needs that are not the
 New Directions 63

caregiver's real job, nor anything she has usually been trained to deal with.
Day care staffs need help recognizing and understanding parents' needs, but
at the same time they need to set limits on the demands made of them. The
combination of responsiveness and limit setting needs to be done at a man-
agerial level and in a way that provides helpful guidelines for parents. The
increasing interest of day care centers in these relationship issues is a real,
but very gradual change.

Changing Psychological Theory

Since Bowlby and Ainsworth's day, attachment has become a focus for
researchers, clinicians, and parents. Attachment could accurately be de-
scribed as an imperialistic schema, taking over and subordinating other cat-
egories. We have certainly seen how the idea of attachment has become the
most important way of thinking about emotional development. However,
as we look at changes in psychological and mental health theories, we can
see that the concept of attachment is now becoming a part of some broader
schemes, hardly thought of in Bowlby's day.
Systems Theory, Transactional Processes, and Attachment. The idea that at-
tachment is a thing inside a person, a state or process that exists indepen-
dently of other factors, is certainly a gross over-simplification of Bowlby's
work. It is not, however, so very far from the ethological view of an im-
printing event that happens quickly and can permanently change the direc-
tion of development. Nor is this simple view much different from some of
the popularized ideas we encountered in Chapter 7.
Modern psychological theory does not deal much with simple ideas. A
fact that becomes ever clearer is that human beings are complex creatures
dwelling in a complex environment. Like other modern disciplines, psy-
chology has come to respect and to use systems theory, and of all aspects of
psychology, the study of development has been most drawn to the dynamic
systems approach.12 Concepts of attachment are now often couched in sys-
tems terms.
Briefly, dynamic systems theory involves the assumption that many fac-
tors work together to create any behavior, emotion, or thought. These
factors are coordinated in their effects and function nonlinearly. That is, it
is possible that some very large change could have only a small influence on
a developmental outcome, but another small change could create a major
effect (or vice versa). Systems theory assumes a natural variation in events
such as emotions and behaviors and posits that not all changes result from
definable causes. The theory also assumes that any group of factors, in the
child or in the environment, could begin to combine in a new way just
because some small change has occurred. Just as one last degree of heat
164 Understanding Attachment

brings a pan of milk to a boil, a brief experience or small maturational event

could seriously alter beliefs, feelings, or behavior.
How do these systems ideas apply to attachment behaviors and emo-
tions? First of all, there are many factors involved in the expression of at-
tachment, and a slight change may have obvious effects. Toddlers may
ignore their parents, wander off, and even deliberately keep their distance
until a stranger appears or the child gets hurt, whereupon the child makes
a beeline for the parent.13 Children may behave quite differently if they
have had similar attachment experiences but different temperaments (see
Chapter 6). A child may show different attachment emotions and behaviors
toward different people, even people who do not seem dramatically differ-
ent from each other.
Attachment also shows normal variations. A given child does not show
exactly the same attachment behavior—like use of a secure base—every
day, even though the situation and the people involved may be very
similar.
Finally, and most importantly, attachment behavior does seem to follow
the systems tenet that rules and functions can reorganize abruptly. Children's
attachment behavior and emotions do not change instantaneously, but
there is often a noticeably rapid transition from the fearless, convivial pe-
riod of infancy to the characteristic behavior of a toddler who screams at
the approach of a well-meaning stranger, or even a familiar family member.
(For example, one child happened to be taken on a two-day trip by his
mother at seven months; he was restless and unfriendly to other people. On
reunion with the father at an unfamiliar campsite, he cried and turned
away, a difficult response for the father.)
A useful concept related to systems theory is that of the transactional
process.14 This idea also provides an excellent framework for attachment.
Transactional processes are repeated interactions, like many that occur be-
tween parents and infants. In each interaction, each person influences and
changes the other a little, and this alteration changes the ways they are able
to affect each other. At the end of many interactions, each person has
changed, and their ways of interacting are different from what they were
originally.
Attachment behaviors and emotions develop as part of a relationship,
and as we saw in earlier chapters, both partners change. To see this as the
result of a transactional process makes a great deal of sense. Take, for ex-
ample, the development of secure base behavior. A parent might initially
assume that a toddler who goes off exploring in a safe playground needs no
special attention or interest; the adult might read the paper or talk to a
friend, occasionally glancing at the child only for safety's sake. The child
might then surprise and distress the parent by returning, climbing into a lap,
 New Directions 165

demanding attention, and so on, before exploring again. After that occa-
sion, the parent would probably become more attentive, watching the
child, being available to return eye contact when the child looks back, and
perhaps speaking to the child. Together, the two would work out how far
away the child might go with comfort to both partners, and how much at-
tention the parent needs to pay in order to support the child's exploration.
After several more park visits, the transactional process would result in ef-
fective secure base behaviors, which involve seeking on the child's part and
cooperating on the parent's. (If, on the other hand, the parent was insensi-
tive to the child's signals, the result might be a degree of risk-taking behav-
ior sufficient to grab the parent's attention.)
Both systems theory and the transactional process are useful ways to ap-
proach the complexity of attachment. Unfortunately, it is quite difficult to
apply these ideas to the design of research on emotional development, or to
interpret research that involves so many factors. In any case, these ways of
thinking have helped us to get past some oversimplified perspectives on
attachment.
Attachment and Theory of Mind. A fascinating and potentially fruitful new
way of looking at attachment links the process with the development of
Theory ofMind.]:> Theory of Mind (TOM) is also referred to as mentaliza-
tion or Reflective Function. The term Theory of Mind describes the ability
of all normal human adults to read minds—not literally, of course, but to
make excellent guesses about what another person knows, wants, be-
lieves, hopes, and intends. We use T O M when we assume that all these
human bodies around us, a lot like ours, have inside them the capacity to
know and experience the world just as we do. In order to use T O M well,
however, we must also understand that another person's knowledge, de-
sires, intentions, and subjective experience are probably not identical to
our own.
Communication with others depends on T O M . If we were to assume
that speech was just mechanical, reflexive noise making, unconnected with
a mind, we could not understand another person's intentions or make our
own clear. When a guest refuses a second helping of mashed potatoes, our
capacity to use T O M enables us to decide with some degree of accuracy
whether to press the offer or let it drop, or even whether to take offense at
the implication that our cooking might not be appetizing.
Of course, we as adults all know that we can do T O M tasks, even though
we do not always do them very well. What does T O M have to do with
attachment or any part of early emotional development? The connection
comes in the course of early development and involves complicated steps.
The description below is a highly simplified summary of a large body of
research and theory, linking parents' abilities to mentalize their emotional
66 Understanding Attachment

involvement with their children and with attachment and T O M develop-

ment in the child.16
Events in development of T O M and attachment begin with inter subjectiv-
ity, the apparently innate tendency for the young baby and parent to respond
to each other in special, human ways. Although the baby may not really
know what a human being is, she acts as if she does, and from birth she looks,
listens, and interacts with a caregiver in quite different ways than she would
with inanimate objects. For example, the baby takes turns in interactions,
vocalizing and looking for a while, then waiting quietly while the caregiver
does the same. The baby also can and does imitate certain adult facial expres-
sions. The caregiver responds by acting as if the baby intends to communi-
cate, even though it seems impossible for us to know whether this is true.
In the course of these interactions with the young baby, the parent be-
gins to mirror, or mimic, the baby's expression, with the imitation follow-
ing a fraction of a second behind the baby. The baby, of course, does not
see herself in a mirror often. The adult's imitation provides a chance for the
baby to learn the connection between the subjective experience of emo-
tion and the external movements that can reveal the internal emotional
state. "I feel like this," the baby may think, "and it looks like that." These
experiences build a foundation of emotion-linked knowledge that will later
help the baby read the mental processes of others.
Can this process go astray? A baby who is weak, sick, or distressed may
not respond well to a caregiver, who may in turn become depressed and
discouraged, and stop engaging with the child. A baby with facial problems
may not present an expression to which an adult can respond. A caregiver
who is depressed to begin with may not look at or listen to the baby, but
may simply go through caregiving routines mechanically. A caregiver who
is overstressed and preoccupied may mirror emotions inaccurately, or her
face may show only her own internal state, which may not be a good match
for the baby's emotions. And a caregiver with poorly developed T O M may
not pay attention to the baby's communications at all, instead behaving as
if the baby has no subjective experience whatever.
As the baby observes the world and begins to see patterns of behavior,
he or she develops a teleological stance—a set of expectations about what a
person will do next. This stance does not initially take into account the
person's intentions, knowledge or other parts of the mental state, but sim-
ply assumes that people do things that make physical sense (like taking the
shortest path to walk across a room). The baby is puzzled and interested
when behavior does not make sense, but has no way to understand the
internal state of mind that is guiding the action. (One 12-month-old,
newly walking, imitated his mother's attempts to swat a fly in every detail,
including swiveling his head as if following the insect's flight. But he paid
 New Directions 167

no attention at all to an actual fly nearby—showing that he did not under-

stand the intention behind the movements.)
At about nine months of age, just as most are beginning to show concern
about separation, infants begin to coordinate new behaviors with adults in
the form of joint attention. The child uses the gaze to get the adult's atten-
tion and to point it to something interesting. For example, she looks at an
adult until the adult looks back. Then, flicking her gaze toward the inter-
esting puppy or new toy, she uses her eyes to point. Next, she looks back
to check the direction of the adult's gaze, and if the adult is not looking in
the right direction, she tries again. The sequence ends with a series of eye
movements: the two look at the interesting sight, then back at each other
with a smile, and then at the object again.
Joint attention may be the beginning of the intentional stance that makes
T O M much more accurate, enabling the child to judge what another per-
son plans and wants to do, not just what might reasonably happen. Paying
attention to the direction someone looks, and to the person's facial expres-
sion, is a great help in understanding what is intended.
The new understanding of intention and desire is normally in place by
about two years of age and alters communication and understanding a good
deal. (When the little boy described above arrived at an intentional stance,
he grabbed and waved the flyswatter, shouting, "Where dat fly!" in imita-
tion of his mother's intention, not just her actions. He omitted the head
swiveling that had been a feature of his earlier performance.) Although
many two-year-olds have this much T O M and know what people want,
they still do not understand that people may fail to do what they intend if
they believe something that is not true. If a three-year-old sees someone
put a cup in the dishwasher before leaving the room and someone else take
the cup out and put it in the cupboard, the child will be mistaken about
what will happen when the first person returns. Asked where the first adult
will look for the cup, the three-year-old replies "in the cupboard," believ-
ing actions are always based on reality.
Not until age four or five do children realize that the adult's false belief
will make her look in the wrong place. At that age, also, children begin to
think about how they or other people know something—whether they
saw it themselves, whether someone told them, whether they just figured
it out. This step enables the child to understand the consequences of false
belief, but also to have some insight into the chances that a belief is false, or
that someone else may think it is.
How are these steps toward mature T O M related to attachment?
First, T O M and attachment seem to develop as a result of similar social
experiences. The sensitive, responsive caregiver, whom we have repeat-
edly mentioned as an important factor in attachment, is also the person
168 Understanding Attachment

who pays attention to the child, quickly and accurately mirroring the
child's expression. The responsive caregiver quickly notices the child's
gaze and joins in joint attention, looking as instructed and exchanging a
smile and glance of pleasure with the child. The caregiver talks to the child
and responds to and joins in pretend play, showing the affectionate, playful
sociability that seems to be a foundation of attachment. At the same time,
the caregiver is providing the language that helps to describe emotional
states ("Is that a scary noise?"), and some practice in understanding inten-
tion and belief. The child may come to think about intention because the
caregiver acts as if it is important. The caregiver's response to the child's
intentions also provides quick comfort and a return to security. A familiar
caregiver will understand the child's signals best and therefore be the best
at comforting the child and providing a secure base.
Development of T O M is essential for negotiation, a significant step in
the attachment process. To negotiate effectively, the child must recognize
that the parent's intentions, desires, and beliefs may be different from the
child's, and this means that to some degree, the child must be capable of
reflecting on his or her own mental state. Negotiation involves attempts to
move toward another's position without abandoning too much of one's
own, and thus requires knowledge of both people's positions, which are
based partly on T O M . Similarly, the development of the goal-corrected
partnership involves progressive change in each partner's assessment of the
other's knowledge and wishes. Parents can and do model this aspect of
T O M by offering compromise, explaining their thinking, and admitting
that they have had a false belief, if this is the case.
Finally, T O M seems to be necessary for development of an effective
internal working model of social relations. Mature T O M involves ways of
thinking both about the self and about other people. These capacities en-
able the individual to be aware of how his own intentions and beliefs are
like those of other people and how they are different. It also enables the
child to understand how others may read him, thus predicting safety, secu-
rity, hostility, nurturing, or rejection by others. Such predictions can in-
clude how others will act and feel about themselves and how one will act
and feel toward someone else. Importantly, mature T O M allows the un-
derstanding of individual differences and needs, a step that creates a differ-
ent position in the internal working model for the vulnerable baby than for
the adult enemy.

CONCLUSION: ATTACHMENT AND POLICY

So far in this chapter, we have seen answers to some important questions
about attachment, but we have also seen that related practice and theory are
 New Directions 169

very much in flux. Is it possible to use what we know about attachment to

establish good policy and legal guidelines for the benefit of children and
their families?
The step from research and theory to policy is one that requires great
care. As Sir Michael Rutter, the eminent researcher who studied the
adopted Romanian orphan group, has warned, it is possible for policy
change to be based far more on evangelism than on systematic evidence.17
Jack Shonkoff, a noted leader in family policy, has given a similar warn-
ing. Shonkoff has pointed out that in establishing policy we have access to
three categories of material: established knowledge based on strict rules of
evidence; reasonable hypotheses that go beyond the limits of what we
clearly know; and unwarranted assertions that can include "blatant distor-
tion or misrepresentation ... of science."18 Policy decisions can be based on
established knowledge and, to some extent, can follow reasonable hypoth-
eses, as long as these are used responsibly and with awareness that they may
not be true. Unwarranted assertions, of course, should never form the basis
of policy decisions.
This book has summarized a great deal of established knowledge and has
presented some reasonable hypotheses about attachment. Some assertions
were also reviewed and identified as unwarranted. Is it possible to use the
information we have discussed to make recommendations about policy?
There are some areas in which we have enough established knowledge
and reasonable hypotheses to support policymaking. These areas include
day care, foster care, some aspects of adoption, and some aspects of divorce
and custody decision-making. We have enough information on these top-
ics to guide judges in certain types of decisions and to establish day care
arrangements that support early emotional development. (We may not
have the money available to do this, of course.)
Here are some important evidence-based conclusions that could be used
to establish policy:

1. Newborn infants are little, if at all, affected by separation from the bio-
logical mother. The mother may be distressed by the loss, but there is
no need to be concerned about a direct effect on the child.
2. Attachment develops as a result of consistent, sensitive, responsive care,
especially in the first two years. For the best developmental outcomes,
foster care and nonparental day care need to be planned with this foun-
dation. Consistency may be the bugaboo of small minds, but is the
essential factor in the social environment of small children.
3. From about six to eight months and on, abrupt long-term separations
are disturbing to the child and interfere with sleeping, eating, playing,
and learning. A child in this age group is old enough to be affected by
170 Understanding Attachment

separation, but can also recover, especially if treated with sensitivity

and responsiveness by new caregivers. Placement in foster care should
be made with as few changes as possible, and foster or adoptive parents
should receive guidance to encourage sensitivity, especially toward
children who may not communicate well. Custody decisions follow-
ing divorce should follow similar rules.
4. Adults' attitudes about relationships strongly influence children's at-
tachment, and caregivers should be selected or guided in ways that
help them show positive attitudes about attachment. The fact that they
have reared their own children successfully is not necessarily evidence
that foster or adoptive parents will be able to do a good job.
5. By the age of five or six years, children are much less influenced by
separation than they were as toddlers, although they will still be sad
about losses. Custody decisions for children of this age or older do not
need to place stress on consistency of care, even though it was neces-
sary earlier. School-age children and adolescents will not be emotion-
ally traumatized by a custody change in and of itself, and other factors
may be more important in such decisions.

Where More Work Is Needed

There are quite a few other areas where we presently do not have
either evidence or sufficiently reasonable hypotheses about attachment
to make policy decisions. Unanswered questions about emotional devel-
opment are still being worked on. 19 For example, we do not know
enough about what parents do in negotiating with preschool and older
children, and how this style of negotiation contributes to the develop-
ment of an internal working model of social relationships. The absence
of this information makes it difficult for us to apply attachment theory to
school policy or to our understanding of foster care and adoption of
older children.
It is also unclear exactly what factors work together with attachment and
loss to influence emotional development. In discussing conclusions drawn
from the Romanian orphan studies, Rutter has made the following state-
ment: "[It] is clear that parental loss or separation carries quite mild risks
unless the loss leads to impaired parenting or other forms of family malad-
apration." Poor family functioning would involve factors like persistent
conflict, especially when it is negative attention aimed at a particular child;
a lack of individualized caregiving; a lack of conversation and play; and
experience with a social group that fosters undesirable behavior.20 How
these factors interact with loss is still so poorly understood, we cannot es-
tablish policy on this basis.
 New Directions 71

Not only are we without answers to important questions, but there is

also little argument that attachment theory and research remain in a state of
flux. Aspects of policy and law fluctuate too, while technology and social
custom follow their own paths of change. A remarkable legal case in 2000
demonstrated clearly how little grasp we still have on the connections be-
tween modern life and family relations.21 In this strange case, two families
were in conflict after a mix-up in a fertility clinic. By error, a white woman
who had come for fertility treatment was implanted not only with her own
embryos but with those of a black couple. The white woman, Donna
Fasano, gave birth to two boys of different skin shades; she named the black
child Joseph and the white one Vincent. The black woman, Deborah
Perry-Rogers, also received implanted embryos, but she did not have a
successful pregnancy.
Mr. and Mrs. Rogers sought legal help and were given permanent cus-
tody of then five-month-old Joseph, renaming him Akiel. During the legal
process, they had agreed to give the Fasanos visiting rights, but when they
requested legal release from the agreement, they obtained it.
Part of the legal decision-making hinged on attachment issues, but not
the sort this book has generally discussed. When the Rogers parents ob-
tained custody of Akiel, "a judge ordered that Akiel and Vincent should
visit each other regularly in the presence of psychologists, who would de-
termine whether they had formed links as brothers during ... months in the
womb [emphasis added] and in the Fasanos' home." However, a later ruling
stated that the Fasanos had known of the mix-up soon after it occurred and
had done nothing, making "any bonds that formed between the boys le-
gally invalid."22
Here we have a startlingly incoherent legal application of the attachment
concept, with nods to concepts like genetic factors in relationships and to
prenatal development of emotional ties. We also seem to have the belief
that psychologists can detect the existence of attachment, not only in chil-
dren of less than six months, but between infants who are far too young to
play the role of attachment figures for each other or give the cues that
might initiate attachment behavior.
A Kabala scholar is supposed to have said, "Nonsense is always nonsense,
but the history of nonsense is scholarship." Although scholarship still has
much to do with the understanding of attachment, we have come far enough
to be able to tell which ideas are nonsense. Application of systematic attach-
ment evidence to law and policy is already possible. It is to be hoped that
the helping and legal professions will join in carrying out this task.
This page intentionally left blank
Notes

CHAPTER 1
1. Jerrold M. Post, Leaders and Their Followers in a Dangerous World (Ithaca,
NY: Cornell University Press, 2004); E. Bumiller, "Was a Tyrant Prefigured by
Baby Saddam?" New York Times, May 15, 2004, B9.
2. Margaret Mahler, "Rapprochement Subphase of the Separation-Individuation
Process," in Rapprochement: Tlie Critical Subphase of Separation-Individuation, ed. R. Lax,
S. Bach, and J. Burland (New York: Aronson, 1980).
3. Marshall H. Klaus, R. Jerauld, N. Kreger, W. McAlpme, M. Steffa, and
J. H. Kennell, "Maternal Attachment: Importance of the First Post-Partum Days,"
New England Journal of Medicine 286 (1972): 460-463.
4. Daniel Yankelovitch Group for Zero to Three and Civitas, "What Grown-
ups Understand about Child Development: A Benchmark Study" (Washington,
DC: Zero to Three, 2000).
5. Jack P. Shonkoff, "Science, Policy, and Practice: Three Cultures in Search
of a Shared Mission," Child Development 7\ (2000): 181-187.

CHAPTER 2
1. Valerie Fildes, Wet Nursing: A History from Antiquity to the Present (New
York: Blackwell, 1988).
2. Ibid., 115, 117.
3. R. de Saussure, "J. B. Felix Descuret," trans. R. de Saussure, Psychoanalytic
Study of the Child 2 (1940): 420.
174 Notes

4. Philip Slater, "Child Rearing During the Early National Period" (PhD
diss., University of California, Berkeley, 1970).
5. Ian Suttie, The Origins of Love and Hate (London: Penguin, 1935).
6. Irenaus Eibl-Eibesfeldt, Ethology: The Biology of Behavior (New York: Holt
Rinehart Winston, 1970).
7. K. M. Wolf, "Evacuation of Children in Wartime," Psychoanalytic Study of
the Child! (1945): 389.
8. Anna Freud and Dorothy T. Burlingham, War and Children (New York:
Medical War Books, 1943).
9. Anna Freud and Dorothy T. Burlingham, Infants Without Families: Reports
on the Hampstead Nurseries, 1935-1945, vol. 3, Writings of Anna Freud (New York:
International Universities Press, 1973).
10. John Bowlby, Attachment (New York: Basic, 1982).
11. Bruno Bettelheim, The Children of the Dream (Cambridge, MA: Harvard
University Press, 1969); S. Diamond, "Kibbutz and Shtetl: The History of an
Idea," Social Problems 2 (1957): 71-99; and M. E. Spiro, Children of the Kibbutz
(Cambridge, MA: Harvard University Press, 1958).
12. Bettelheim, Children of the Dream, 33.
13. Ibid., 108.
14. K.Johnson, H. Banghan, and W. Liyao, "Infant Abandonment and Adop-
tion in China," Population and Development Review 24, no. 3 (1998): 469-510.
15. Michael Rutter, J. Kreppner, J. O'Connor, and English and Romanian
Adoptees (ERA) Study Team, "Risk and Resilience Following Profound Early
Global Deprivation," British Journal of Psychiatry 179 (2001): 97-103.
16. Michael Rutter, "Nature, Nurture, and Development: From Evangelism
through Science Toward Policy and Practice," Child Development 73 (2002): 1-21.

CHAPTER 3
1. Rene Spitz, "Hospitalism: An Inquiry into the Genesis of Psychiatric Con-
ditions in Early Childhood," Psychoanalytic Study of the Child 1 (1945): 53-74.
2. Jeremy Holmes, John Bowlby and Attachment Theory (London: Routledge,
1993); Inge Bretherton, "The Origins of Attachment Theory: John Bowlby and
Mary Ainsworth," Developmental Psychology 28 (1992): 759-775.
3. Melanie Klein, Contributions to Psychoanalysis, 1921-1945 (London: Hogarth,
1948).
4. Bretherton, "Origins of Attachment Theory."
5. John Bowlby, "Forty-Four Juvenile Thieves: Their Characters and Home
Life," International Journal ofPsychoanalysis 25 (1944): 19-52, 107-127.
6. James Robertson and Joyce Robertson, Nine Days in a Residential Nursery
(Van Nuys, CA: Child Development Media, 1953) video.
7. Harry Harlow, "The Nature of Love," American Psychologist 13 (1959):
573-585.
8. John Bowlby, "The Nature of a Child's Tie to his Mother," International
Journal of Psychoanalysis 39 (1958): 350-433; John Bowlby, "Separation Anxiety,"
 Notes 175

International Journal of Psychoanalysis 41 (1960): 89-113; and John Bowlby, "Grief

and Mourning in Infancy and Early Childhood," Psychoanalytic Study of the Child
15 (1960): 9-52.
9. Bretherton, "Origins of Attachment Theory."
10. Ibid.
11. Mary Ainsworth, M. Blehar, E. Waters, and S. Wall, Patterns of Attachment
(Hillsdale, NJ: Erlbaum, 1978).
12. D. Teti and M. Nakagawa, "Assessing Attachment in Infancy: The Strange
Situation and Alternative Systems," in Interdisciplinary Assessment of Infants, ed.
E. Gibbs and D. Teti (Baltimore: P. H. Brookes, 1990), 91-214.
13. Mary Main andj. Solomon, "Discovery of an Insecure-Disorganized/Dis-
oriented Attachment Pattern," in Affective Development in Infancy, ed. T. B. Brazelton
and M. Yogman (Norwood, NJ: Ablex, 1986), 95-124.
14. Everett Waters and K. Deanne, "Defining and Assessing Individual Differ-
ences in Attachment Relations: Methodology and the Organization of Behavior in
Infancy and Early Childhood," Monographs of the Society for Research in Child Devel-
opment 50 (1985): 41-65.
15. C. George, N. Kaplan, and M. Main, "Adult Attachment Interview" (un-
published manuscript, University of California, Berkeley, 1995).
16. M. J. Bakermans-Kranenburg and M. van IJzendoorn, "A Psychometric Study
of the Adult Attachment Interview," Developmental Psychology 29 (1993): 870-879.
17. Daniel Yankelovitch Group for Zero to Three and Civitas, "What Grown-
ups Understand about Child Development: A Benchmark Study" (Washington,
DC: Zero to Three, 2000).
18. Jean Mercer and Clyde McMurdy, "A Stereotyped Following Behavior in
Young Children," Journal of General Psychology 112 (1985): 261-266.
19. J. L. Gewirtz, "Potency of a Social Reinforcer as a Function of Satiation
and Recovery," Developmental Psychology 1 (1969): 2—13.
20. John Bowlby, Darwin: A New Life (New York: W. W. Norton, 1990).

CHAPTER 4
1. Rene Spitz, "Relevancy of Direct Infant Observation," Psychoanalytic Study
of the Child 5 (1950): 66-73.
2. W. Ernest Freud, "Prenatal Attachment and Bonding," in Course of Life,
vol. 1, Infancy, ed. S. I. Greenspan and G. H. Pollock (Madison, CT: International
Universities Press, 1989), 467-484.
3. Donald Winnicott, "Transitional Objects and Transitional Phenomena," in
Collected Papers: Through Pediatrics to Psychoanalysis, ed. D. Winnicott (Middlesex,
England: Penguin, 1953), 1-25.
4. L. B. Adamson and C. L. Russell, "Emotion Regulation in the Emer-
gence of Joint Attention," in Early Social Cognition, ed. P. Rochat (Mahwah, NJ:
Erlbaum, 1999), 281-297.
5. Rene Spitz, "Hospitalism: An Inquiry into the Genesis of Psychiatric Con-
ditions in Early Childhood," Psychoanalytic Study of the Child 1(1945): 53-74.
176 Notes

6. R. Weiss, "Loss and Recovery," in Handbook of Bereavement, ed. M. Stroebe,

W. Stroebe, and R. Hansson (New York: Cambridge University Press, 1994),
271-284.
7. John H. Kennell and Marshal H. Klaus, "Caring for the Parents of a Still-
born or an Infant Who Dies," in Parent-Infant Bonding, ed. M. H. Klaus and J. H.
Kennell (St. Louis: Mosby, 1982), 151-266.
8. Marshal H. Klaus, R. Jerauld, N. Kreger, W. McAlpine, M. Steffa, and
J. H. Kennell, "Maternal Attachment: Importance of the First Post-Partum Days,"
New England Journal of Medicine 286 (1972): 460-463.
9. Referred to by David M. Levy, "The Concept of Maternal Overprotec-
tion," in Parenthood: Its Psychology and Psychopathology, ed. E. T. Anthony and
T. Benedek (Boston: Little, Brown, 1970), 387-409.
10. A. Sharma, M. McGue, and P. Benson, "The Psychological Adjustment
of United States Adoptive Adolescents and Their Non-Adopted Siblings," Child
Development 69 (1998): 791-802.

CHAPTER 5
1. Bowlby, "Forty-Four Juvenile Thieves," 19-52, 107-127.
2. Jerome Kagan, Three Seductive Ideas (Cambridge, MA: Harvard University
Press, 2000).
3. M. A. Easterbrooks and W. A. Goldberg, "Security of Toddler-Parent
Attachment," in Attachment in the Preschool Years, ed. M. Greenberg, D. Cicchetti,
and M. Cummings (Chicago: University of Chicago Press, 1990), 221-244.
4. Mary Main and D. Weston, "The Independence of Infant-Mother and
Infant-Father Relationships: Security of Attachment Characterizes Relationships,
Not Infants," Child Development 52 (1981): 932-940.
5. Vivian B. Shapiro, Janet R. Shapiro, and Isabel Paret, Complex Adoption and
Assisted Reproductive Technology (New York: Guilford, 2001).
6. Nancy Scheper-Hughes, Death Without Weeping (Berkeley, CA: University
of California Press, 1992).
7. Ibid., 471.
8. Marinus van IJzendoorn, "Adult Attachment Representations, Parental
Responsiveness, and Infant Attachment: A Meta-Analysis on the Predictive
Validity of the Adult Attachment Interview," Psychological Bulletin 117 (1995):
387-403.
9. Mary Main and E. Hesse, "Parents' Unresolved Traumatic Experiences
Are Related to Infants' Insecure-Disorganized/Disoriented Attachment Status: Is
Frightened or Frightening Behavior the Linking Mechanism?" in Attachment in the
Preschool Years, ed. M. Greenberg, D. Cicchetti, and M. Cummings (Chicago:
University of Chicago Press, 1990), 161-184.
10. Robertson and Robertson, Nine Days in a Residential Nursery.
11. Jay Belsky and M. J. Rovine, "Nonmaternal Care in the First Year of Life
and the Security of Infant-Parent Attachment," Child Development 59 (1988): 157-167;
Jay Belsky and J. M. Braungart, "Are Insecure-Avoidant Infants with Extensive
 Notes 177

Day-Care Experience More Independent in the Strange Situation?" Child Develop-

ment 62 (1991): 567-571.
12. Inge Bretherton, B. Golby, and E. Cho, "Attachment and the Transmis-
sion of Values," in Parenting and Children's Internalization of Values, ed. T. Grusec
and L. Kuczynski (New York: Wiley, 1997), 103-134.
13. R. Furman, "A Child's Capacity for Mourning," in The Child in his Family,
vol. 2, ed. E. Anthony and C. Koupernik (New York: Wiley, 1973), 225-232.
14. Judith Wallerstein, "Children of Divorce: The Psychological Tasks of the
Child," American Journal of Orthopsychiatry 53 (1983): 230-243.
15. R. Weiss, "Loss and Recovery," in Handbook of Bereavement, ed. M. Stroebe,
W. Stroebe, and R. Hansson (New York: Cambridge University Press, 1994),
27-284.
16. S. Murphy, "Identifying Pattern and Meaning in Sibling-Infant Relation-
ships: Using Multiple Forms of Family Data," in Qualitative Methods in Family Re-
search, ed. J. Gilgun, K. Daly, and G. Handel (Newbury Park, CA: Sage, 1992),
146-171.
17. Murphy, "Identifying Pattern and Meaning."
18. P. Silverman and J. Worden, "Children's Reaction to the Death of a Par-
ent," in Handbook of Bereavement, ed. M. Stroebe, W. Stroebe, and R. Hansson
(New York: Cambridge University Press, 1993), 300-316.
19. E. M. Hetherington, M. Bridges, and G. Insabella, "What Matters? What
Does Not? Five Perspectives on the Association Between Marital Transitions and
Children's Adjustment," American Psychologist 53 (1998): 167-184.
20. Wallerstein, "Children of Divorce."
21. Ibid.; Hetherington, "What Matters? What Does Not?"
22. M. Cunliffe, C. Lee, A. Bashe, and M. Elliot, "'Thanks, Mom!' or 'Mom?
No Thanks.' Emotion Regulation and Attachment in Adolescence," poster pre-
sented at Biennial Meeting of the Society for Research in Child Development
(Albuquerque, N M , April, 1999).
23. David Elkind, The Hurried Child (Reading, MA: Addison-Wesley, 1981).
24. A. C. Huth-Bocks, A. A. Levendosky, C. A. Bogat, and A. von Eye, "The
Impact of Maternal Characteristics and Contextual Variables on Infant-Mother
Attachment," Child Development 75 (2004): 480-496.

CHAPTER 6
1. American Psychiatric Association, Diagnostic and Statistical Manual of Mental
Disorders, 4th ed. (Washington, DC: American Psychiatric Association, 1994).
2. Rutter, "Nature, Nurture, and Development," 1—21.
3. Mary Main and J. Solomon, "Discovery of an Insecure-Disorganized/Dis-
oriented Attachment Pattern," in Affective Development in Infancy, ed. T. B. Brazelton
and M. Yogman (Norwood, NJ: Ablex, 1986), 136.
4. Ibid., 137.
5. Ibid., 138.
6. Ibid.
78 Notes

7. Ibid., 139.
8. Main and Hesse, "Parents' Unresolved Traumatic Experiences," 161-184.
9. Ibid., 175.
10. Ibid.
11. Ibid., 176.
12. Ibid.
13. John Bowlby, M. Ainsworth, M. Boston, and D. Rosenblith, "The Effects
of Mother-Child Separation: A Follow-Up Study," British Journal of Medical Psy-
chology 29 (1956): 211-247.
14. American Psychiatric Association, Diagnostic and Statistical Manual.
15. Rochelle F. Hanson and E. G. Spratt, "Reactive Attachment Disorder in
Children: What We Know About the Disorder and Implications for Treatment,"
Child Maltreatment 5, no. 2 (2000): 137-145.
16. Frederick R. Volkmar, "Reactive Attachment Disorder," in DSM-IVSource-
book, ed. T. A. Widiger, A. J. Frances, H. A. Pincus, R. Ross et al. (Washington,
DC: American Psychiatric Association, 1997).
17. World Health Organization, ICD-10 (Geneva, Switzerland: W H O , 1992).
18. Charles H. Zeanah, O. Mammen, and A. F. Lieberman, "Disorders of At-
tachment," in Handbook of Infant Mental Health, ed. C. H. Zeanah (New York:
Guilford, 1993), 332-349.
19. Ibid., 343.
20. Ibid., 344.
21. CarlJ. Sheperis, R. A. Dogget, N. E. Hoda, T. Blanchard, E. L. Renfro-
Michael, S. H. Holdiness, and R. Schlagheck, "The Development of an Assessment
Protocol for Reactive Attachment Disorder," Journal of Mental Health Counseling
25, no. 4 (2003): 291-310.
22. Constance G. Dalenberg, Countertransference and the Treatment of Trauma
(Washington DC: American Psychological Association, 2000).
23. Mary Dozier, K. C. Stovall, K. Albus, and B. Bates, "Assessment for Infants
in Foster Care: The Role of the Caregiver's State of Mind," Child Development 72
(2001): 1467-1477.
24. Alicia F. Lieberman, "The Treatment of Attachment Disorders in Infancy
and Early Childhood: Reflections from Clinical Intervention in Later Adopted
Foster Care Children," Attachment and Human Development 5, no. 3 (2003): 279-282.
25. Ibid., 281.
26. Ibid., 282.
27. Ibid.
28. Mary Dozier, "Attachment-Based Treatment for Vulnerable Children,"
Attachment and Human Development 5, no. 3 (2003): 253-257.
29. Ibid., 255.
30. Ibid., 256.
31. Stanley I. Greenspan, Infancy and Early Childhood (Madison, CT: Interna-
tional Universities Press, 1992).
32. Jean Mercer, L. Sarner, and L. Rosa, Attachment Therapy On Trial (Westport,
CT: Praeger, 2003).
 Notes 79

33. Ronald S. Federici, Help for the Hopeless Child (Alexandria, VA: Ronald S.
Federici and Associates, 2003).
34. Vivian B. Shapiro, Janet R. Shapiro, and Isabel Paret, Complex Adoption and
Assisted Reproductive Technology (New York: Guilford, 2001).
35. Daniel S. Schechter, "How Post-Traumatic Stress Affects Mothers' Per-
ceptions of their Babies: Brief Video Feedback Intervention Makes a Difference,"
Zero to Three 24, no. 3 (2004): 43-49.
36. Ibid., 43.
37. Ibid.
38. Ibid., 47.
39. Ibid., 48.

CHAPTER 7
1. J. Sears, M. Sears, R. Sears, and W. Sears, The Baby Book: Everything You
Need to Know About Your Baby From Birth to Age Two (New York: Little, Brown,
2003).
2. Gary Ezzo and R. Bucknam, On Becoming Babywise: Parenting Your Pretod-
dler, Five to Fifteen Months (Sisters, O R : Multnomah Books, 1995).
3. Robert W. Zaslow and M. Menta, The Psychology of the Z-Process (San Jose,
CA: San Jose State University Press, 1975).
4. Mercer et al., Attachment Therapy On Trial.
5. Nancy Thomas, "Parenting Children with Attachment Disorders," in
Handbook of Attachment Interventions, ed. T. M. Levy (San Diego, CA: Academic,
2000), 67-111.
6. The mental health practitioner who gave this description has given permis-
sion for its use, but has asked not to be named.
7. Elizabeth Randolph, Manual for the Randolph Attachment Disorder Question-
naire (Evergreen, C O : The Attachment Center Press, 2000).
8. Jirina Prekop, Haettest du mich festgehalten (Muenchen: Koesel, 1991).
9. J. Goldstein, A. Solnit, S. Goldstein, and A. Freud, The Best Interest of the
Child: The Least Detrimental Alternative (New York: Free Press, 1996).
10. American Psychiatric Association, Diagnostic and Statistical Manual.
11. D. E. Arredondo and L. P. Edwards, "Attachment, Bonding, and Recipro-
cal Connectedness," Journal of the Center for Families, Children, and the Courts 146
(2000): 109-127.

CHAPTER 8
1. Everett Waters, K. Kondo-Ikemura, G. Posada, andj. Richters, "Learning
to Love: Mechanisms and Milestones" in Minnesota Symposia on Child Psychology,
vol. 23, Self-Processes and Development, ed. M. Gunnar and L. Sroufe (Hillsdale, NJ:
Erlbaum, 1991), 217-255.
2. C. L. Bokhorst, M. J. Bakermans-Kranenburg, R. M. Pasco Fearon,
M. van IJzendoorn, P. Fonagy, and C. Schuengel, "The Importance of Shared
180 Notes

Environment in Mother-Infant Attachment Security: A Behavioral-Genetic

Study," Child Development 74 (2003): 1769-1782.
3. Daniel Bar-On, Fear and Hope: Life-Stories of Five Israeli Families of Holocaust
Survivors, Three Generations in a Family (Cambridge, MA: Harvard University Press,
1995).
4. Glen Elder, "The Life-Course as Developmental Theory," Child Develop-
ment 69 (1998): 1-2.
5. Vivian Shapiro, J. Shapiro, and I. Paret, Complex Adoption and Assisted Re-
productive Technology (New York: Guilford, 2001).
6. Jay Belsky and M. J. Rovine, "Nonmaternal Care in the First Year of Life
and the Security of Infant-Parent Attachment," Child Development 59 (1988): 157—
167; Jay Belsky and K. M. Braungart, "Are Insecure-Avoidant Infants with Exten-
sive Day-Care Experience More Independent in the Strange Situation?" Child
Development 62 (1991): 567-571.
7. K. A. Clarke, "Infant Day Care: Maligned or Malignant," American Psy-
chologist 44 (1989): 266-273.
8. G. R. Munoz, "A Relational and Sociocultural Approach in Services to a
Mexican Family Across the Generations," Zero to Three 23, no. 3: 26-32.
9. F. Stott and R. Halpern, "Listening to the Voices of Families: Thoughts,
Hopes, and Fears in a Latino Community," Zero to Three 23, no.3 (2003): 16—21.
10. R. Lurie-Hurvitz, "A Champion for Babies: An Interview with Joan
Lombardi," Zero to Three 21, no. 6 (2001): 40-44.
11. Jeree Pawl, "Infants in Day Care: Reflection on Experiences, Expectations,
and Relationships," Zero to Three 10, no. 3 (1990): 1-6.
12. R. Aslin, "Commentary: The Strange Attractiveness of Dynamic Systems
to Development," in A Dynamic Systems Approach: Applications, ed. L. Smith and
E. Thelen (Cambridge, MA: MIT Press, 1993), 385-399.
13. Jean Mercer and Clyde McMurdy, "A Stereotyped Following Behavior in
Young Children," Journal of General Psychology 112 (1985): 261—266.
14. Arnold Sameroff, "Factors in Predicting Successful Parenting," in Minimiz-
ing High Risk Parenting, ed. V. Sasserath (Skillman, NJ: Johnson and Johnson),
16-24.
15. P. Fonagy, G. Gergely, E. L. Jurist, and M. Target, Affect Regulation, Men-
talization, and the Development of the Self (New York: Other Press, 2002).
16. Ibid.
17. Rutter, "Nature, Nurture, and Development," 1-21.
18. Shonkoff, "Science, Policy, and Practice," 183.
19. Ross Thompson, "The Legacy of Early Attachments," Child Development
71 (2000): 145-152.
20. Rutter, "Nature, Nurture, and Development," 8.
21. A. Newman, "Visiting Rights Denied in Embryo Mix-Up Case," New
York Times, Metro section, October 27, 2000.
22. Ibid.,B3.
Bibliography

Adamson, L. B., and C. L. Russell. "Emotion Regulation and the Emergence of

Joint Attention." In Early Social Cognition, edited by P. Rochat. Mahwah,
NJ: Erlbaum, 1999.
Ainsworth, Mary, M. Blehar, E. Waters, and S. Wall. Patterns of Attachment.
Hillsdale, NJ: Erlbaum, 1978.
American Psychiatric Association. Diagnostic and Statistical Manual of Mental
Disorders. 4th ed. Washington, DC: American Psychiatric Association,
1994.
Arredondo, D. E., and L. P. Edwards. "Attachment, Bonding, and Reciprocal
Connectedness." Journal of the Center for Families, Children, and the Courts
146 (2002): 109-127.
Aslin, R. "Commentary: The Strange Attractiveness of Dynamic Systems to
Development." In A Dynamic Systems Approach to Development: Applications,
edited by L. Smith and E. Thelen. Cambridge, MA: MIT Press, 1993.
Bakermans-Kranenburg, M. J., and M. van IJzendoorn. "A Psychometric Study of
the Adult Attachment Interview: Reliability and Discriminant Validity."
Developmental Psychology 29 (1993): 870-879.
Bar-On, Daniel. Fear and Hope: Life-Stories of Five Israeli Families of Holocaust
Survivors, Three Generations in a Family. Cambridge, MA: Harvard University
Press, 1995.
Barr, R. "Reduction of Infant Crying by Parent Carrying." In Advances in Touch,
edited by N. Gunzenhauser. Skillman, NJ: Johnson and Johnson, 1996.
Beck, M. "Birthfather Registries." http://www.adoptivefamilies.com/article/
PrinterFriendly.php?aid=233 (accessed May 4, 2004).
82 Bibliography

Belsky, Jay, a n d j . M. Braungart. "Are Insecure-Avoidant Infants with Extensive

Day-Care Experience More Independent in the Strange Situation?" Child
Development 62 (1991): 567-571.
Belsky, Jay, and M. J. Rovine. "Nonmaternal Care in the First Year of Life and the
Security of Infant-Parent Attachment." Child Development 59 (1988): 157-167.
Bettelheim, Bruno. The Children of the Dream. New York: Macmillan, 1969.
Bokhorst, C. L., M. J. Bakermans-Kranenburg, R. M. Pasco Fearon, M. van
IJzendoorn, P. Fonagy, and C. Schuengel. "The Importance of Shared
Environment in Mother-Infant Attachment Security: A Behavioral-
Genetic Study." Child Development 74 (2003): 1769-1782.
Bowlby, John. Attachment. New York: Basic, 1982.
Bowlby, John. Darwin: A New Life. New York: W.W. Norton, 1990.
Bowlby, John. "Forty-Four Juvenile Thieves: Their Characters and Home Life."
International Journal of Psychoanalysis 25 (1944): 19-52, 107-127.
Bowlby, John. "Grief and Mourning in Infancy and Early Childhood." Psycho-
analytic Study of the Child 15 (1960): 9-52.
Bowlby, John. "Separation Anxiety." International Journal of Psychoanalysis 41
(1960): 89-113.
Bowlby, John. "The Nature of a Child's Tie to his Mother." International Journal
of Psychoanalysis 39 (1958): 350-433.
Bowlby, John, M. Ainsworth, M. Boston, and D. Rosenblith. "The Effects of
Mother-Child Separation: A Follow-Up Study." British Journal of Medical
Psychology 29 (1956): 211-247.
Bretherton, Inge. "The Origins of Attachment Theory: John Bowlby and Mary
Ainsworth." Developmental Psychology 28 (1992): 759-775.
Bretherton, Inge, B. Golby, and E. Cho. "Attachment and the Transmission of
Values." In Parenting and Children's Internalization of Values, edited by
T. Grusec and L. Kuczinski. New York: Wiley, 1997.
Bumiller, E. "Was a Tyrant Prefigured by Baby Saddam?" New York Times, May
15, 2004, Metro section.
Clarke-Stewart, K. A. "Infant Day Care: Maligned or Malignant?" American Psy-
chologist 44 (1989): 266-273.
Cunliffe, M., C. Lee, A. Bashe, and M. Elliot. April 1999. 'Thanks, Mom!' or
'Mom? No Thanks.' Emotion Regulation and Attachment in Adolescence.
Poster presented at Biennial Meeting of the Society for Research in Child
Development, Albuquerque, NM.
Dalenberg, Constance. Countertransference and the Treatment of Trauma. Washington,
DC: American Psychological Association, 2000.
Daniel Yankelovitch Group for Zero to Three and Civitas. "What Grownups
Understand About Child Development: A Benchmark Study." Washington,
DC: Zero to Three, 2000.
De Saussure, R. "J. B. Felix Descuret." Psychoanalytic Study of the Child 2 (1946):
17-424.
Diamond, S. "Kibbutz and Shtetl: The History of an Idea." Social Problems 2 (1957):
71-99.
 Bibliography 83

Dozier, Mary. "Attachment-Based Treatment for Vulnerable Children." Attach-

ment and Human Development 5, no. 3 (2003): 253-257.
Dozier, Mary, K. C. Stovall, K. Albus, and B. Bates. "Attachment for Infants in
Foster Care: The Role of Caregiver State of Mind." Child Development 72
(2001): 1467-1477.
Easterbrooks, M. A., and W. A. Goldberg. "Security of Toddler-Parent Attach-
ment." In Attachment in the Preschool Years, edited by M. Greenberg,
D. Cicchetti, and E. M. Cummings. Chicago: University of Chicago
Press, 1990.
Eibl-Eibesfeldt, Irenaus. Ethology: The Biology of Behavior. New York: Holt,
Rinehart, Winston, 1970.
Elder, Glen. "The Life-Course as Developmental Theory." Child Development 69
(1998): 1-2.
Elkind, David. The Hurried Child. Reading, MA: Addison-Wesley, 1981.
Emde, Robert N. "Social Referencing Research." In Social Referencing and Social
Construction of Reality in Infancy, edited by S. Feinman. New York: Plenum,
1992.
Ezzo, Gary, and R. Bucknam. On Becoming Babywise. Book Two: Parenting Your
Pretoddler Five to Fifteen Months. Sisters, O R : Multnomah Books, 1995.
Federici, Ronald S. Help for the Hopeless Child. Alexandria, VA: Dr. Ronald S.
Federici and Associates, 2003.
Fildes, Valerie. Wet Nursing. New York: Blackwell, 1988.
Fonagy, P., G., Gergely, E. L. Jurist, andM. Target. Affect Regulation, Mentalization,
and the Development of the Self. New York: Other Press, 2002.
Freud, Anna, and D. T. Burlingham. Infants Without Families and Reports on the
Hampstead Nurseries, 1935-1945. New York: International Universities
Press, 1973.
Freud, Anna, and D. T. Burlingham. War and Children. New York: Medical War
Books, 1943.
Freud, W. Ernest. "Prenatal Attachment and Bonding." In The Course of Life, vol.
1, Infancy, edited by S. I. Greenspan and G. H. Pollock. Madison, CT:
International Universities Press, 1973.
Furman, R. "A Child's Capacity for Mourning." In The Child in his Family, vol. 2,
edited by E. Anthony and C. Koupernik. New York: Wiley, 1973.
Gardner, R. "Guidelines for Assessing Parental Preference in Child-Custody
Disputes." Journal of Divorce and Remarriage 30, nos. 1-2 (1999): 1-9.
George, C , N. Kaplan, and M. Main. "Adult Attachment Interview." Unpub-
lished manuscript, University of California, Berkeley, 1995.
George, J. "Long Thought Dead, A Child is Found Alive and Healthy." New York
Times, March 3, 2004, sec. A.
Gewirtz, J. L. "Potency of a Social Reinforcer as a Function of Satiation and
Recovery." Developmental Psychology 1 (1969): 2-13.
Gilbert, M., and R. Schulzinger. Relinquishing Custody: The Tragic Result of Failing
to Meet Children's Mental Health Needs. Washington, DC: Bazelon Center
for Mental Health Law, 2000.
184 Bibliography

Goldstein, J., A. Solnit, S. Goldstein, and A. Freud. The Best Interest of the Child:
The Least Detrimental Alternative. New York: Free Press, 1996.
Greenspan, S. I. Infancy and Early Childhood. Madison, CT: International Universi-
ties Press, 1993.
Hanson, R. F., and E. G. Spratt. "Reactive Attachment Disorder in Children:
What We Know About the Disorder and Implications for Treatment."
Child Maltreatment 5 (2000), no. 2: 137-145.
Harlow, Harry. "The Nature of Love." American Psychologist 13 (1959): 5 7 3 -
685.
Hart, A. "Chinese Parents Not Tricked, Judge Says in Custody Case." New York
Times, May 13, 2004, sec. A.
Hetherington, E. M., M. Bridges, and G. Insabella. "What Matters? What Does
Not? Five Perspectives on the Association Between Marital Transitions and
Children's Adjustment." American Psychologist 53 (1998): 167-184.
Holmes, Jeremy. John Bowlby and Attachment Theory. London: Routledge, 1993.
Huth-Bocks, A. C , A. A. Levendosky, G. A. Bogat, and A. von Eye. "The Impact
of Maternal Characteristics and Contextual Variables on Infant-Mother
Attachment." Child Development 75 (2004): 480-496.
Johnson, K., H. Banghan, and W. Liyao. "Infant Abandonment and Adoption in
China." Population and Development Review, no. 3 (1998): 469—510.
Kagan, Jerome. Three Seductive Ideas. Cambridge, MA: Harvard University Press,
2000.
Kennell, J. H., and M. H. Klaus. "Caring for the Parents of a Stillborn or an Infant
Who Dies." In Parent-Infant Bonding, edited by M. H. Klaus and J. H.
Kennell. St. Louis: Mosby, 1982.
Klaus, M. H., R. Jerauld, N. Kreger, W. McAlpme, M. Steffa, andj. H. Kennell.
"Maternal Attachment: Importance of the First Post-Partum Days." New
England Journal of Medicine 286 (1972): 460-463.
Klein, Melanie. Contributions to Psychoanalysis, 1921—1945. London: Hogarth,
1948.
Leon, I. "Adoption Losses: Naturally Occurring or Socially Constructed?" Child
Development 73 (2002): 652-663.
Levy, David M. "The Concept of Maternal Overprotection." In Parenthood: Its
Psychology and Psychopathology, edited by E. J. Anthony and T. Benedek.
Boston: Little, Brown, 1970.
Lieberman, Alicia F. "The Treatment of Attachment Disorders in Infancy and
Early Childhood: Reflections from Clinical Intervention in Later Adopted
Foster Care Children." Attachment and Human Development 5, no. 3 (2003):
279-282.
Lurie-Hurvitz, E. "A Champion for Babies: An Interview with Joan Lombardi."
Zero to Three 21, no. 6 (2001): 40-44.
Mahler, Margaret. "Rapprochement Subphase of the Separation-Individuation
Process." In Rapprochement: The Critical Subphase of Separation-Individuation,
edited by R. Lax, S. Bach, and J. Burland. New York: Aronson, 1980.
 Bibliography 85

Main, Mary, and D. Weston. "The Independence of Infant-Mother and Infant-

Father Attachment Relationships: Security of Attachment Characterizes
Relationships, Not Infants." Child Development 52 (1981): 932-940.
Main, Mary, and E. Hesse. "Parents' Unresolved Traumatic Experiences Are
Related to Infants' Insecure-Disorganized/Disoriented Attachment Status:
Is Frightened or Frightening Behavior the Linking Mechanism?" In
Attachment in the Preschool Years, edited by M. Greenberg, D. Cicchetti, and
M. Cummings. Chicago: University of Chicago Press, 1990.
Main, Mary, andj. Solomon. "Discovery of an Insecure-Disorganized/Disoriented
Attachment Pattern." In Affective Development in Infancy, edited by T. B.
Brazelton and M. Yogman. Norwood, NJ: Ablex, 1986.
Mercer, Jean, and C. McMurdy. "A Stereotyped Following Behavior in Young
Children." Journal of General Psychology 112 (1985): 261-266.
Mercer, Jean, L. Sarner, and L. Rosa. Attachment Therapy On Trial. Westport, CT:
Praeger, 2003.
Munoz, G. R. "A Relational and Sociocultural Approach in Services to a Mexican
Family across Three Generations." Zero to Three 23, no. 3 (2003): 26-32.
Murphy, S. "Identifying Pattern and Meaning in Sibling-Infant Relationships:
Using Multiple Forms of Family Data." In Qualitative Methods in Family
Research, edited by J. Gilgun, K. Daly, and G. Handel. Newbury Park, CA:
Sage, 1992.
Newman, A. "Visiting Rights Denied in Embryo Mix-Up Case." New York Times,
October 27, 2000, Metro section.
Pawl, Jeree. "Infants in Day Care: Reflections on Experiences, Expectations, and
Relationships." Zero to Three 10, no. 3 (1990): 1-6.
Post, J. M. Leaders and their Followers in a Dangerous World. Ithaca, NY: Cornell
University Press, 2004.
Prekop, Jirina. Haettest du mich festgehalten. Muenchen: Koesel, 1991.
Randolph, Elizabeth. Manual for the Randolph Attachment Disorder Questionnaire. 3rd
ed. Evergreen, C O : The Attachment Center Press, 2000.
Robertson, James, andj. Robertson. Nine Days in a Residential Nursery. VHS. Van
Nuys, CA: Child Development Media, 1953.
Rutter, Michael. "Nature, Nurture, and Development: From Evangelism Through
Science Toward Policy and Practice." Child Development 73 (2002): 1—21.
Rutter, Michael, J. Kreppner, T. O'Connor, and the English and Romanian
Adoptees (ERA) Study Team of 2001. "Risk and Resilience Following
Profound Early Global Deprivation." British Journal of Psychiatry 179 (2001):
97-103.
Sameroff, Arnold. "Factors in Predicting Successful Parenting." In Minimizing
High-Risk Parenting, edited by V. Sasserath. Skillman, NJ: Johnson &
Johnson, 1983.
Schechter, Daniel S. "How Post-Traumatic Stress Affects Mothers' Perceptions of
their Babies: Brief Video Feedback Intervention Makes a Difference." Zero
to Three 24, no. 3 (2004): 43-49.
186 Bibliography

Scheper-Hughes, Nancy. Death Without Weeping. Berkeley, CA: University of

California Press, 1992.
Sears, J., M. Sears, R. Sears, and W. Sears. The Baby Book: Everything You Need to
Know About Your Baby from Birth to Age Two. New York: Little, Brown,
2003.
Shapiro, Vivian B., J. R. Shapiro, and I. Paret. Complex Adoption and Assisted
Reproductive Technology. New York: Guilford, 2001.
Sharma, A., M. McGue, and P. Benson. "The Psychological Adjustment of
United States Adoptive Adolescents and their Non-Adopted Siblings."
Child Development 69 (1998): 791-802.
Sheperis, CarlJ., R. A. Dogget, N . E. Hoda, T. Blanchard, E. L. Renfro-Michael,
S. H. Holdiness, and R. Schlagheck. "The Development of an Assessment
Protocol for Reactive Attachment Disorder." Journal of Mental Health
Counseling 25, no. 4 (2003): 291-310.
Shonkoff, Jack P. "Science, Policy, and Practice: Three Cultures in Search of a
Shared Mission." Child Development 7\ (2000): 181-187.
Silverman, P., a n d j . Worden. "Children's Reactions to the Death of a Parent."
In Handbook of Bereavement, edited by M. Stroebe, W. Stroebe, and
R. Hansson. New York: Cambridge University Press, 1993.
Slater, Philip. "Child Rearing During the Early National Period." Doctoral
dissertation, University of California, Berkeley, CA, 1970.
Spiro, M. E. Children of the Kibbutz. Cambridge, MA: Harvard University Press,
1958.
Spitz, Rene. "Hospitalism: An Inquiry into the Genesis of Psychiatric Conditions
in Early Childhood." Psychoanalytic Study of the Child 1 (1945): 53-74.
Spitz, Rene. "Relevancy of Direct Infant Observation." Psychoanalytic Study of the
Child 5 (1950): 66-73.
Stott, F., and R. Halpern. "Listening to the Voices of Families: Thoughts, Hopes,
and Fears in a Latino Community." Zero to Three 23, no. 3 (2003): 16—21.
Suttie, Ian. The Origins of Love and Hate. London: Penguin, 1935.
Teti, D., and M. Nakagawa. "Assessing Attachment in Infancy: The Strange
Situation and Alternative Systems." In Interdisciplinary Assessment of Infants,
edited by E. Gibbs and D. Teti. Baltimore: P. H. Brookes, 1990.
Thomas, Nancy. "Parenting children with Attachment Disorders." In Handbook of
Attachment Interventions, edited by T. M. Levy. San Diego, CA: Academic
Press, 2000.
Thompson, Ross A. "The Legacy of Early Attachments." Child Development 71
(2000): 145-152.
Tomasello, M. "Social Cognition Before the Revolution." In Early Social Cognition,
edited by P. Rochat. Mahwah, NJ: Erlbaum, 1999.
Van IJzendoorn, Marinus. "Adult Attachment Representations, Parental R e -
sponsiveness, and Infant Attachment: A Meta-Analysis on the Predictive
Validity of the Adult Attachment Interview." Psychological Bulletin 117
(1995): 387-403.
 Bibliography 187

Van IJzendoorn, Marinus, and M. Bakermans-Kranenburg. "Attachment Disor-

ders and Disorganized Attachment: Similar and Different." Attachment and
Human Development 5, no. 3 (2003): 313-320.
Volkmar, Frederick R. "Reactive Attachment Disorder." In DSM-IV Sourcebook,
vol. 3, edited by T. A. Widiger, A. J. Frances, H. A. Pincus, R. Ross,
et al., Washington DC: American Psychiatric Association, 1997.
Wallerstein, Judith. "Children of Divorce: The Psychological Tasks of the Child."
American Journal of Orthopsychiatry 53 (1983): 230-243.
Waters, Everett, and K. Deanne. "Defining and Assessing Individual Differences
in Attachment Relations: Methodology and the Organization of Behavior
in Infancy and Early Childhood." Monographs of the Society for Research in
Child Development 50 (1985): 41-65.
Waters, Everett, K. Kondo-Ikemura, G. Posada, and J. Richters. "Learning to
Love: Mechanisms and Milestones." In Minnesota Symposia on Child Psy-
chology, vol. 23, Self-Processes and Development, edited by M. Gunnar and
L. Sroufe. Hillsdale, NJ: Erlbaum, 1991.
Weiss, R. "Loss and Recovery." In Handbook of Bereavement, edited by M. Stroebe,
W. Stroebe, and R, Hansson. New York: Cambridge University Press,
1994.
Whitehead, Mary Beth. A Mother's Story. New York: St. Martin's, 1980.
Winnicott, Donald. "Transitional Objects and Transitional Phenomena." In Col-
lected Papers: Through Pediatrics to Psychoanalysis, edited by D. Winnicott.
Middlesex, England: Penguin, 1953.
Wolf, K. M. "Evacuation of Children in Wartime." Psychoanalytic Study of the
Child 1 (1945): 389-404.
World Health Organization. ICD-10. Geneva, Switzerland: W H O , 1992.
Zaslow, Robert W., and M. Menta. The Psychology of the Z-Process. San Jose, CA:
San Jose State University Press, 1975.
Zeanah, Charles H., O. Mammen, and A. F. Lieberman. "Disorders of Attach-
ment." In Handbook of Infant Mental Health, edited by C. H. Zeanah.
New York: Guilford, 1993.
This page intentionally left blank
Index

adolescent parents, 8, 9, 100, 125, 156 125, 138; parent's response to

adolescents: romantic attachments, child, 71, 119
150; separation concerns, animal studies, 4, 13, 17-20, 36
4, 5, 68-70, 77, 98-101 Assisted Reproductive Technology,
adoption: attitudes, 140, 156, 157; 84, 124, 155
early vs. late, 64, 87, 88, ATTACh, 136
144-148; foreign, 8, attachment (definition), 2, 3, 47
26—31; parent training attachment behavior, 3, 4, 5, 62, 66
and personality, 8, 26, attachment emotions, 3, 59, 62
74-75, 119-121, 170 attachment objects, 162
Adult Attachment Interview, 46—47, attachment parenting, 127-134
85, 118 attachment theories, 15-20, 23, 31,
age periods, 38-40, 55-77, 169-170 33-40,47, 52, 163-168
aggression, 63, 71, 76, 106, 113-116, attachment therapy, 115-117, 124,
126, 139 134-142
Ainsworth, Mary Salter, 35, 40-45, attachment thoughts, 3
52, 163 autism, 80, 104-109, 112, 115, 120
"Ali Bowlby and the Forty Thieves," autonomous-secure attachment
36 pattern, 47, 85, 118
Alternative Attachment Disorder autonomy, 61, 67, 69, 95, 99, 133
Diagnoses, 111-114
ambivalence, 4, 41, 68, 77, 84, 110 Baby Jessica, 143, 144
anger: child's response to separation, BabyM, 143
63, 67,98, 107, 114, 116, babywearing, 131
190 Index

Babywise parenting, 134 cultural factors, 111, 131, 152

Baltimore study, 41, 42 cultural issues, 160
behaviors, 80, 85-86, 95, 96, 113, 164 custody decisions, 140, 141, 144,
beltloop technique, 124 146-148, 170
best interest of the child, 141 custody recommendations, 140-148
bipolar disorder, 106
boarder babies, 87, 112 D category, 86
bonding, 5, 6, 70-76, 84, 85, 129, 130 Darwin, Charles, 52
Bowlby, John: practical experience, day care, 88, 157-161
35, 86, 109; psychoanalytic death: of Candace Newmaker, 135; of
background, 35, 36; research child or significant other, 74,
orientation, 36, 40-42; theory 84, 100; of parent, 93-94,
of attachment, 23, 37-40, 97, 121
47-49, 80, 134 depression, 90, 92, 93, 106, 125, 145
brain development, 11-12 diagnosis, 104, 109, 110
breastfeeding, 14-16, 25, 63-65, dismissing attachment pattern, 9, 47
74-75, 129-134, 157 disobedience, 83, 92, 116, 124, 138
Burlingham, Dorothy, 22, 23 disorganized/disoriented attachment,
45, 107-109
caregivers: animals, 18; child as divorce: changing patterns, 153—154;
parent's, 98, 114; in day care child's adjustment, 5, 9-10,
center, 158; in kibbutz 91-93, 110; child's age,
culture, 24-26; relationships 91-93, 96, 100
with, 7, 49, 59, 62-66, domestic violence: effect on
85-86, 89, 140; in traditional attachment status, 106; effect
societies, 86 on emotional life, 9, 114,
Ceausescu, Nicolae, 27 115, 121, 125, 133-134;
child care, 7, 10, 62, 103, 106, 111, foster care, 121
132, 160 drugs and alcohol, 84, 86, 87, 106,
Chinese adoptees, 28, 29 132, 133
Christian thought, 133-134 DSM (Diagnostic and Statistical
clinginess, 49, 63, 90, 103, 113, 162 Manual), 104, 109-112,
comforting, 25, 39, 58-59, 60, 76 115, 116
communication: adoptive parents, 118, dyadic self-regulation, 58
121-123; in day care, 158, dynamic action system, 73
160—162; emotional disorders,
104-105; feeding, 64, 96; early contact, 5-6, 71-73
relationship to attachment elementary school, 66, 67
and bonding, 67, 73, 76, 96 emotional refueling, 61
confounded variables, 82 engagement behaviors, 38, 57
co-sleeping, 131-133 English and Romanian Adoptees
countertransference, 118 (ERA) Study Team, 30, 31
criminal behavior, 80, 109, 115, 116 ethology, 17-20, 163
critical period, 20 evacuation, 21-23
cross-cultural work, 41, 152 evolution, 150
 Index 191

evolutionary psychology, 52, 53 attachment and, 65, 68, 77,

exploration, 33, 61, 62, 165 95, 99-100
eye contact, 160 individual differences, 83
Ezzo parenting. See Babywise infant day care, 7, 88
parenting infant determinism, 11, 81
inherited behaviors, 17—20, 38,
false belief, 167 52, 150
falsifiability, 34 insecure-ambivalent attachment, 45
family patterns, 153 insecure-avoidant attachment, 82
family system, 118-119, 163-165 insecure-disorganized/disoriented
fear, 74, 75, 91 attachment, 45, 107-109
fearlessness, 57 instinct, 17-19, 38, 70
fertility issues, 26 intensity of reaction, 82
floor time (DIR), 122-123 intentional stance, 167
foreign adoptions, 26 internal working model: Bowlby's
foster care: facilitation of attachment, attachment theory, 39-40;
83, 92, 134; parent training changes with age, 64—70;
and personality, 119, changes with experience, 74,
121—122; programs, 7—8, 80, 9 1 - 9 3 , 9 7 - 1 0 1 , 122, 126,
106, 143, 152 151, 155; definition of, 3, 18,
foster parents, 121, 122, 152, 170 48-49, 153
Freud, Anna, 22, 23 intersubjectivity, 166
Freud, Sigmund, 15-17, 75, 76
Freudian theory, 20, 81 joint attention, 62
joint planning, 62, 66
genetic factors, 104, 105
goal-corrected partnership, 65, 67, kibbutz movement, 23—25
69, 132 Kindertransport, 22, 23
grandparents, 64, 91, 97 kinship care, 8, 146
Greenspan, Stanley, 123 Klein, Melanie, 35, 40
grief, 3, 4, 63, 66, 74, 84, 85, 93, 97
language, 67, 93, 105, 141
Hague Convention, 29 legal decisions, 140-148
handicapping conditions, 9 Lieberman, Alicia, 112, 114, 119,
Harlow, Harry, 36 120-121, 124
historical material, 13-26 London Child Guidance
holding, 135, 139 Clinic, 35, 36
hormones, 74 Lorenz, Konrad, 17, 33
hospitalism, 33, 63
hurried child, 99 malnutrition, 85
maternal depression, 74, 84, 105
ICD-10, 111 maternal instinct, 70
imprinting, 1 9 , 2 0 , 7 1 , 7 3 , 163 Matthey, Viktor, 142
independence: parental values and, 49, measurement, 40-48, 81
85, 133-134; secure mental illness, 10, 103-106, 109
192 Index

mentalization, 165—168 Reflective Function, 125

mourning. See Grief relationship-based practice, 160
religious beliefs, 66, 133, 134, 139, 140
National Association for the Education reorganization, 59-60
of Young Children restraint, 135
(NAEYC), 89, 160, 161 reunion behavior, 4, 107
negotiation, 64-66, 68, 77, 92-93, Romanian adoption, 27
99, 168 Romanian orphans, 27, 105, 170
Russian adoptees, 28
obedience, 136, 137 Rutter, Sir Michael, 105, 169, 170
object-relations theory, 15-16, 35
operationism, 34 same-sex couples, 124, 147, 154
Oppositional and Defiant Disorder schizophrenia, 104-106, 115
(ODD), 106, 115 second-parent adoption, 146, 155
orphanages, 120, 169 secure attachment: causes of, 85,
88-89,92, 101, 118-119;
parent education, 120 measurement of, 45-47;
parenthood, 70, 74, 80, 101, 124 results of, 49
peers, 65, 66, 67, 68, 81,95, 109 secure base behavior, 4, 61, 62, 69,
personality, 80-83, 87, 91, 101 106, 112, 162, 164
pervasive developmental disorder, sensitivity, 83, 86, 101, 128, 131, 132
105, 106, 110 separation anxiety, 4, 61, 62, 132
policy, 168-170 sexuality, 69, 99
population differences, 151, 152 Shonkoff, Jack, 11-12, 169
poverty, 24, 26, 85 sibling relationships, 68, 95, 96
power issues, 68, 69, 98 situational factors, 50, 74, 84
pregnancy, 74, 101 sleep problems, 90
prenatal emotion, 56, 101, 138 social referencing, 60
preoccupied attachment pattern, 47 Spitz, Rene, 33, 40, 55, 63
preschool period: attachment Stockholm Syndrome, 75—76
measurement, 46; stranger anxiety, 4, 62, 103, 105
negotiation, 64-66, 76, 81; Strange Situation, 42-46, 81, 82,
separation, 92-93 107, 108, 142
psychoanalysis, 15-16, 33, 35 strangers, 57, 60, 61, 106, 108, 113, 133
psychological parent, 142, 148 strong sitting, 136
psychotherapy, 103, 116-127, Suttie, Ian, 16-17
135, 136 systems theory, 163-165

Q-sort, 46 tantrums, 4, 5, 7, 61, 67, 77, 113,

115, 119
Randolph Attachment Disorder Tavistock Clinic, 36
Questionnaire (RADQ), 139 teleological stance, 166
Reactive Attachment Disorder, tender years doctrine, 141
109-112, 115, 116, 135, 141 Theory of Mind (TOM), 165-168
reciprocal connectedness, 143 therapy. See psychotherapy
 Index 93

threat, 43, 51, 60, 66, 76, 94 visiting relationship, 89-92

time-out, 93 voice patterns, 108
Tinbergen, Nikolaas, 33
toddlers, 51 wariness, 59
traditional societies, 157 wet nursing, 14-15, 16
transactional process, 59, 163 Winnicott, Donald, 57, 128
transference, 117, 118 World Association for Infant Mental
transitional object, 4, 162 Health, 160
trauma, 118, 125, 145 World War II, 20-23, 26, 33, 76

Uganda study, 41, 42 Zaslow, Robert W., 135

Zeanah, Charles, 46, 111-116
video feedback, 125 Zero to Three, 7, 160-161
About the Author

JEAN M E R C E R is a professor of Psychology in the Division of Social

and Behavioral Sciences at Richard Stockton College and President of the
New Jersey Association for Infant Mental Health. She is also coauthor of
Attachment Therapy On Trial (Praeger, 2003).