BECOMING

ATTAC H
E D
First Relationships and How They Shape Our Capacity to Love

ROBERT KAREN, Ph.d.

 Digitized by the Internet Archive
in 2016 with funding from
Kahle/Austin Foundation

https://archive.org/details/becomingattachedOOkare
Becoming Attached
ALSO BY ROBERT KAREN

When the Shooting Stops

Top Dog/Bottom Dog

Becoming Attached

First Relationships and

How They Shape
Our Capacity to Love

ROBERT KAREN

Oxford University Press

New York Oxford
1998
Oxford University Press
Oxford New York

Athens Auckland Bangkok Bogota Bombay Buenos Aires

Calcutta Cape Town Dar es Salaam Delhi Florence

Hong Kong Istanbul Karachi Kuala Lumpur Madras Madrid

Melbourne Mexico City Nairobi Paris Singapore Taipei

Tokyo Toronto Warsaw

and associated companies in

Berlin Ibadan

Copyright ©1994 by Robert Karen

First published by Warner Books Inc., 1994

First issued as an Oxford University Press paperback, 1998

Oxford is a registered trademark of Oxford University Press
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or
transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or
otherwise, without the prior permission of Oxford University Press.

Grateful acknowledgment is given to quote from the following:

The Continuum Concept © 1977 by Jean Liedloff. Reprinted by permission of Addison-Wesley Pub-
lishing Company, Inc.

Hospitals and Children: A Parent’s eye View by James Robertson. Copyright © 1962 by the Tavistock
Institute of Human Relations. Reprinted by permission of International Universities Press, Madison,
CT, and by Victor Gollanz, London.
Infancy in Uganda by Mary Ainsworth. Used by permission of The Johns Hopkins University Press.

The Interpersonal World of

by Daniel Stern
the Infant ©
1987 by BasicBooks, Inc. Reprinted by per-
mission of BasicBooks, a division of HarperCollins Publishers, Inc.

Maternal Care and Mental Health (1951) by John Bowlby (WHO Monograph Series No. 2). Re-
printed by permission of the World Health Organization, Geneva.

The Nature of by Jerome Kagan. Copyright

the Child ©
1984 by BasicBooks, Inc. Reprinted by per-
mission of BasicBooks, a division of HarperCollins Publishers, Inc.

Your Child Is a Person by Stella Chess, Alexander Thomas and Herbert Birch. Copyright © 1965 by
Dr. Stella Chess, Dr. Alexander Thomas, and Dr. Herbert G. Birch. Used by permission of Viking
Penguin, a division of Penguin Books USA Inc.

Excerpts from Bowlby’s The Making and Breaking of Affectional Bonds, which appear on pages
J.

48^9, 50-51, and 363-364, were originally published in Sutherland, J.D. (Ed.). (1958). Psycho-
analysis and Contemporary Thought. London: Hogarth Press.

Library of Congress Cataloging-in-Publication Data

Karen, Robert.
Becoming attached: first relationships and how they shape our capacity to love / Robert Karen,
p. cm.
Originally published: Warner Books, 1994.
Includes bibliographical references and index.
ISBN 0-19-511501-5 (pbk.)
Mother and
1. infant. 2. Attachment behavior. 3. Mother and child. I. Title.
(BF720.M68K37 19971
306.874' 3—dc21 97-24822

98765432
Printed in the United States of America
on acid-free paper
For Rafie.
m
 CONTENTS

INTRODUCTION: How Do We Become Who We Are? 1

PART I. What Do Children Need?

1. Mother'Love: Worst'Case Scenarios 13

2. Enter Bowlby: The Search for a Theory ot Relatedness 26

3. Bowlby and Klein: Fantasy vs. Reality 40

4. Psychopaths in the Making:

Forty^four Juvenile Thieves 47

5. Call to Arms: The World Health Report 59

6. First Battlefield: ‘‘A TwO'Year^Old Goes to Hospital 67

7. Of Goslings and Babies:

The Birth of Attachment Theory 67

8. “What’s the Use to Psychoanalyze a Goose?’’

Turmoil, Hostility, and Debate 103

9. Monkey Love: Warm, Secure, Continuous 1 19

via Contents

PART II. Breakthrough: The Assessment of Parenting Style

10. Ainsworth in Uganda 129

11. The Strange Situation 143

12. Second Front: Ainsworth’s American Revolution 162

PART III. The Fate of Early Attachments

13. The Minnesota Studies:

Parenting Style and Personality Development 177

14. The Mother, the Father, and the Outside World:

Attachment Quality and Childhood Relationships 191

15. Structures of the Mind:

Building a Model of Fluman Connection 202

16. The Black Box Reopened:

Mary Main’s Berkeley Studies 210

17. They Are Leaning Out for Love:

The Strategies and Defenses of Anxiously Attached
Children, and the Possibilities for Change 220

18. Ugly Needs, Ugly Me:

Anxious Attachment and Shame 238

19. A New Generation of Critics: The Findings Contested 248

PART IV. Give Parents a Break! Nature-Nurture Erupts Anew

20. Born That Way? Stella Chess and the Difficult Child 269
21. Renaissance of Biological Determinism:
The Temperament Debate 289
22. A Rage in the Nursery: The Infant Day-Care Wars 313

23. Astonishing Attunements:

The Unseen Emotional Life of Babies 345

PART V. The Legacy of Attachment in Adult Life

24. The Residue of Our Parents:

Passing on Insecure Attachment 361

 Contents IX

25. Attachment in Adulthood:

The Secure Base vs. The Desperate Child Within 379

26. Repetition and Change:

Working Through Insecure Attachment 394

PART VL The Odyssey of an Idea

27. Avoidant Society:

Cultural Roots of Anxious Attachment 411

28. Looking Back: Bowlby and Ainsworth 426

APPENDIX: Typical Patterns of Secure

and Anxious Attachment 443

ACKNOWLEDGMENTS 446

NOTES 449

BIBLIOGRAPHY 469

INDEX 487
ll
Becoming Attached
 INTRODUCTION
How Do We Become
Who We Are?

Love is the good we all search for, and yet we have different conceptions

(and misconceptions) about what it is, ambivalence about how close we

want to get to it, doubts about whether we can achieve it or even deserve

it. Some of us repeat futile patterns with intimates, mates, and children
to the point where we may question whether we are capable of close, sat-
isfying relationships at all. At times it feels as if the shadow of our par-

ents hangs over us like a fate we cannot elude. And we wonder: How
much do our childhoods, and especially the quality of our first loving

bonds, determine whether we can get love right as adults?

When my son was two and a half, 1 saw a slightly older boy, perhaps

three, fall off a slide in the playground and bloody his lip. He stood,

threw his arms in the and started wailing, but he made no effort to
air,

approach or attract or even look toward his mother who was sitting
nearby on the bench. She promptly retrieved him, though, and sat him
on her lap. The little fellow continued to wail miserably, his body stiffly

unengaged with hers such that one might have thought he was being
her bag
held by a stranger. The mother, meanwhile, rummaged through
looking for a tissue, all the while lecturing in a nonstop, mechanical.
2 Introduction

somewhat irritable way that seemed strangely disconnected from the boy
struggling in her arms.
The child’s cut was eventually attended to, but plainly his emotional
needs were not. His mother held him with his lower body dangling awk'
wardly, and she never paused to respond to his cries with a cuddle or a
kiss or a loving gaze. Rather she kept up her strange stream of muttered
comments parallel to his ceaseless cries. I have heard people in therapy
say, “The only way my mother could show love is when I was hurt or

sick.” But this mother seemed unable to show love in the case of injury.
And I felt sure that the pain of not being heard and responded to was
greater for this little boy than the pain of his cut lip.
What causes a parent and child to interact this way? What impact
does the mother’s emotional coldness have on her boy? Did his panicky
response to his wound and his failure to approach his mother or even
look at her throughout his ordeal reflect his expectation, based on previ'
ous experience, that she would not be available emotionally? If what I

witnessed in the playground — the mother, in effect, rebuffing her son at

a moment when his need for emotional comfort was high — is typical of
her parenting style, how much of it is influenced hy her son’s tempera-
ment, which may be unusually demanding or irritable? How much by the
quality of her relationship with her husband (assuming she has one),
which may be distant and unsupportive? How much hy the responsibili-
ties on her and the backing she does or does not get from oth-
that press
ers? Could her response have been predicted hy the quality of parenting
she herself received?
Finally, to return to the issue raised above, what is this boy going to be
like in the future? Assuming, again, that what 1 saw with his mother is

typical, does it necessarily mean that his emotional development will be

compromised? That his intimate relationships will he poisoned? That he
will he clingy or isolated? Depressed when alone? Unable, despite a thou-
sand efforts, to feel good about himself? That he will be a remote or
sadistic father? Or will he manage to leave his early experience with his
mother behind, take nourishment from other sources, and become whole
in the process? What will determine which path predominates and can
he — or someone who loves him — do anything to alter it?

These aresome of the paramount questions facing psychologists today.

They are part of a many-pronged debate about what children need and
how to provide it, a debate about parenting style, day care, adoption, and
basic psychology that roars through the halls of academia and out into
 Introduction 3

the legislatures, the press, and the streets. It is a debate that effects not

only our children hut our hopes for ourselves, spilling over into
some'
psychother'
times bitter disputes about how people change and whether
dynamics or
apy should attempt to deal with early experience and inner
chemistry and behavior. The concept of attachment, born in
just
some forty years ago and nurtured to near maturity
British psychoanalysis

in the developmental psychology departments of American universities,

spins at the center of these scientific and social
whirlwinds. It encom-

passes both the quality and strength of the parent'child

bond, the ways

in which it forms and develops, how it can be damaged and repaired, and
the long'term impact of separations, losses, wounds, and
deprivations.

Beyond that, it is a theory of love and its central place in human life.

struggle to understand the parent'child bond touches us deeply

The
because we intuitively sense that our first relationships hold many clues

to how we’ve become who we are. We suspect that what happens be'
tween the little boy in the playground and his mother matters a great
deal. And the terrible certainty some of us have that we will re'enact the

worst aspects of our upbringing with our own children is not only wide'

spread but seems distressingly welhfounded. A disproportional number of

suggests that
child abusers were abused children, and clinical evidence
many lesser miseries are passed on as well. ^X/hat mother hasn t had the
chilling experience of making what she assumes will he a legitimate diS'
her
ciplinary statement to her child only to hear some hated aspect of
own mother’s voice leap from her throat?
There is no shortage of theories to explain this unwanted inheritance.
Indeed, until
But empirically verifiable explanations have been elusive.
scientific authority
the last two decades, nothing could be said with
let alone how
about almost any dimension of the parent'child bond,
aspects of relatedness, good and bad, are transmitted. What early experi'

feel that the world of people is a positive place

ences enable a child to
and that he has value? How does he become equipped with enough con'
to rebound trom
fidence to explore, to develop healthy peer relations,
adversity? Which of us are at risk of being parents
who will raise insecure
what can he done to support or change us? Today, with
children, and
mothers spending time in the home, and with families falling apart
less

reconstituted in new shapes and combinations,

understanding
and being
all this seems more urgent than ever.
Much begins in infancy, a time when certain fundamental themes of
4 Introduction

personality and relatedness originate. We have to move beyond our need

to be at the center of the universe, to be constantly adored and minis-
tered to. We need to learn to manage our rage, which in the infant is like

a cataclysm that seems to destroy everything in its wake. We need to

learn that forgiveness and repair are possible. We need to develop the
faith that we can nourish and be nourished by others. And yet because
our own early experiences are shrouded in amnesia, and because babies
cannot give reliable reports on what’s going on inside them, infants are
strangely alien beings about whose inner processes we can only speculate.
Who they really are, whether they’re truly “human,” how they traverse
the developmental stages that psychologists have bequeathed to them,
and what their parents really mean to them have always been matters of
guesswork, romance, and fantasy, some of it of the most absurd kind. The
primitive level of discourse is evident if we recall the starkness of the
nature-nurture debate as it existed as recently as fifty years ago.
One strain of thought, represented by the eugenics movement, saw
the infant as essentially a genetic construction and believed that the
human race could be perfected if substandard seedlings — those with bad
health, low intelligence, or poor moral character — could be weeded out.
Behaviorally or emotionally disturbed children, like mentally retarded
children, were seen as unfortunate products of heredity who should not
be allowed to propagate. The genetic view achieved its greatest respecta-
bility in Arnold Gesell, the eminent American pediatrician and pioneer
in developmental psychology, who first brought attention to the child’s
inborn maturational timetables. Although Gesell believed that children
should be treated kindly, he was convinced that children will pretty
much be what they will be regardless of how they are raised: “The inborn
tendency toward optimum development,” he wrote, “is so inveterate that
Ithe childl benefits liberally from what is good in our practice, and suffers
less than he logically should from our unenlightenment.”*
At the opposite end of the sp’iectrum was behaviorist John B. Watson,
a psychologist who believed that children were entirely products of their
environment, pieces of clay — or blank slates, as John Locke had written
two centuries earlier — that parents and other social forces mold into the
final shape. He held that a mother’s affection was potentially dangerous
to the child’s character, that picking up a child when it cried or feeding
iton demand were pernicious forms of coddling. In his famous 1928
book on child rearing (published the same year as Gesell’s book quoted
above), Watson wrote:
 Introduction 5

Treat them as though they were young adults. Dress them,

hathe them with care and circumspection. Let your behavior
always be objective and kindly firm. Never hug and kiss them,
never let them sit on your lap. If you must, kiss them once on
the forehead when they say goodnight. Shake hands with
them in the morning. Give them a pat on the head if they

have made an extraordinary good job of a difficult task.^

Yet a third position was taken by psychoanalysts. Although certain

aspects of psychoanalytic thinking stressed the hereditary makeup of the
child — in particular the strength of his sexual and aggressive drives and
his ease or difficulty in handling them — many analysts saw the infant'

mother relationship important to healthy emotional development.

as

They, too, were environmentalists; but, unlike Watson, they expected

the environment to be manifestly loving. Of this latter group, the late

John Bowlby was among the most important. Bowlby, the founder of
attachment theory, believed that it is in our first relationship, usually

with our mother,* that much of our future welLbeing is determined. In

sharp contrast to both Watson and Gesell, he believed that the infant
needs and actively seeks affectionate relationships. He wrote:

When a baby is born he cannot tell one person from another

and indeed can hardly tell person from thing. Yet, by his first

birthday he have become a connoisseur of people.

is likely to

Not only does he come quickly to distinguish familiars from

strangers but amongst his familiars he chooses one or more
favorites. They are greeted with delight; they are followed

when they depart; and they are sought when absent. Their

loss causes anxiety and distress; their recovery, relief and a

sense of security. On this foundation, it seems, the rest of his

emotional life is built — without this foundation there is risk

for his future happiness and health.^

*For simplicity’s sake, I

primary caregiver as the mother, as Bowlby
will usually refer to the child’s
nonrelated adult can also
and many others have done, but it should he understood that a father or a
will usually refer to the child as “he.” This convention,
which has become
play this role. Similarly, 1

think, the most convenient for the reader, especially in sentences

where
controversial lately, is still, 1

would cause confusion. But

mother and child are being discussed and using a second female pronoun
referring to children ot both sexes.
whether the child is called “he,” “she,” or “it,” 1 am almost always
6 Introduction

Bowlby was part of a broad midcentury trend in philosophy and psy^

choanalysis that was moving away from seeing people as isolated energy
systems and toward seeing them as embedded in relationships. Unlike
other analysts, however, Bowlby was also a research scientist who bot'
rowed from animal studies, systems theory, cognitive psychology, and
behaviorism in a huge intellectual effort to bring Freud’s early theories of

motivation up to date. His work was controversial and rejected by most

of his peers.
But in the early 1960s Mary Ainsworth, a Canadian psychologist who
had worked with Bowlby and shared his views, began conducting re-

search on babies at Johns Hopkins University that seemed to verify

much of what Bowlby had been claiming. In a technique that was ex'
tremely unusual at the time, Ainsworth and her colleagues closely
observed mothers and children in their homes, paying careful attention
to the mother’s style of responding to the infant in a number of funda^
mental areas: feeding, crying, cuddling, eye contact, smiling. At twelve
months the infant and mother were taken to the lab and observed in
what Ainsworth called a “strange situation.” What she discovered there
would have repercussions throughout psychology, psychiatry, and psycho^
analysis, making the paired names “Bowlby and Ainsworth” a cause for
reverence and rebuke to the present day.
Ainsworth was able to demonstrate that something she called “secure
attachment” between infant and mother was of crucial importance to the
child’s psychological development and that a certain style of mothering
— warm, sensitive, responsive, and dependable — was the key ingredient
in bringing this about. Secure attachment was seen as a source of emo'
tional health, giving a child the confidence that someone will be there
for him and thus the capacity to form satisfying relationships with others.
Insecure (or “anxious”) attachment, on the other hand, could reverber'
ate through the child’s life in the form of lowered selhesteem, impaired
relationships, inability to seek help or to seek it in an effective way, and
distorted character development. Subsequent research suggested that
close to a third of the children in middle^class American homes suffer

from insecure attachment and that it tends to be transmitted from one

generation to the next. In poor, unstable homes the percentage is higher.
Attachment research aroused great excitement in the field, partly
because it finally offered empirical evidence for the origins of some of our
psychological problems, but also because questions about child rearing
 Introduction 7

that had only been speculated about could now be answered with greater
authority. For years, to cite an obvious example, mothers bad been

warned against picking up their babies when they cried. It seemed con-
trary to nature and intuition, but behaviorist theory asserted that picking

up the kid reinforced the crying, and enough you d have a

if you did it

monstrous crybaby on your bands. Attachment research seems to have

disproved this, at least as a general principle.

Indeed, the great promise of attachment theory has been the prospect
of finally answering some of the fundamental questions of human emO'
tional life: How do v/e learn what to expect from others? How do
we
come to feel the way we do about ourselves in the context of an intimate
relationship? How do we come to use certain futile strategies in a vain

we (often unconsciously) feel was denied us as chib

effort to get the love

dren? How do we pass on our own parenting style to our kids? Attach-
ment researchers have examined the fantasies that accompany insecure
attachment. They have attempted to show how different types of in-
security get played out in various childhood disturbances — including
excessive aggressiveness, withdrawal, clinginess, and inability to attend
to schoolwork. Perhaps its most startling and controversial claim is that
insecure attachment, which shows up at twelve months, is predictive of

behavior not only at three, five, seven, or fourteen years of age which
has been well established in research — but also at twenty, thirty, and

seventy, as people make romantic choices, parent their own children, get

into marital squabbles, and face the loneliness of old age. Equally impor-

tant, attachment researchers have attempted to show how insecure pat-

terns of attachment can change, whether in childhood, as
adjustments

are made in the family, or later, as the adult attempts to work

through his

early experiences.
The debate over all this has been heated and at times filled with

antipathy, as science took a back seat to politics and personal

feelings.

Not only have established theories been challenged, but styles of child
raising, too; and psychologists, like others, have children, not always
raised according to the Bowlhy-Ainsworth precepts. The day-care issue
tull-time
has been most explosive. Attachment workers tend to see
day care in the first year as a risk, especially for certain children, and
Pennsylvania State University,
Jay Belsky, an attachment researcher at
voiced the concern that if you put your baby in substitute care for
more

than twenty hours a week, you are invoking a serious risk of his becom-
8 Introduction

ing anxiously attached — which could skew all his subsequent efforts to

relate with the outside world. Such assertions, needless to say, have
drawn tremendous fire and bristle with political implications.
Attachment theory has also been faulted by feminists and working
mothers for its nearly exclusive emphasis on the child’s relationship with
the mother. This has been a problem throughout developmental psychol-
ogy, partly because mothers are usually the primary caregivers and partly
because they are, therefore, more accessible to study. The emphasis on
the primary caregiver and on infant experience is important, because so
much gets established in that first relationship, so many psychological
forces are set in motion. As important as it is, however, observing it in a
vacuum leaves out a lot: the father’s relationship with the child, which
can have a potent effect, especially, in many families, as the child gets
older and the father takes a more active role; the parents’ relationship
with each other, which may be warm, hostile, supportive, or undermin-
ing; sibling relationships and sibling order; general family dynamics,
which may include other important figures, like grandparents who live in
the home; not to mention community and cultural values. When these
influences are not included, mothers often feel that the full weight of the
child’s psychological well-being rests on their shoulders. Most attach-
ment theorists do not believe this — but there has been an evangelical
enthusiasm for its findings which, as it spread into the popular realm,
included an eagerness for mother improvement and mother blaming.
Classical analysts, behaviorists, and those who emphasize biology and
heredity have raised numerous objections as well. Opponents question
attachment theory’s Darwinian claim that the newborn is “prewired” to
seek a relationship with the mother. They reject attachment’s explana-
tion of how infants understand, retain, and incorporate experience.
Perhaps most important, they dismiss the idea that the vast complexity
of the infant-mother experience can be reduced to three or four attach-
ment styles. They are perturbed by attachment theory’s failure to pay
adequate tribute to inborn temperament, which, some say, may be the
real culprit in the tenacity of the patterns that attachment research
has described. Indeed, in recent years, with the study of identical twins
separated at birth, many of whom show remarkable psychological simi-
larities despite having been raised in different environments, and the
study of adoptive versus biological children in the same family, the
genetic view has had its own renaissance, which can neither be ignored
nor underestimated.
 Introduction 9

Although fur still flies, the debates provoked by attachment theory

and research have caused all sides to make refinements in their thinking,

so that now more than any time before, the nature-nurture question has
the potential to go beyond its simplistic extremes and toward a synthesis
that brings us closer to true understanding.
But I am getting ahead of my story, which is about the discoveries and
intellectual battles of the last fifty years that brought us from knowing
precious little about the infant-mother bond and parent-child relation-
ships in general — indeed, about whether family ties mattered to the
child at all — to the current flowering of knowledge. The story is told

through the work of John Bowlby, Mary Ainsworth, and their followers,
but at every step of the way their work has been enriched by others,
including opponents, and I’ve tried to give a sense of the many tribu-

taries that have flowed into and helped form this large river of under-

standing. I expect that for many readers this will be not just a voyage of

discovery in child emotional development and its pertinence to adult life

but a voyage of personal discovery as well, for it is almost impossible to

read about this material without reflecting on one’s own life as a child, a

parent, and an intimate partner in love or marriage.

It often happens in science — for example, in brain studies, where the

destruction of a particular organ or structure has proved to be the break-
through in understanding its function — that advances in knowledge
come in the wake of traumatic circumstances. In the case of the infant-

mother bond, the trauma was maternal deprivation — children who’d

been separated from their mothers at a very early age or who had no
mother at all. In the history of the attachment idea, their plight might be
considered the starting point.
 Part I

What Do Children Need?

 1

Mother-Love :

Worst-Case Scenarios

In 1937 New York psychoanalyst David Levy reported the case of an

eight'yeat'old girl I’ll refer to as Anna whom he’d observed at the Insti'
tute for Child Guidance in New York City. Anna had been adopted at
six and a half after an illegitimate birth and years of having been shifted
around in foster care. The new parents apparently found her to he an
appealing and responsive child. But they were nonplussed by the fact
that when she was shown her new home and her room, Anna had
first

no apparent emotional reaction. Her adoptive mother soon had further

cause for concern. While she continued to find Anna “affectionate on
the surface,” after a few weeks she complained that the child seemed
incapable of real affection. She “would kiss you but it would mean noth'
ing.” Her husband assured her that there was nothing wrong with the
child, that she only needed time; hut after a few months he felt the same

way. In fact, during this period, he had come to view Anna as deceitful

and evasive. Levy said he felt particularly confident that these were lov'
ing parents. He noted that their natural child, four years older than
Anna, was affectionate and well adjusted.
The parents tried to correct Anna’s deceitful behavior, hut nothing
14 What Do Children Need?

they did seemed to have an impact. A psychoanalyst was consulted. He

recommended that all disciplinary action cease and that the parents suh'
stitute greateramounts of warmth. After a while, they felt that this, too,
was ineffective. Anna’s teacher complained that although she had good
intelligence and did well in her subjects, she was inattentive and lacked
pride in her work. A year and a half after adoption, none of Annas ici'

ness seemed to have thawed. “You can’t get to her,” the father told Levy.

The mother said, “1 have no more idea today what’s going on in that

child’s mind than 1 knew the day she came.”^

In perhaps the most upsetting sentence of his report. Levy said that
cases like this were “familiar to everyone who works with children.” But,

in fact,no one had ever reported them before. Levy, one of the leaders of
the child guidance movement in the United States, whose purpose was
to effect social reforms in order to prevent emotional damage to chiL
dren, would go on to great prominence in psychiatry. He was one of a

handful of pioneers, a surprising number of them working in New York

City at around the same time, who first brought attention to such deeply
disturbed children and to a problem that would soon he an international
cause: “maternal deprivation.”

One senses intuitively that for the tiny child, motherTove, whether it

comes from the biological mother or someone who has taken her place,

promotes welLbeing. In the sunshine of her love, we grow and develop,

take an interest in things and people, learn, acquire skills, become a

proud member of the family. We babble in response to her, smile with her
and at her, play special games with her in which we learn to take joy in

interaction, check back earnestly to make sure she’s still there when
we’re old enough to crawl away. We feel in a sense a part of her, her won^
derfulness making us wonderful, too. Later, as we become aware of our

separateness, we feel known by her in a way that helps us know ourselves

and that warms us, making us feel loved and secure. These very early
years would seem to represent an age when, for many infants, all the

corny things ever recited about mother^love seem to glow with a truth
untarnished by the ironies, failings, and complexities of later years.

This, despite the fact that, from the baby’s point of view, mother^love
is not always shining. There are times when mother does not come,
when mother is insensitive, when mother does not please us because she

doesn’t respond or is too stimulating or because we are beyond soothing.

Terrible hatreds boil up inside us and the earth darkens. We become
 Mother-Love 15

filled with murderous wishes. We wail inconsolably, until our blood veS'
sels seem ready to burst. Hell, pure bell, is everywhere. Mother is hell.

Baby is hell. But then the sun comes out and the steaming toxins evapo'
rate. Mother^love, pure and unclouded, shines again.
Of course, for many babies, even from their earliest days, mother^love
is compromised in fundamental ways. Mother is cold and unresponsive
much of the time; mother is unable to meet the needs of a very demand'
ing or temperamental infant; mother is too preoccupied with herself and
her feelings of deprivation to give of herself fully; mother is hitter and
angry, and her rage boils over in fits. And yet even a troubled, inconi'
petent, or neglectful mother is present in important ways. She feeds,
she changes diapers, she takes baby for a walk, she rocks him to sleep.

Through all her failings, failings that may haunt her child into his adult'

hood and old age, rays of mother'love shine through, and the baby has a
place where he knows he belongs.
But what is it like for infants who lack mother'love entirely or have it

taken away prematurely? Should we follow the analogy all the way, and

assume that without it they are as perishable as a plant without sunshine?

Do such infants lack the essential template on which to develop the ahih
ity to love and to relate to others? What if a child had its basic physiO'

logical needs met — food, shelter, a clean and comfortable environment

— but lacked the loving relationship? What if he had to wait a while for

a mother, till he was a year or two old? Would he he able to recover lost

time and become like other children? What if he was shifted around in

his early years, posted like a parcel from one mother figure to another, as

often happens in foster care or when an extended family cares for a baby
after his mother’s death? Would a series of mothers be as good as one?
What if the child was finally adopted, like Anna, at six or eight? Could
normal life commence then? It was in the examination of such extreme
questions that our understanding of the infant'inother bond began.
In 1760 a Spanish bishop wrote in his diary, “In the foundling home
the child becomes sad and many of them die from sadness.”^ But to many
physicians and scientists of the next two centuries, observations such as
this seemed needlessly sentimental and even anthropomorphizing: These
little creatures were not yet people and could not have real human feeh

ings. They didn’t become sad or lonely. If they deteriorated, there were
other reasons: They were not hardy; they were broken by disease. But

they certainly did not need others in any emotional or psychological way.
Even with the romance of motherhood and maternal bliss that acconi'
16 What Do Children Need?

panied the industrial revolution — when fathers and their work were
removed from the home environment and the home of popular imagina^
tion became a feminine haven, free of the competitive harshness of the

male workplace — this detached professional attitude toward maternally

deprived infants did not appreciably change. Orphaned children or those
who had been given up by their parents were still typically raised in insti^

tutions in their early years rather than placed immediately into foster
care or an adoptive family. Eugenics proponents opposed early adoption
on the grounds that it was better to let the child grow a bit and see if it

was defective in any way. Prospective parents were encouraged to put off
adopting until the child was old enough to match his characteristics with
theirs — would the blue eyes stay blue? What color would the hair be

when it came in? Was the intelligence quotient high enough?

The preoccupations with antisepsis had meanwhile led to sterile emo'
tional conditions wherever children were cared for, with minimal han^
dling or stimulation. Hospital policy typically forbade parental visits
when a child was ill, and it was not uncommon for infants and young
children to be in the hospital for over a year without ever seeing their
mother. Across a broad front, maternal care and its substitutes placed

second to other concerns, and few questioned whether anything was lost

in the process.

But as child psychiatry came into being in the 1930s, as the first hoS'
pitals began to set aside psychiatric wards for children, and as research
attention was focused on children who were brought into the child guid'
ance centers that had sprung up around the nation, a number of disturb-
ing trends were noticed among children raised without a stable mother
figure, trends that, if taken together, constituted a silent plague.

David Levy encountered children like the affectionless Anna somewhat

accidentally. He was studying another problem, “maternal overprotec-
tion” — that is, the emotional impact of mothers who are anxious, overly

cautious, and generally infantilizing of their young. As part of his study,

he had set up a control group of children who had lacked maternal care
in their early years and had failed to make bonds with adoptive parents
later on. This control group soon seized his attention. Other cases were
as disturbing as Anna’s. One concerned the illegitimate son of a woman
of high status. The infant had been placed with an agency shortly after

birth, put in an orphanage at the beginning of his second year, trans-

ferred to a boarding home at twenty-seven months, and finally adopted

 Mother'Love 17

four months before his third birthday. The adoptive mother complained
that the child did not respond to affection, rejected fondling, was un-
manageable. By the time the boy was brought to the Child Guidance
Center, Levy said, “The mother felt she had been taking punishment for
a year and could stand it no longer.”^

Levy’s cases shared many similar features: children shifted around,

adopted after several years, often pleasant and affectionate on the sur-

face, indeed indiscriminately affectionate, but seemingly indifferent

underneath; lacking pride; and displaying incorrigible behavior problems
that often included sexual aggressiveness, fantastic lying, stealing, tem^
per tantrums, immature or infantile demands, and failure to make mean-
ingful friendships. And then there were the parents: pained, frustrated, at

wit’s end. None of the children seemed able to respond to psychotherapy.

In one extraordinary case, a child who had shown no progress in therapy

was sent by her adoptive parents to Vienna, where she lived with an
Adlerian psychiatrist who treated her for three years. The child returned

home with the same emotional flatness and more alarming aggressive
symptoms.
Levy concluded that the children he’d observed were suffering from a
form of emotional starvation he called “primary affect hunger.” By this

he did not simply mean hunger for affection, but rather for the full spec-

trum of human feelings, even including hostility, that arise from daily
interaction with a mother. “Is it possible,” he asked of these maternally
deprived children, “that there results a deficiency disease of the emo-
tional life, comparable to a deficiency of vital nutritional elements with-
in the developing organism?”^ Levy’s paper on this subject, published in
the American Journal of Psychiatry, made little stir and was quickly forgot-

ten. But unbeknownst to Levy, his question was being asked by other
researchers as well.
One of them was Loretta Bender, who headed the new child psychia-

try service at New York’s Bellevue Hospital. Many of the children on

Bender’s unit had debilitating organic psychiatric syndromes, like pos-
tencephalitic syndrome or autism. Others had no known organic deficits

but were nevertheless severely behaviorally disturbed. The majority of

them came from two major adoption agencies.

The children from the Jewish Foundling Homes were given excellent
pediatric care and were physically healthy. But they had had few social

contacts and almost nothing to play with. Bender believed that this
accounted for their delayed speech and retarded behavior. After living
18 What Do Children Need!

in these conditions for three or three and a half years, she wrote, “they

are placed in a foster home, to which they often cannot adjust, and are

subsequently tried in five or six homes, each time becoming more of a

problem.’’^
Stella Chess, who worked on Bender’s unit at the time, recalls that
these children had numerous problems, from inadequate toilet training,
to ignorance of how the world worked, to superficial and indiscriminate
affectionateness. “They were affectionate to everybody,’’ she says. “The
foster parents would find that they would somebody Daddy who they
call

saw for the first time in the street. One mother got new linoleum, which
was the height of luxury in those days, and the kid the next day was
pounding holes all through it — he was imitating the guy who put it in.

These children had no experience whatsoever, no experience with what

was good or bad, so pretty soon they’d be moving to another foster home.
And pretty soon they were unplaceable. And at the age of five or six,

they’d come onto our unit.’’^

The second organization, the Angel Guardian Home, a Catholic

run institution, made “no effort to make the child feel secure.” Like
many such home believed it was wrong for a
institutions of the day, the

child to form an attachment to a person who wasn’t going to be a perma^

nent parent. They therefore moved the children from one home to the
next every six months. By the time the children were five years old, they

not only lacked attachments, they appeared to lack any feeling for others
at all.^

They have no play pattern [Bender wrote], cannot enter into

group play but abuse other children, and cling to adults and
exhibit a temper tantrum when cooperation is expected. They
are hyperkinetic and distractible; they are completely con^
fused about human relationships, and . . . lose themselves in a

destructive fantasy life directed both against the world and

themselves.^

She classified them as psychopathic personalities and attributed their

condition to emotional deprivation during their formative years.

In the pediatric ward at Bellevue, as on all pediatric wards, there was

great concern for sterility. Infections had been known to spread viru-
 Mother^Love 19

lently wherever children lived in groups, and since Pasteur’s discovery of

germs in the mid-nineteenth century, there was a constant search for

ways to make the infant environment as antiseptic as possible. In

Bellevue, for instance, a single white coat was used by each staff member
who handled a particular baby. The coat was then hung on a hook inside

out, until the next person attending to that child put his arms through
it. Signs admonished staff members on the need for sterility, and han-
dling a baby was considered dangerous to its health. Babies in many
institutions, like the Angel Guardian Home, were typically propTed, the

bottle propped up for them so that they wouldn’t have to be held during
feeding. This was considered ideally antiseptic, and it was labor-saving
as well.

That infants in such circumstances did not do well was no secret. It

had been reported, for instance, in 1915 that infants admitted to ten asy-
lums in the eastern United States had mortality rates of from 31.7 per-
cent to 75 percent by the end of their second year. In 1920 a German
medical journal reported that in one of the great foundling homes of
Germany seven of ten infants died in their first year. Although condi-
tions had greatly improved, particularly in the realm of hygiene and
nutrition, there was still a 10 percent mortality rate in the best European
and American institutions. The condition leading to death was some-

times called hospitalism, or failure to thrive, and seemed to be accompa-

nied by what looked like depression and lost hope. From 1912 to 1922
malnutrition was the most frequent diagnosis of children in Bellevue’s
infant wards, for which a high death rate was reported. As nutrition
improved, the new diagnosis used to account for the sickliness of the

ward infants was infection, and those babies that withered away were
assumed to be suffering from chronic infections.
In 1942 pediatrician Harry Bakwin, in a paper poignantly titled “Lone-
liness in Infants,’’ reported:

To lessen the danger of cross infections, the large open ward of

the past has been replaced by small, cubicled rooms in which
masked, hooded and scrubbed nurses and physicians move
about cautiously so as not to stir up bacteria. Visiting parents
are strictly excluded, and the infants receive a minimum ot

handling by the staff. Within recent years attempts at isola-

tion have been intensified, and a short time ago there was
20 What Do Children Need?

devised a box equipped with inlet and outlet valves and sleeve
arrangements for the attendants. The infant is placed in this

box and can be taken care of almost untouched by human

hands.^

Why, asked Bakwin, do infants in hospitals fail to gain weight when

given a diet they would thrive on at home? Why do they sleep less, smile

rarely, hardly babble? Why don’t they respond to a smile or a coo? Why
do they seem so listless and unhappy? Why don’t they seem to take any

pleasure in eating? Why do infections that last a day or two at home pet'

sist for months on end on the ward?

When Bakwin took over the pediatric unit at Bellevue in 1931, he
changed the routines. He took down the old signs emphasizing antisepsis

(“Wash your hands twice before entering this ward’’), and put up new
ones: “Do not enter this nursery without picking up a baby.” Nurses were
instructed to fondle babies periodically and to sit them on their laps.

Infection rates went down.^^

In Bellevue child psychiatry a similar change was occurring. “I still

remember,” Stella Chess says, “one skin^andTones infant, a failure to

thrive, who landed on the psychiatric unit. This was a kid who was
placed in a group home for infants, and he was starving. Loretta Bender
put a single nurse in charge of him, and he gradually began to eat more
and more. I never knew what happened to him in the end, but I watched
him fill out. The nurse on the unit got terribly attached to this kid and
the kid to her — you could see it. She’d change his diaper and the poor
little thing, who had hardly the energy to move, would try to lift his butt

up so that she could slip the diaper underneath.”^

Bakwin similarly found that the symptoms of hospitalism began to diS'

appear when the baby was placed in a good home, often with a rapidity
that he described as “amazing.” He wrote that “the baby promptly
becomes more animated and responsive; fever, if present in the hospital,
disappears in twentyTour to seventy^two hours; there is a gain in weight

and an improvement in the color.”

Even intelligence was found to suffer in the institutional environ^

ment. Another researcher. Bill Goldfarb, a psychologist (and later a

physician and psychoanalyst) who worked with Jewish Family Services,
compared children raised in institutions with those raised in foster
homes. In a remarkable series of studies, he found that institutionalized
children at almost three years of age not only suffered from major deficits
 Mother'Love 21

in their ability to form relationships and function maturely, but had a

mean IQ of 68, which is mildly retarded, while the mean IQ of a
matched group of children in foster care was 96, which is average.
Bender had come to similar conclusions regarding the pernicious effect

of institutionalization on intelligence.

The issue of intelligence and retardation was a controversial one,

which affected adoption policies. Arnold Gesell, the great Yale psycholo'
gist, was among those who opposed early adoption because only time

would tell which institutionalized infants were retarded and therefore

undesirable to prospective parents. The pioneering child psychiatrist
Milton Senn, also at Yale, argued that Gesell’s practice of diagnosing
institutionalized infants as suffering from “minimal brain damage” was
itself mistaken and that it only served to stigmatize the children and pre-
vent their adoption. In a revolutionary move, Senn favored early adop'
tion, which he believed would prevent retardation, a policy that was

sharply criticized by Gesell and others.

The debate might have been settled by Harold Skeels and his cob
leagues at the Iowa Child Research Welfare Station, if their work during
the 1930s had not been dismissed and forgotten after being scathingly
attacked for lack of scientific rigor. Skeels, who was the first to note the

language deficits of deprived children, began with a study of children in

severely depriving homes. Observing that older children in such homes
had IQs that were consistently lower than their younger siblings, Skeels

concluded that, over time, a severe lack of adequate loving attention

took its toll on intellectual functioning. In his next study Skeels took a

group of preschool children in a sterile orphanage and gave them an

ongoing nursery-school experience. Their IQs, which averaged 82,
remained stable during this period, while those of a control group from
the same orphanage declined. Might they improve even more if they had
something approaching real maternal care?
This question led to Skeels’ most original and daring study. He placed
thirteen institutionalized children, all under the age of two and a half, in

a home for older feeble-minded girls. The children were each “adopted”
by one of the older girls (or in some cases by an attendant), who took
over many typical mothering functions. The touching simplicity of this

study lay in the fact that the retarded hut affectionate girls proved in
these circumstances to be adequate surrogate mothers for children this
age. Over a mean span of nineteen months the average IQs of the mater-
nally deprived toddlers in their care rose from 64 to 92. Seven of the
22 What Do Children Need!

thirteen were later adopted by outside families, where they maintained

their intellectual functioningd^ Some thirty years later, when Skeels

received the Joseph P. Kennedy Award in belated recognition of his work

with disturbed and retarded children, he was introduced by a welhspO'
ken, tuxedoed young man with a master’s degree. He reported that he
had been one of those children, depressed, withdrawn, deteriorating,
“who sat in the corner rocking.’’ As part of Skeels’ study, he had been
given over to a retarded girl for care and later adopted.

Despite the evidence from the emerging fields of child psychology and
psychiatry and from the work of a few radical pediatricians like Harry

Bakwin, few people in or out of the medical profession had any idea of
the suffering that infants endured when they did not get the routine flow
of affectionate care that normal parents cannot help but give. The iner^

tia of old views and old ways of doing things combined with a resistance

to painful truths had quietly smothered the evidence. The preliminary

medical and scientific views also helped people to distance themselves
from children’s suffering. It was temporary. It was hereditary. It wasn’t

really so bad. And the studies that suggested otherwise were flawed and
should thus be dismissed. The resistance to seeing the pain of deprived,

neglected, or abused children has a long history. Indeed, there has been
some resistance to recognizing that children suffer at all. Many psychia-

trists long denied that children were capable of experiencing depression.

And, as we know, only recently has our society recognized the preva-

lence of child abuse, which was until the last decade thought to be so
rare as not to be worth considering.
Regardless of what theories people adhered to in order to maintain
this emotional distance, it seems plain now that the distance was needed

because the suffering of children, especially when it is great, is terrible to

witness. Even to see a little hahy not have his spontaneous gestures
responded to is heartrending; and the decline that follows is much worse.

People who work with suffering children are sometimes the most hard-
ened. To have to experience for forty hours a week the agony of a child

who is too depressed to eat is a great deal to endure. Some way must he

found to defend oneself from empathizing with such pain or life will not
seem livable.

Also, in the case of institutionalized children, budgets and routines

would he threatened with disruption by any effort to give each child
individualized care. How would such a thing he paid for? To allow par-
 Mother-Love 23

ents to visit their hospitalized children would disturb the nurse s smooth
routines, especially since it was known that children who had furiously
protested their parents’ departure and had, at some length, finally settled

down to a state of relative calm became upset and unmanageable after a

parental visit. To change adoption policies such that children would be

given away at birth, before they suffered institutionalized deprivation,
would require a reeducation of numerous bureaucracies, public and pri'

vate, not to mention the parents they deal with.

But then, in 1947, an unusual film, which would eventually be seen by
thousands upon thousands of health care workers and student trainees,
was shown to a group of physicians and psychoanalysts at the New York
Academy of Medicine. It had been made by Rene Spitz, a psychoanalyst

and a European emigre who was said to have been analyzed by Freud.
He was something of an original, considered eccentric by some of his

peers, not least of all because he often carried with him a camera, either
16mm or still. He also had an unusual interest in observing infants and
children. He had studied with two prominent Viennese psychologists,
Charlotte Biihler and Lotte Schenk-Danziger, in the early 1930s, and he
was trained to administer baby tests. Once in America, he began his own
research, much of it supported by the fees earned from wealthy analytic
patients. Spitz had already published a paper on which the film he was

showing this night was based, but, like those written by Levy, Bender,
Bakwin, and Goldfarb, it had caused little interest.

The film was called Grief: A Peril in Infancy. It was silent, crude,

black-and-white, with occasional title cards to explain what was being

seen. We are in an unnamed institution. In grainy, flickering images, we

are shown Jane, a little black baby, just after her mother had been forced
to leave her for what would turn out to be a three-month period. She’s a
happy, approachable baby, smiling and giggling as an adult observer plays
with her. We are then shown Jane one week later. It is painful to recog-

nize that this is the same child — depressed, eyes searching, completely

unresponsive except, finally, for a tremendous, hopeless, frowning wail.

A kindly male observer (actually Spitz himself) cannot soothe the child.

She kicks and sobs in terrible agony. A title card tells us that the despair-
ing expression, the tears, and the moaning are unusual at this age and
that they lasted the entire three months her mother was gone. Looking
at this child, we feel we are experiencing sadness at its ultimate depth,

the most profound grief imaginable.

We are now shown a seven-month-old white baby with a happy.
24 What Do Children Need?

delightful face and inquisitive eyes — an engaging, irresistibly sweet

child, playing with Spitz, shaking hands, allowing herself to be picked

up, exploring his face. She, too, is a new arrival. The scene changes to a

few weeks later. The baby now looks fretful and sad, has bags under her

eyes. She is not interested in playing and doesn’t respond to a little bell.

She looks as been through an unspeakable ordeal. She allows her^

if she’s

self to be picked up by Spitz, but her crying does not stop. She seems to

be in a state of terror.

Another child, lying dejected on a cot. She does not respond to an

observer’s efforts to engage her. She refuses to make eye contact. After

raising her head briefly and without much interest, she puts it back down
on the cot. She is reminiscent of hospitalized mental patients who are so

depressed they don’t want to get out of bed. The camera lingers on the
little girl’s blank, dazed face. At last her mouth forms a wide, hollow cry.

The babies we are viewing are in a Mexican foundling home where

they receive excellent nutritional, medical, and hygienic attention, but
where there is only one head nurse and five assistant nurses for forty-five
babies.

A title card explains that if the mother returns in less than three
months, the recovery is rapid. But if the separation lasts much longer, the

baby continues to display “a frozen, passive, apathetic attitude.” Such

children, we are told, no longer weep, just wail thinly. They cannot sit or
stand. They cannot be engaged. They do not explore. We are now pre-

sented with the evidence.

The next baby looks like a concentration camp victim. It is eight

months old, but appears to be only three. It seems almost nonhuman.

Now a baby of nine and a half months. Can he really be that old? His
shaking is pitiful, his little fist up against his head. A title card explains

that his motions are bizarre and “reminiscent of psychiatric distur-

bances.” His despair seems beyond the ability of language to convey.
A ten-and-a-half-month baby. She is emaciated and looks strangely
disturbed. She’s a beautiful baby, with big, dark eyes, but in another
world.
Now a fifteen-month-old who is the size of a three-month-old. She
seems to have a skin disease. She smiles at Spitz’s approach and then
becomes fretful. She just lies there now, as if sick, crying and gnawing her
hand.
Title card: “The cure: Give mother hack to baby.”
 Mother'Love 25

It’s Jane again, the very first haby, who had seemed lost in incon-
solable grief within a week of her mother’s absence. Now, three months
later, her mother has returned, jane is playing with a female observer,
clapping and laughing and allowing herself to be held. Her world seems
once again a place of happiness and sunshine.
In his paper. Spitz had compared these babies to another set of infants

that were being raised at a prison nursery in upstate They New York.

were the children of delinquent women who were pregnant when sen^
tenced. Unlike the founding home babies, the babies in the prison nurS'

ery had regular access to their mothers. They thrived.

The prison babies are shown in the film, exploring, climbing, toddling

and flopping about, a stark contrast to the broken babies we’ve just seen.

A healthy and determined little boy is climbing over his crib wall into
another baby’s crib, head first, legs in the air. The end.
The effect of the film on the audience was emotionally catastrophic.

Nothing like it had been seen before. When it was over, one prominent
New York analyst approached Spitz with tears in his eyes. “How could
you do this to us?” he said.^®

Spitz, whose reputation as a leading analyst and developmentalist

would grow steadily over the next two decades and whose book. The First

Year of Life, would be read by professionals worldwide, was soon the chief
spokesman in the United States on the dangers of maternal deprivation,
and he continued his assault on the issue for many years. The impact of

the film was not immediate, and the acceptance of its basic tenets was

hardly widespread. Its conclusions were hotly debated for a long time. As
late as 1981, a textbook in developmental psychology disputed them.^^
But a spark had been ignited. Another maternahdeprivation researcher,
a self'Confident young Englishman named John Bowlby, would fan it into

a fire.
 2

Enter Bowlby:
The Search for
A Theory of Relatedness

In a long'forgotten paper read before the British Psycho'Analytic So^

ciety in 1939 and published the following year, John Bowlby, then thirty^

three, outlined his views on the sorts of early childhood experiences that
lead to psychological disorders. He noted that analytic literature had
given only meager attention to this subject and politely suggested that
the reason might be that most analysts, because their time is spent sitting
with adult patients, had little opportunity to investigate what goes on
with children in their early lives. He insisted, nonetheless, that it was
important for psychoanalysts to make a scientific study of childhood
experience and relationships, as important as it is for “the nurseryman to
make a scientific study of soil and atmosphere.”^
No one, perhaps, had done more than Freud to spread the view that
the child is the father to the man. By this time much of the educated

world thought of psychoanalysis not only as the promoter of the idea of

unconscious motivation but also of the notion that a good deal of what
we are is a result of what our parents did or did not do to us when we
were young. Who had not heard that an overly strict or punitive toilet
training, to cite one obvious example, could lead to such things as fastid'
 Enter Boiulhy 27

iousness, compulsiveness, or anxiety about dirt in adult life? Even more

widely known perhaps was the idea of early traumatic experience by
which Freud had once explained the origin of devastating neurotic con-

At Bellevue Harry Bakwin had the habit of telling distressed

ditions.

mothers that “there are no behavior-disturbed children, just behavior-

disturbed parents”^ (hardly a comforting idea for parents whose children
were organically damaged), and this pronouncement, too, was consid-
ered very psychoanalytic.
But, in fact, although psychoanalysis stood firmly for the idea that the
roots of our emotional life are found in infancy and childhood, it had
expended little effort in working out the effects of upbringing on charac-
ter development; and the trauma theory, although never forgotten by the
public, had been largely abandoned by Freud and his followers. Although
informally concerned with the quality of parenting and with the things
parents could do to make it easier for their children during the difficult

early years, psychoanalysts generally did not view such matters as a seri-

ous aspect of their work, and little was written about them in their pro-
fessional journals. What really interested them now was the developing
child’s psychic structures and fantasy life, and instead of theorizing about
why certain family conditions caused certain children to become dis-

turbed, they sought the bigger picture: the internal conflicts that bedev-
iled all children as a result of the universal conditions of infancy and
early childhood.
Nevertheless, a concern for the child’s home life continued to grow
markedly in the early decades of the century. Freud’s trauma theory had
struck a chord, as did the ideas of Adolph Meyer, the great Swiss-born

psychiatrist who immigrated in 1892 to the United States. Meyer helped

promote both the mental hygiene movement and the development of
child psychiatry.^ Both would come to see the child’s early environment
as critical in determining later mental health.
But despite the growing concern about the child’s home life among
child health care workers, no one really knew for certain what aspects ot

his home life mattered. Certain obvious things were focused on when a

child was brought into a guidance center with behavior problems: Did he
come from a broken home? Was the house well kept? Was there enough
to eat? Were either of the parents drinkers? Did they establish a proper

moral environment? Etc. As far as John Bowlby was concerned, however,

such questions were almost entirely irrelevant, often reflecting no more
than the prejudices of the day. Bowlhy argued that in concerning them-
28 What Do Children Need?

selves with such issues, child care workers overlooked critical factors of

psychological importance. Their reports frequently concluded, “The

environment appears satisfactory,” when, from Bowlby’s point of view, it

was not satisfactory at all. “It is surprising what vital facts can be over-
looked in a perfunctory interview,” he wrote, “the mother being in a
TB. sanatorium for six months when the child was two, the grand-
mother dying in tragic circumstances in the child’s home, the fact that a
child was illegitimate and attempts had been made to abort the preg-
nancy Intentionally or not, he said, parents often conceal such
unhappy experiences and an interviewer must probe for them.
What mattered, Bowlby said, was not the physical or religious but the
emotional quality of the home. And not just the emotional quality at the

moment when the child was brought in for treatment, but going back to
birth and even before. He pointed to a recent study of criminals in which
the authors found in one case that delinquency had “no relationship to
early or later unsatisfactory environment,”^ when, in fact, the child was
illegitimate and had been born in a Salvation Army home, facts that

begged for further investigation.

While Bowlby believed that heredity could play a role in emotional

disturbance, he doubted that hereditary difficulties would lead to neuro-
sis unless the environment had somehow exacerbated them. And having
worked in a child guidance clinic for several years, he found it rare that a

child brought in for treatment had had an even average psychological

environment.
Two environmental factors were paramount in early childhood,
Bowlby said. The first was the death of the mother or a prolonged separa-
tion from her. To buttress this point, he offered examples of children who
had had lengthy separations from their mothers when very young and
who subsequently became cunning, unfeeling, thieving, and deceptive
qualities that were similar to what Levy, unbeknownst to Bowlby, had
reported in the United States two years earlier.
The second factor was the mother’s emotional attitude toward the

child, an attitude that becomes apparent in how she handles feeding,

weaning, toilet training, and the other mundane aspects of maternal care.

One group of mothers demonstrates an unconscious hostility toward the

child, Bowlby said, which often shows up in “minor pin-pricks and signs
of dislike.” Such mothers often compensate for their hostility with an
overprotecting attitude
— “being afraid to let the child out of their sight,

fussing over minor illness, worrying lest something terrible should happen
 Enter Boivlby 29

to their darlings. . . The underlying hostility emerges, however, “in

unnecessary deprivations and frustrations, in impatience over naughti'
ness, in odd words of bad temper, in a lack of the sympathy and under-
standing which the usually loving mother intuitively has.” Another
group of mothers is neurotically guilty and cannot withstand a child’s
hostility or criticism. “Such mothers will go to endless lengths to wheedle
affection from their children and to rebuke in a pained way any show of
what they call ingratitude.”^ In either case, the results for the child are
lasting emotional damage.
This briefly summarizes the themes of Bowlby’s first professional paper.

The twin concepts presented there — of maternal separation and nega-

tive maternal attitude — would prove rich quarries for Bowlhy and those
who would eventually follow him. In formulating these ideas, he laid out
a point of view to which he would adhere implacably for the next fifty

years of his life.

When 1 met Bowlhy in January of 1989, he was a soft-featured man of

eighty-two with bushy white eyebrows, thinning white hair, and a proper,
somewhat detached, upper-class bearing. He had what Victoria Hamil-
ton, who worked with him for many years, described as “penetrating but
responsive eyes beneath raised eyebrows which to me expressed both
interest and a slight air of surprise and expectation.”^ He still had an

office at the Tavistock Clinic, where he’d worked since shortly after the

Second World War, and he lived in an old rambling house opposite

Hampstead Heath that he’d acquired around the same time. One of his

four children, Richard Bowlhy, lived next door with his family.
When Bowlhy died nearly two years later, an outpouring of reminis-
cences paid tribute to the affection, loyalty, and respect he’d engendered.
There was some mention, too, of qualities like headstrongness, which
might help explain how a young man so new to his chosen field could

take positions of such strength despite the opposition of top people. He

was often considered aloof and emotionally distant — a quality some
attributed to shyness or awkwardness,^ others to a protective shell that
made it difficult for him to express his feelings. Indeed, he rarely spoke

of his feelings, was “completely inarticulate” when he tried, and seemed

almost without curiosity about himself. One longtime colleague noted

that “he was perfectly able to ‘take turns,’ the essential ingredient of con-
versation”; hut seemed oddly touching that she should have felt it rele-
it

vant to state that.^^ Those who came to work under him at the Tavistock
30 What Do Children Need?

Clinic in later years, having learned in advance of his haughtiness and

stubborn, sometimes pugnacious, adherence to his views, were often sur^
prised by his gentle availability and deep fund of affection.

Intellectually, Bowlhy was efficient, focused, and formidable

— “the
most formidable man I ever met,” his wife Ursula would later say.^^ At
some point in his life, he seems to have become the sort of person who
never wastes a minute, never suffers through down time, never fails to

understand and integrate everything he’s read or studied. Ursula Bowlhy

thought of his mind as a “smoothly functioning Rolls'Koyce.”^"^ But it

was a Rolls Royce with artillery. His aggression showed plainly at times,
as when he barked “Bowlhy!” into the phone when disturbed by an
unwanted call;^^ but he could also manage it masterfully, as when he
fielded questions from unfriendly members of an audience with shrewdly
pointed replies. In old age Bowlhy admitted to having been “a rather
arrogant young man,” to which Ursula Bowlhy adds, “He was also an
arrogant middle-aged man and an arrogant old man (he knew he was
right, in fact).”^^ Yet he was also very direct, admirably, almost touch-
ingly, incapable of being devious, and possessed, according to his wife
and others, of an unshakable integrity. He was also very well-mannered
and had an unusual ability to maintain relations with those who held
opposing views. He was in almost every respect an old-fashioned English
gentleman.
This, then, was the upstart who emerged on the psychoanalytic scene
in the late 1930s. Bowlhy was brilliant, confident, impatient, decidedly
off-putting at times, with a tremendous sense of purpose and not at all

inclined to worship existing theories or their makers. In the coming years

he would get under a lot of people’s skin.

Many of the early child psychiatrists came to the field via pediatrics.

That was not the case with Bowlhy. Born in 1907, the son of a promi-
nent baronet and surgeon to the king,^^ Edward John Mostyn Bowlhy
was the fourth child in a family of three girls and three hoys. “Mine was a
very stable background,” he once announced with typical British final-

ity.’^ may have been trying to convey, “stability”

But whatever Bowlhy
here should not he taken to mean warm, secure, emotionally responsive
or any of the other qualities that Bowlhy believed were so important to a

developing child. His parents were conventional upper-class people of

their day, with a belief in intellectual rigor and a stiff-upper-lip approach
to all things emotional. Although Bowlhy never discussed the matter and
seems to have put it out of his mind, he did not have happy relations
 Enter Bowlby 31

with either of them. His mother was a sharp, hard, selhcentered woman
who never praised the children and seemed oblivious to their emotional
lives; his father, although rarely present, something of an inflated bully.

Both parents set themselves utterly apart from their children, handing
over their care to nannies and a governess. The children ate separately

until each one reached the age of twelve, when, if the child still lived

athome, he or she was permitted to join the parents for dessert. The
nannydom consisted of a head nanny, herself a somewhat cold creature
and the only stable figure in the children’s lives, and an assortment of
undernannies, mainly young girls who did not stay very long. Bowlby was
apparently very attached to one of these young nannies and pained when
she left.^° On the other hand, he and his brother Tony were his mother’s
favorites, taken on many outings from which the others were excluded.
This may have contributed to his uncommon selhconfidence.^^

At eight, Bowlby was sent away to boarding school where he joined

Tony, only thirteen months older, with whom he shared a close and
fiercely competitive relationship. Bowlby, who would never criticize his

been sent away because the family was concerned

parents, later said he’d
that the German zeppelins would drop bombs on London. But since the
other children remained behind, it is more likely that this is simply what
upper-class English families did. In any case, he was unhappy, and he
later told his wife, in a rare moment of candor, that he wouldn’t send a

dog to boarding school at that age. Although he never said as much and
was probably unaware of it, almost everything he wrote in later years
about the needs of young children could be seen as an indictment of the
type of upbringing to which he’d been subjected and to the culture that
had fostered it.

Bowlby studied Dartmouth Royal Naval College and Trinity

at the

College, Cambridge. When he enrolled at Cambridge, he was not espe-

cially interested in taking up his father’s calling but “didn’t know quite

what else to do” and so studied medicine. He read psychology during his

third year, however, was intrigued, and “decided to take it up — whatever

that meant!”^^
In the summer of 1928, Bowlby found himself drawn to the phenome-
non of “progressive education” — a radical alternative to the philosophy

by which he himself had been raised and educated. The British progres-
residen-
sive schools, first started about ten years earlier, were essentially

tial schools for maladjusted children and were considered quite beyond
the fringe by mainstream educators. The most famous was Summerhill,
32 What Do Children Need!

founded and run by A. S. Neill, who argued that a disciplinary regime

was exactly the opposite of what children needed, that it quashed their
natural inquisitiveness and stunted their individuality. Instead, children
at his school were pretty much allowed to do as they pleased, as long as
they didn’t impinge on others; and teachers were given special training
so that they could be gently available rather than figures of fear and
authority. This amalgam of anarchism, utopian socialism, and Freudian^
ism must have struck the proper young Bowlby as quite a good mix, for it

remained a cornerstone of his own views for the rest of his life. Almost
thirty years later, in a lecture on child care, he would say:

An immense amount of friction and anger in small children

and loss of temper on the part of their parents can be avoided

by such simple procedures as presenting a legitimate plaything
before we intervene to remove his mother’s best china, or
coaxing him to bed by tactful humouring instead of demand'
ing prompt obedience, or permitting him to select his own
diet and to eat it in his own way, including, if he likes it, hav'
ing a feeding bottle until he is two years of age or over. The
amount of fuss and irritation which comes from expecting
small children to conform to our own ideas of what, how, and
when they should eat is ridiculous and tragic — the more so
now that we have so many careful studies demonstrating the
efficiency with which babies and young children can regulate
their own diets and the convenience to ourselves when we
adopt these methods.

So attracted was Bowlby to the progressive philosophy that he aban^

doned his medical education and worked as a volunteer at two Neilhlike
institutions for the next year. Bowlby had little to say about the first

school except that it was run hy an “inspired manic-depressive” (and vet-

erinary surgeon) named Theodore Faithfull.^'^ At the second, a small
school in Norfolk, he met John Alford, a troubled war veteran (and later
an art teacher in Toronto) who had himself been through analysis and
who took the young Bowlby under his wing, turning his attention to all

those issues that would become central to the Bowlby canon. Most
important, Alford explained the connections between the disturbed
behavior that Bowlby was observing at the school and the unfortunate
early histories of the children involved. Bowlhy joined the staff without
 Enter Bowlby 33

pay, receiving board and lodging for six months. He apparently con^

nected well with some of the children, one of whom followed him every'

where and was known as his shadow.^^ He would later say that

“everything has stemmed from that six months.”

In the of 1929, at Alford’s urging, Bowlby, then aged twenty^two,
fall

enrolled at University College Hospital Medical School and began ana-

lytic training, which included his own analysis. Four
years later, after

completing his degree, he went on to train in psychiatry, while continu-

ing his training in psychoanalysis.
His analyst was Joan Riviere, a close friend and follower of Melanie
Klein, whose views were causing a sensation in British psychoanalysis
at

the time. Bowlby and Riviere were apparently not a good match. She
no
doubt found him a tough nut to crack, and she complained about
Bowlby ’s questioning attitude toward analytic theory
critical, as if,

she said prophetically, he was “trying to think everything out from

scratch.”^^ She was also known to be something of a bully^® which
could

not have sat well with this patient. Their sessions must at times have

seemed like polite wrestling matches.

Riviere no doubt saw Bowbly s persistent intellectual protests as resis-
tance to the treatment, which they may well have been. Indeed, al-
though Bowlby was over seven years in analysis with Riviere, seeing her

almost daily, she was never satisfied with his progress; while he never
It was
gave any indication that she had the slightest impact on his life.^^
only with Riviere’s reluctant approval —
probably arrived at after consid-

erable pressure from her determined young patient that he qualified for

associate membership in the British Psycho- Analytic Society in 1937.

When his new wife told him, in partial jest, that she couldn’t see how he
could “afford both a wife and to continue an analysis which had
already

seven years (and used up most of his capital), Bowlby apparently

lasted

took this as just cause for putting an end to the treatment. (Charac-
teristically, Bowlby spoke little of Riviere afterward.
“The only thing he
told [mel about her,” Ursula Bowlby later said, “was
that she was a lady,

i.e. out of the top drawer like him.”^0

1936 Bowlby had gone to work half-time at the
Meanwhile, in

London Child Guidance Clinic at Canonbury. The child guidance

movement had been more or less exported to England through financial
grants by the Commonwealth Fund, which supported the movement in

British psychiatrists to
the United States. Bowlby was one of the first

in child guidance, and he found that it provided him

become involved
34 What Do Children Need?

with a singularly compatible home.^^ His three years at Canonbury rep'

resented a return to all the things Alford had taught him regarding the
impact of early parent'child relationship. His social work colleagues,
Molly Lowden and Nance Fairbairn, who had had some analytic training
themselves, were taking psychoanalysis in a practical direction that was
uniquely suited to a family mental health center. They would ultimately
have a greater impact on Bowlby’s thinking than any of his teachers or

supervisors in analytic and psychiatric training.^^

Lowden and Fairbairn introduced Bowlby to the idea that unresolved

conflicts from the parents’ own childhoods were responsible for the hos'
tile and deficient ways in which they sometimes treated their children.
As a result, the social workers gave therapeutic attention to the mothers
as well as the children, a process that struck Bowlby as immensely sensi'
hie. Later he would recall two examples from that period.

In one a father was deeply concerned about his 8'yeat'old

son’s masturbation and in reply to my inquiries explained how,
whenever he caught him with his hand on his genitals, he put
him under a cold tap. This led me to ask father whether he
himself had ever had any worry about masturbation, and he
launched into a long and pathetic tale of how he had battled
with the problem all his life. In another case a mother’s puni'
tive treatment of her 3'year'old’s jealousy of the new baby was
as quickly traced to the problem she had always had with her
own jealousy of a younger brother.^"^

But, according to Bowlby, this approach was not mainstream, neither in

child guidance, child psychiatry, nor psychoanalysis, where, indeed, it

was looked down upon by his analytic superiors and caused him troubling
professional conflicts.

Psychoanalysis had certainly played an important part in sensitizing the

public to the dangers of early wounds. In the United States analysts and
analytically oriented workers were frequently among those who insisted
that a child’s behavior is a reflection of his home life. A. S. Neill (a
friend and supporter of analyst Wilhelm Reich) and John Alford were
both solidly in the psychoanalytic camp, and almost all of those who did
the pioneering work on maternal deprivation were analysts, as Bowlby
 Enter Boivlby 35

himself would soon discover. But, for the most part, analysts tended to
impact of problems around feeding, toilet train-
limit their focus to the

ing,and exposure of the infant or young child to sexual intercourse be-

tween its parents. They were not interested in making a serious science
of the way parents treated a child or of the quality of relationships in the

family.
trauma.
Freud had originally argued that neurosis was caused by early
His female patients who suffered from hysteria which included such
symptoms as dizziness, delirium, fainting spells, paralysis
of some part of

the body— had apparently all recalled having been sexually molested
they were small children, often by their fathers, and Freud
de-
when
termined that that was the cause of their condition. But in a
famous

about-face, which has in recent years become the source of immense con-
troversy, Freud announced, in 1897, that he had gotten it wrong the first

time. He said that the unconscious is unable to distinguish between real

memories and fantasies, and, finding it impossible to believe that so many

of his patients had been seduced by their upstanding bourgeois
fathers—

and apparently distressed by the thought that his own father might
be

among the offenders — he concluded that the memory of seduction was

actually the memory of a wish that had been played out in his patients

imagination. Young children, he argued, have a potent erotic drive that

naturally causes them have sexual love with their opposite-sex

to want to

parent and to do away with the same sex parent. Here was
born Freud’s
with the guilty
theory of infantile sexuality and of the Oedipus complex,
feelingsand neurotic tensions that are often left in its wake. Although
and real
Freud always acknowledged the possibility of real seduction
again, and he
trauma, he never seriously considered the parenting factor
seemed to have little sense of the intricate connections that could exist

between the parent’s emotional problems and the child s.

It is now impossible, of course, to
know whether Freud’s hysterical
fathers or anyone else.
patients were indeed seduced or molested by their
But even if they were and Freud made a grave error (as
Bowlby and oth-

ers came to believe), the alternative view he put forth did not inherently
contradict the first one and could easily have lived alongside it: Some
abused or suf-
people become disturbed because they have been sexually
to assim-
feredother traumatic blows that their young minds were unable
ilate at the time; but most others who develop neurotic conflicts have
not experienced such overt traumas. Many considered the new view to

be a victory for common sense. It was, as Charles Rycroft, hardly a

36 What Do Children Need?

Freudian apologist, says, “the beginning of a new era, one in which it

became possible to elucidate the way in which fantasies can distort mem-
ory and in which infantile sexual wishes and parental attitudes combine
to generate what we now call the Oedipus complex. But although
“parental attitudes” may have been an implicit part of the new equation,
with the abandonment of the trauma theory, orthodox analysts became
disenchanted with almost all environmental issues.

The Oedipus complex, nevertheless, proved to be a gold mine for

Freud, because rather than dealing exclusively with the traumatized few
it spoke to the human condition and the conflicts inherent in emotional
life. In the near-universal triangle of mother, father, and child, love,
hatred, and jealousies arise that generate considerable inner conflict, the
only difference between the mentally healthy person and the neurotic
being one of degree, neurotics exhibiting “on a magnified scale feelings
of love and hatred to their parents which occur less obviously and less

intensely in the minds of most children. The new view was not only
more universally applicable, it was more revolutionary and, in a sense,
more humane, for it narrowed the distance between neurotic and ordi-
nary experience, between “us” (doctors, normals, upstanding citizens)
and “them” (women, weaklings, defectives).
For children, the oedipal period (about three to five years of age in the
standard view) is often a critical point of passage, and for many adults
unresolved oedipal feelings are disturbing and frequently distorting of
their lives. As a boy grows into a man, his relationships with women and
with other men and his attitude toward himself as a man, including
whether he is anxious about surpassing his father, are inevitably affected
by how he worked through the competitive feelings that arose in the
oedipal triangle.
One of the paradoxes of the debate over whether neurosis was caused
by the child’s own fantasies or actual molestation was the unspoken
agreement by both camps that the oedipal theory somehow absolves par-
ents. The assumption is that if a little girl, naturally in love with her

father because of her own erotic drives, is haunted in later life hy irra-

tional guilt make unnecessary reparations,

and the need to the fault for
such an unhappy development lies in her. The reason why she, but not
another girl with the same natural 'drives, ends up with neurotic symp-
toms in later life is that she must have had a constitutional disposition
(perhaps her erotic drive was too strong) that made it impossible for her
to resolve her Oedipus complex and move on up the developmental lad-
 Enter Boiulby 37

der. But, in fact, such a conclusion is usually unwarranted and today, at

least, few analysts would hold to it.

True, neurotic conflicts can arise in the most caring environment.

Although parental behavior almost always contributes to them in some
way, it need not be behavior that we would consider seductive,
manipU'

lative, or rejecting. Achieving a completely untroubled adulthood is a

rare, if not impossible, accomplishment in any
environment. But it is
to be a trou'
also true that if a child has oedipal problems and grows up

hied adult with irrational guilt, disturbing fantasies, and neurotic symp'

toms, something was probably amiss in the parenting. Clearly,

some
parents handle oedipal issues in a way that helps the child develop
his

own strength and personhood while maintaining a strong connection to

each of them. Others compete for the child’s affections or use him as a
pawn in their struggles. A
mother may be dependent on her son’s affeC'
tions and subtly seduce him emotionally, so that he remains
caught in

her web rather than free to be his own person and to seek new
rela^

tionships.She may allow her son to observe her humiliating his father,
thereby not only damaging his sense of maleness but leaving him
with

guilty feelings over vanquishing his father. A father may be so dictatorial

with his son as to force him into an unhealthy alliance with his mother.

And so on, ad infinitum, with parallel problems for girls.

The fact that
the child’s fantasy life may be filled with all sorts of
distortions of fact —
his father in
that his father hates him, for instance, or that he has injured
some way — in no way alters the fact that parental behavior has left him
in a stew.
an oedipal and away from a trauma view of the etiology
Freud’s shift to

of neurosis does not, therefore, have to be seen as

blaming children or

letting parents off the hook. Unfortunately,

psychoanalysis — with much
of psychiatry in its tow — became so taken with the problems of the

child’s fantasy life that reahlife events were considered fundamentally

important. Analysts became fascinated by how our
unconscious
less
into healthy
sexual and aggressive drives get bent, twisted, sublimated
channels; how they get hidden by reaction formations (compulsive help'
fulness masking hostility) or allowed their pleasures through compromise
making do where physical aggression is de-
activities (verbal aggression

sired). Relationships and life experiences were inevitably assumed to play

a part in this process, but that was an afterthought. The main focus was
on the individual and the workings of his unconscious.
as almost outside
In those days, Bowlby later wrote, ’’it was regarded
38 What Do Children Need?

the proper interest of an analyst to give systematic attention to a person’s

real experiences.” The standard view was that “anyone who places em^
phasis on what a child’s real experiences may have been . . . was regarded
as pitifully naive. Almost by definition it was assumed that anyone inter^

ested in the external world could not be interested in the internal world,
indeed was almost certainly running away from it.”^®

It was an odd situation, since, in fact, many analysts were well aware
that early relationships had an impact, often a deleterious one, and many
were sensitive to this issue in their individual practices. Their published
case studies attested to this. But important theory making was reserved
for the unfolding of the internal world in what analyst Heinz Hartmann
would call “the average expectable environment,” and to leave aside
issues of variation in upbringing. In the writings of leading classical ana^

lysts, the nature of a patient’s relationships, past or present, often seemed

like an incidental matter.
This gap in mainstream analytic thinking brought to the fore new
schools of thought. Some Freudian loyalists, like Erik Erikson, attempted
to adjust Freud’s developmental stages by making them more attuned to
social issues. Thus, in Erikson’s hands, for instance, the oral stage (when
the mouth is the center of the child’s biological drives) retains all the
Freudian contours but also becomes the time when one does or does not
learn basic trust, according to the type of parenting one receives. Other
thinkers insisted on more substantial revisions. Object relations theorists
(in psychoanalysis the unfortunate word “object” usually means “per^
son”) like Melanie Klein, Ronald Fairbairn, Michael Balint, and Donald
Winnicott in England, interpersonal and social theorists like Karen
Homey, Harry Stack Sullivan, and Erich Eromm in the U.S., and later
on family systems theorists, who were mainly nonpsychoanalytic, were
all struggling over the relational ground left uncharted by the classical
Freudian model. They argued that people are motivated by more than
the desire to satisfy instinctual impulses, like hunger and sex; that they
also have a primary need to be meaningfully connected to others.
Bowlby’s years of analytic training coincided with the early deveh
opment of object relations thinking among psychoanalysts in Britain,
and it was to the development of this theory that he hoped to make a
contribution. The object relations movement, begun by Melanie Klein,
reflected not only a concern with neglected relationships, but a desire
to move beyond neurotic symptoms, like irrational guilt, which was
 Enter Bowlby

believed to derive from oedipal conflicts, to the more fundamental prob-

lem of character.
This was also a time when analysts first began to see a new type

of patient, less sure of himself and what he should be than the well-to-
do bourgeois whom Freud usually treated. The new patients often felt

empty, didn’t know who they were or how they should live, projected

what Winnicott called “false selves,” and were assumed to be suffering

from personality distortions that began in infancy, before the oedipal
triangle became an issue. Their arrival in analytic offices demanded
a

greater understanding of early relationships and their impact on

person-

ality.

Ronald was probably the most compatible with

Fairbairn’s theory

Bowlby’s thinking and must have influenced him. In a bold move,

Fairbairn was in the process of abandoning Freud’s drive theory,
which
being motivated mainly by unconscious forces, like sexual-
saw people as

ity and aggression, which build up inside us and emerge in various ways,

many of them quite disguised. In Freud’s theory, the id was the repository

of these powerful amoral forces; it only knew desire, pleasure, and the
urge for immediate gratification. The ego struggled to tame these forces
and find a way to live in the real world, where gratification often has to

be delayed and impulses controlled, and to accomplish this without

vio-

argue that
lating the strictures of the superego. Fairbairn was the first to
was the need for other people.
what Freud had underestimated in all this

He argued that the libido, or sexual energy, was not pleasure-seeking, as

the classical theory held, but person-seeking and that
psychopathology

originated in disturbances in early relationships.

But Fairbairn, up in Edinburgh, was not a major player in the Psycho-
Analytic Society of London. His theoretical revisions were considered
impudent and his analytic training substandard. With little more fore-

thought than the toss of a coin, Bowlby threw his lot in with Klein.
 3

Bowlbv and Klein:

Fantasy vs. Reality

Melanie Klein was a brilliant Vienna^born analyst and originator of psy-

choanalytic play therapy who many believe to be the most important
and original contributor to psychoanalysis after Freud. Klein perceived
that children in their play communicate a great deal about their inner
lives, their relationships, their anxieties — that, in effect, they free-asso-

ciate through play — and that the analyst, by engaging in and reflecting
aloud on their play, could have a profound therapeutic impact. This huge
and original contribution was enough to earn her a place in the analytic
pantheon, but it was only the beginning of a much larger body of insights
that strongly influenced and divided the analytic world.
Freud had observed that people were troubled by conflicting wishes
and feelings toward others, and even those who are very close to them
are often loved and despised at the same time. He argued that in child-
hood such conflicts and unwelcome feelings are the rule not the excep-
tion, and Bowlby himself would later emphasize this point by speaking of
“the stark nakedness and simplicity of the conflict with which humanity
is oppressed — that of getting angry with and wishing to hurt the very
person who is most loved.’’ ^ The oedipal theory became the focal point
 Boivlhy and Klein 41

of Freud’s insights about such ambivalence and inner conflict. Melanie

Klein took them much further.

Klein’s work with children led her to develop new theories about early

mental processes, how they develop, how they are reflected in later life,

both normal and abnormal, and how they manifest themselves in the

adult analytic situation. It also caused her to see the child s relationship

with its mother in a new light. She believed that the child s early rela^
tionship with its mother lives within the child and that it becomes a
template for future relationships.
Klein began discussing the infant’s inner life in unheard of ways. She
was one of the few Freudians to take Freud’s death instinct seriously and,
in so doing, energized his ideas about the links between love and hate
and the ambivalence that haunt every relationship. Klein did
feelings of

away with the idea of the blissful innocent infant and spoke instead of
infant rage, paranoia, and agony. She believed that the infants first rela^
tionship is with the breast and that it projected onto the breast all its
innate capacities for hatred and love. Indeed, she argued that projection
and the distorting power of fantasy are such pervasive forces that the
“mother” who lives in the child’s fantasy world is considerably different

and more important in many ways than the real mother and that
far

what the real mother does or doesn’t do may be completely uncorrelated

with the child’s later emotional condition.
Because Anna Freud, the other pioneer in child analysis (who a deC'
ade later would also emigrate to London), took a sharply different view
of theory and technique, the British Psycho- Analytic Society held a
sym-

posium 1927 in which adherents of both sides had a chance to express

in

their views. Joan Riviere, Bowlby’s analyst, succinctly summarized

the

Kleinian position on the distorting power of childhood fantasy: “The boy

wishes to be immensely big, powerful, rich, sadistic, as in his imagination
beautiful and
his father appears to be. The girl wishes to be radiantly

adored, possessed of unlimited jewels, finery, children, and so on.

Every

day we see how little these phantasies tally with what the parents
really

were and really did and had.”^

The very young infant is unable to grasp the idea of a person. Even
though modern research reveals that newborn babies know the scent
of

their own mothers and prefer her to others, the baby’s neurologic organi-

zation is apparently not advanced enough to make sense of individual

The world still composed of smaller parcels breasts, faces,

people. is

hands— to which the baby responds with pleasure, anger, fear, and what-
42 What Do Children Need?

ever other emotions it is capable. Only in the second half of the first

year, according to the analytic view, do whole people come into view. It

takes much longer yet for the child to integrate good and bad feelings
about its mother or anything else in its world: Mother is good when she
is satisfying the child, bad when she is not, hut never, at this early stage,

a whole person comprised of good and had traits. Drawing on this theO'

retical knowledge and on the evidence gained from play therapy, Klein
assumed that during early infancy the most fundamental “being” in the

infant’s world is mother’s breast and that the infant has a love^hate rela^
tionship with it, much as he will later have with mother.
There are moments in his relationship with the breast when what we
think of as mother^love is shining brightly. The breast is available, the

milk is flowing and sweet, mother is in a relaxed and cooperative mood,

and the bahy is not overtired, congested, or being pricked by a pin. Now
the breast is good. There are other moments, according to Klein, when
the relationship sours and the baby not only pushes the breast away hut
actively hates the breast and feels unloved and rejected by it.

The moments of bliss and misery that the baby experiences at the
breast — and what hreastTeeding mother hasn’t seen both? — are not
simply matters of physiological satisfaction or pain, according to Klein.
They are relationship moments as well, in which the bahy teels cared for
and loved or deprived and persecuted. Klein argued, however, that, like

many severely disturbed adults, the infant cannot tolerate the idea that
its beloved and alhimportant breast, the giver of sustenance and satisfaC'

tion, is sometimes evil, that there are times when the child wants to
destroy the very breast it loves and totally depends on. That would gen^
erate impossible levels of anxiety. And so, in the baby’s immature mind,
a necessary split takes place. It imagines two breasts, not left and right,

hut good and had, one to love and one to hate. The good breast of the
infant’s fantasy is alMoving and giving; and this positive image reflects

the infant’s own capacity for love. In its relationship with the good
breast, life itself seems good. The had breast, on the other hand, is hostile
and persecutory, reflecting the infant’s innate capacity for envy, hatred,

and aggression.
That a tiny infant, unable yet to grasp the idea of a whole person,
4

might have a love affair with the breast seems inherently plausible. But
hating the breast and seeing it as a persecutor seems odd. To get a better
sense of Klein’s vision it helps to imagine what is going on in the child’s
mind when its needs are not being met, when it feels irritable, when it is

screaming itself blue and cannot he comforted. Would this not he a time
 .

Boivlby and Klein 43

to hate whatever it was that one held responsible for the conditions of

one’s life? To feel deprived and persecuted hy it? Klein believed that “the

young infant, without being able to grasp it intellectually, feels every diS'

comfort as though it were inflicted by hostile forces.”^ And in the early

months, before whole persons exist for the child, the breast is felt to be

omnipotent and the cause for all that’s good and bad in the baby’s world.

In short order, mommy herself will be seen to have this power and he
similarly divided in the infant’s mind.

Seeing small children acting out fantasies of mutilating, poisoning,

and setting fire to their parents in play therapy gave Klein a new vision
of the frustrated infant’s violent inner life when the primitive capacities

for love and hate rule in a pure and unevolved form, when good and evil

are entirely separate, not yet softened by the ambiguities of real life:

Although psychology and pedagogy [she wrote] have always

maintained the belief that a child is a happy being without
any conflicts, and have assumed that the sufferings of adults

are the results of the burdens and hardships of reality, it must

be asserted that just the opposite is true. What we learn about

the child and the adult through psychoanalysis shows that all

the sufferings of later life are for the most part repetitions of
these earlier ones, and that every child in the first years of life

goes through an immeasurable degree of suffering."^

Much of the child’s psychic suffering — arising from its own innate,

orahsadistic impulses projected onto the environment consists of tears

“of being devoured, or cut up, or torn to pieces, or its terror of being sur-

rounded and pursued by menacing figures.’’ Of these terrors, Klein wrote:

“1 have no doubt from my own analytic observations that the identities

behind these imaginary, terrifying figures are the child’s own parents, and
that those dreadful shapes in some way or other reflect the features of its

father and mother, however distorted and fantastic the resemblance may
be. .

In 1935 Klein enunciated her theory of the “depressive

position,’’

which described psychic pain that the child suffers when it dawns on him
that the hated mother he wants to destroy is also the beloved
mother

who cares for him. Whereas before the child had been frightened ot being
destroyed by hostile forces, it now dreads its own aggression and tears that

it will destroy what it loves. Because he still has ditticulty distinguishing

reality from fantasy, his destructive wishes bring on feelings ot “guilt, irre-
44 What Do Children Need?

trievable loss, and mourning.”^ This depressive position remains a part of

the child and can be activated in later life, especially at times of loss. For
some people it forms a prominent part of their adult psychology.
This, briefly and inadequately, is what Klein was beginning to make of
the child’s internal world based on her analytic work and her own cre-

ative leaps, as well as the inclinations of her own personality. Klein’s

description of infant fantasy life — seething with envy and aggression and
filled with “part objects,” like breasts and penises, and also excrement
were seen by many as profound forays into the infantile mind, as well as a

way to account for some of the mental processes that had been observed
in severely distressed or disturbed adults.

But others saw her theories as incredible, even repulsive, and they
were soon a source of intense controversy. Could little babies have such
fantasies? Could they experience such hatreds and malice? Were they
born with such destructive inclinations? Klein had become the adoptive
parent of Freud’s disturbing and largely ignored death instinct, which
pleased Freud tosome extent but alienated those who were relieved that
Freud had never made too much of it. Her personality was also a matter
of controversy. She was an eccentric and difficult person who tended to
ensnare and enmesh those around her. She encircled herself with disci-
ples and, unable to tolerate opposition, sometimes organized attacks
against those who disagreed with her.^ She was estranged from her daugh-
ter, whom she had analyzed as a child, an act of therapeutic incest that
later attracted much (The daughter, Melitta Schmideberg, her-
criticism.

self a British analyst, became an implacable foe of Klein and an early

friend and supporter of Bowlby. She later abandoned psychoanalysis
altogether.) Some, including Klein’s biographer, Phyllis Grosskurth, ques-
tioned whether her theories about the instinctive hostility, destructive-
ness, and (most strangely) envy of infants might not be her way of
evading her own shortcomings, apparently quite severe, as a mother.® To
this day she is remembered by many for being devious and nasty
— “a
frightfully vain old woman who manipulated people”^ was John Bowlby ’s
final assessment.

Klein’s interests were almost totally opposite to Bowlby’s and could in

a sense be seen as complementary, although neither of them viewed it

that way. Although she acknowledged “that a good relation to its mother
and to the external world helps the baby to overcome its early paranoid
anxieties, she certainly did not see those deathly anxieties as originat-
ing in reaction to the mother’s behavior. Indeed, the mother’s behavior
 Bowlby and Klein 45

was of little consequence to her. She did not really care too much about
the child’s environment or how its parents treated it. Psychic reality was

more important to her than material reality; it was her passion, it was all

she really wanted to explore, and she had a genius for it. Riviere echoed

this Kleinian disposition when she issued, in the same symposium men-
tioned earlier, these uncompromising lines:

Psycho-analysis is . . . not concerned with the real world, nor

with the child’s or the adult’s adaptation to the real world [a

direct slap at Anna Freud and her followers, who were con-
cerned with just this adaptation], nor with sickness or health,
nor virtue or vice. It is concerned simply and solely with the

imaginings of the childish mind, the phantasied pleasures and

the dreaded retributions.^^

By the time Klein came London in 1926 at the age of forty-four,

to

such views had already made her well known there among analysts, who
were excited by her brilliance, creativity, and uncanny ability to find in

infant experience antecedents for so much of what they themselves had

seen in extremely disturbed adults. Having received the unqualified sup-
port of Ernest Jones, who headed the British society, she soon had a
major following. was thus a great honor when Bowlby was informed in
It

the spring of 1938 that Melanie Klein herself would supervise him.
The very first case set the tone. “1 was seeing a small hyperactive boy,
five days a week,” Bowlby recalled. “He was anxious, in and out of the
room, all over the place. His mother used to bring him, and her job was
to sit in the waiting room and take him home again. She was an
extremely anxious, distressed woman, who was wringing her hands, in a

very tense, unhappy state.” In the child guidance clinic, where Bowlby
had now been working for three years, it was customary in such cases to
meet with parents and to deal, as far as possible, with their emotional

problems, as Bowlby did with the father who was distressed about his

son’s masturbation and the mother who was anxious about her first

child’s jealousy of the newborn. “But,” Bowlby said, “1 was forbidden by

Melanie Klein to talk to this poor woman. It was just something that
wasn’t done, mustn’t be done. Well, 1 found this a rather painful situa-

tion really.”^^
Klein took this extreme stance because the child’s relationship with
the mother did not concern her as much as how it had become internal-
46 What Do Children Need?

ized in the hoy. Besides, she did not much believe in the efficacy of work'
ing with neurotic parents, since giving them advice on how to care for

their children would probably only increase their guilt and anxiety and
make their attitude toward their child worse. “1 do not, in the light of my
own experiences, put much faith in the possibility of affecting the child’s

environment,” she wrote. “It is better to rely upon the results achieved in
the child itself.”^

To Bowlby, who felt he had seen good results not so much advising
parents —
which he agreed was useless —
hut in treating them therapeuti'
cally and helping them understand the roots of their own feelings, her

rigidity was infuriating and unforgivable. “1 held the view that reahlife
events — the way parents treat a child — is of key importance in deter'
mining development, and Melanie Klein would have none of it. The
object relations she was talking about were entirely internal relation'
ships” — that is, fantasy. “The notion that internal relationships reflect
external relationships was totally missing from her thinking.
“After three months the news reached me that the mother had been
taken to a mental hospital — she had a state of anxiety or depression or

both, which didn’t surprise me. And when came

1 to report this to
Melanie Klein, her response was, ‘What a nuisance, we shall have to find

another case’” — since there was no one to bring the child to treatment
anymore. “The fact that this poor woman had had a breakdown was of
no clinical interest to her whatever; it might have been the man in the

moon who was bringing this hoy. So this horrified me, to he quite frank.
And from that point onwards my mission in life was to demonstrate that
reahlife experiences have a very important effect on development.
A few months later all of Britain was mobilized and Bowlhy was in the
army, washing his hands of Klein (as shortly before he’d done with
Riviere) and happily avoiding a confrontation for which he did not feel

prepared. He and Klein were poles apart, each taking for granted, and in
a sense, dismissing the very things that the other thought most necessary
to explore. But despite Bowlhy’s distaste for Klein, he had been influ'

enced by her. He resonated with her belief in the infant’s capacity to

form early relationships and he was affected not only by her emphasis on
the powerful engine of childhood fantasy and its ambivalent swings
toward kwe and hate, hut her focus on loss, mourning, and depression as

well. Meanwhile, his determination to create a science of the early envi'

ronment and its effects on the developing child had been given a potent
impetus by her imperious opposition.
 4
Psychopaths in the Making:
Forty-four Juvenile Thieves

When a child cries, “1 hate you!” to his mother, the outburst is likely to

induce a storm of inner conflict. Will his venomous hatred destroy the
person that he loves and depends on? Will he himself he destroyed for
his wickedness? The child’s conflict will be different at various stages

of his emotional growth — depending, for instance, on how developed

his moral sense is — but the conflict will be powerful nonetheless and
monopolize a great deal of psychic energy. What is he to do with his teeh

ings of hatred? What is he to do with his murderous wishes? How can he

hate and love and desperately want to be cared for by the same person?
To one degree or another, every child has these feelings of ambivalence
toward his parents and the internal struggle that goes with it. By the time
he becomes an adult, the inner conflict he experiences over love and
hate, the anxiety it produces, and his peculiar way of handling it will

have helped define his character.

If he follows a favourable course [Bowlhy said], he will grow up

not only aware of the existence within himself of contradic'
tory impulses but able to direct and control them, and the
48 What Do Children Need?

anxiety and guilt which they engender will be bearable. If his

progress is less favourable, he will be beset by impulses over

which be feels he has inadequate or even no control; as a

result, he will suffer acute anxiety regarding the safety of the

persons he loves and be afraid, too, of the retribution which

he believes will fall on his own head. This way lies danger. . .
.^

The danger Bowlby perceived was the development of what was

known in psychoanalysis as “primitive defenses.” Defenses help wall un^
wanted feelings out of consciousness. A mother who wants to strangle a
screaming infant may not be able to acknowledge to herself the extent of
her murderous wishes. But if she sweetly sings “Rock-a^Bye Baby,” aware
that the lyrics reflect her own desire to see the impossible child crash

down from the treetops, she is managing her anger in a healthy way. She
allows herself to remain at least partly conscious of her feelings, she is

able to handle them in a way that neither harms the child nor divorces
her from her true self, and she perhaps fends off the intensity of her

wrath with humor, which is considered a mature defense.

Like humor, sublimation (hitting a tennis ball instead of the baby) is

considered one of the hallmarks of mature functioning. Other defenses,

like repression (“I’m not aware of ever having wanted to strangle him”)
and reaction formation (“I only have the most loving, caring impulses
toward my child at all times”), are seen as neurotic, hut are common
nevertheless in many welLfunctioning people. More primitive defenses,
like passive^aggressiveness (“Gee, I guess I just didn’t hear him cry”) or
acting out (“I must have pinched him too hard”), while likely to show up
in almost anyone, are typical of immature or adolescent personalities.
The most primitive defenses, like delusional projection (“That baby
wants to harm me”) or the outright denial of reality (“We had a lovely
afternoon”), are often seen in people with psychotic tendencies.^
In the child’s case, if he finds his conflicted feelings toward his mother
or father too terrible to vocalize or even to admit into consciousness, he
may resort to one of the more primitive defense mechanisms to keep
those feelings at hay. In many cases it is not just the hatred that is dis-

avowed in this way but the love as well.

What exactly does the child do .^’Unable to tolerate his hateful wishes
toward his parent, he may displace these feelings onto a sibling or a pet
and torment them with sadistic behavior. He may dissociate from his
feelings in such a way that he is strangely bland when talking about
 Psychopaths in the Making 49

things that would normally be upsetting. He may go through a psychic

convolution which enables him to believe that he is innocent of aggres-
sive wishes and that they originate in the minds of others, who wish to

persecute and destroy him. Or, as we shall see, he may act out his aggres-

sion in symbolic ways, like theft, which may simultaneously enact his

wish for love. But under no condition must he become aware of his true
feelings or the nature of his conflict. As such primitive defensive styles

get worked into his developing character, the child becomes a disturbed

personality with marked inhibitions, troubling symptoms, and a distorted

view of himself and the world.
To Bowlby, of course, the key question in all this was: What condi-

tions in the child’s home life might make a favorable adjustment more or
less likely? And, virtually alone among analysts, he now went about try-

ing to define these conditions. One problem that he had already identi-

fied was the old-fashioned, very British style of parenting, which was
impatient with the child’s emotional demands, which held that the
greatest sin was to spoil children by showing too much concern for their

outbursts, protests, or plaints, which was insensitive to the harm done by

separating the child from its primary caregiver, and which held that strict

discipline was the surest route to maturity. All this he vehemently

opposed, advocating in its stead a warmer, more tolerant household,

whose hallmark is the ability of the parents to accept the expression of

negative emotions:

Nothing helps a child more than being able to express hostile

and jealous feelings candidly, directly, and spontaneously [he
said], and there is no parental task more valuable, 1 believe,

than being able to accept with equanimity such expressions of

filial piety as, ‘1 hate you, mummy,’ or ‘Daddy you’re a beast.’
By putting up with these outbursts we show our children that
we are not afraid of hatred and that we are confident it can be
controlled; moreover, we provide for the child the tolerant

atmosphere in which self-control can grow.^

What many parents do, however, is just the opposite. Believing that

feelings like hatred and jealousy are dangerous and must be stamped out,

they either punish any expression of it or demonstrate their disapproval

by shaming the child and exploiting his guilt. How can he be such an
ingrate? How can he cause so much pain to his devoted parents? The
50 What Do Children Need?

effect is to drive the unacceptable feelings underground, creating a guilty

and fearful child. What’s more, both methods, Bowlby argued, “tend to
create difficult personalities, the — punishment — promoting
first rebels

and, if very severe, delinquents; the second — shame — and anxi' guilty

ety^ridden neurotics.’’"^
As Bowlby’s views about the impact of parental behavior on children
evolved, he was aware that they were considered highly speculative by
other analysts, especially his Kleinian colleagues, who steadfastly held

that the child’s fantasies correlated poorly with reality. He, therefore,
sought some means to prove his point. He had hardly begun working at
the Child Guidance Center when an opportunity arose.
Years earlier, while still at the Norfolk school with John Alford,
Bowlby had encountered an intelligent boy who was dour and remote
and who’d been expelled from one of the better schools for stealing. The

assumption of Bowlby’s colleagues had been, of course, that this boy’s

thievery was a product of his early experience, which included a rigid

nurse and no real mother figure. Now, at the Child Guidance Center,
Bowlby began to encounter other young thieves, with similar personalia

ties, and this seemed fortuitous: Could he find something fundamentally

similar in their backgrounds?

“1 remember one particular small boy who was in constant trouble and
seemed impervious to praise or blame. He was very delinquent, and he’d
had a disastrous early life. He’d been for nine months in a hospital with'
out any visiting between the age of eighteen months and two years three
months, and he had never made any emotional relationship with his par-
ents since. And then another child turned up — a little girl — with a very

similar condition and a very similar story.” Bowlby believed he saw a

cause-and-effect relationship between the early deprivations and the

criminal character, an idea, that, he said, “was regarded as mad at the

time.”^ While at the clinic, from 1936 to 1939, he collected forty-four

such cases aged six to sixteen. Some of the thieving children were
depressed, some hyperactive, and some displayed the same detached
quality that Levy and Bender were discovering in New York. Bowlby
called these detached children “affectionless.”
When social workers at the clinic examined the parents of the forty-

four young thieves, they found attitudes and behaviors that were more
drastically disturbed than the misguided notions of many otherwise
normal British families. Claud’s mother was “immoral, violent, and nag-
 Psychopaths in the Making 51

ging.” Clifford’s mother “extremely anxious, fussing, critical, and hy-

pochondriacal.” Lily’s mother was “drunken and cruel” and “did not
want the child.” Fred’s mother “shouts and terrifies the children.” Kath-

leen’smother was “extremely unstable and jealous,” “had curious sexual

ideas about the children and had been seen thrashing dogs in a sadistic

way.” Leo’s mother “alternates between violent domination and senti'

mentality.”
Several of the children had been blamed for a sibling’s death. Cyril’s
mother “openly stated that she wished he had died instead of the baby.”
Fathers’ attitudes were also problematic in a number of cases. Winnie’s

father “often beat her” and openly favored her younger sister. Reginald’s

father was “a morose man, who hated the child, shouted him and gave
at

him neither affection nor presents.” Other fathers took no interest in

their kids, and some made it plain they hadn’t wanted them. Still, the

mothers were obviously the primary problem, since they spent the most
time with the children. All told seventeen of the twenty^seven children
in this group had disturbed mothers.^
In trying to understand these children, Bowlby did not make a simple

equation between bad mothering and bad behavior. Rather, he wanted to

show, in modified Kleinian fashion, how mistreatment was filtered

through the child’s inevitable fantasies and distortions to produce the

unhappy behavioral outcome. He argued that when a mother is irritable,

nagging, and critical, when she unnecessarily interferes and frustrates the
child, he will become not only angry and aggressive but greedy both for

affection and for the things that represent affection to him, like sweets.

In such aggression and greed lie the roots of theft. The child’s hostility

and avarice will at first be directed at the frustrating mother, which will
make her all the more irritable, nagging, and critical. And thus a vicious

cycle is established that will be incorporated into the child’s view of the
world and later relationships.

Having disturbed and emotionally abusive parents was not, however,

unique to the young thieves. Indeed, there was very little difference in

this respect between the thieves and the fortyTour nonthieving diS'
turbed children who also attended the clinic and constituted the control
group for the study. Bowlby was thus unable to say anything conclusive
about the etiology of thievery based on parental behavior. But he was reh
atively certain that if he were to look at a third set of forty-four normal
children, he would not find these gross parental failings. In these prob'
52 What Do Children Need?

lem parents he felt he had located a clear, if general, source of child emo-
tional disturbance. But he didn’t have a third set of children and so even
this aspect of his analysis remained speculative.
Bowlhy sensed that such a study was beyond his reach in any case.

While it is easy enough to get anecdotal evidence of parental neglect and

abuse, making a comparison that has any claim to scientific validity is

difficult. How do you rate neglect, mistreatment, unkindness, unrespon-

siveness, abuse, not to mention subtler or more manipulative forms of

bad parenting? How can you be sure that the people who do the ratings
are looking at the same things in the same way? What’s more, in inter-

views, some parents are more forthcoming, some more concealed; and
observing parents in the home over a long period of time is a costly

proposition, which can also be skewed by deception and efforts to stay on

good behavior. In this respect Bowlby’s entire goal of creating a develop-
mental psychopathology based on a scientific evaluation of the environ-

ment seemed doomed from the start.

But Bowlhy found one environmental factor that was easy to docu-
ment and not open to misinterpretation — namely, prolonged early sepa-

rations of child and mother, and this was where he turned his attention.

Six-year-old Derek B. was the first young thief Bowlhy met at the clinic.

He had been referred for persistent thieving, truancy, and staying out
late. He appeared to come from a normal, happy family, with sensible,

affectionate parents and an older brother who seemed perfectly well

adjusted. Derek’s infancy was unremarkable. But at eighteen months he
contracted diphtheria and was sent to the hospital. He remained hospi-
talized for nine months without ever seeing his parents, which was the
standard procedure at the time. The hospital staff adored him, but when
he came home, he seemed a stranger. He called his mother “nurse” and
he showed no affection. “It seemed like looking after someone else’s

bahy,”^ his mother said. Derek so stubbornly refused to eat that he was
finally allowed to starve for a time. After a year and a half, he seemed to
settle down, but he remained strangely detached, unmoved by either
affection or punishment. His mother described him as hard-boiled.

This was not the first impression he gave to Bowlhy, who found Derek
to he sociable and engageable, “a genial, attractive little rogue.”® But his
play was often violent and destructive, and he frequently stole the toys.
In fact, far from being genial, Derek seemed to care for no one, except
perhaps his brother. He preferred to play alone, fought frequently with
 Psychopaths in the Making 53

other children, destroyed his and their toys, lied to his teachers, and stole
from everyone. He spent the money he stole on sweets, which he shared
with his brother and other children. He was repeatedly beaten, by both

his parents and school authorities, but such punishment seemed to have
no impact, except to make him cry for a few moments.
Bowlby diagnosed Derek as an “affectionless character,” but he might
just as easily have used Loretta Bender’s diagnosis of psychopathic per^
sonality disorder. Fourteen of his forty^four thieves fell into this affec'

tionless category. Of those fourteen, fully twelve had, like Derek, suffered

prolonged, early separations. This struck Bowlby as a stunning finding.

Only five of the remaining thirty young thieves, who were not affection-

less,had experienced such separations, and only two of the forty-four

controls. This discovery —
this association between prolonged early sepa-

ration and affectionless character — would determine the future course of

Bowlby ’s career, placing him in the forefront of the small group of mental
health workers who were attempting to warn their peers about the dan-
gers of maternal deprivation.
Derek’s history was no more poignant than the other affectionless

thieves, the main difference being that he appeared to have come from a

less troubled home, most of the others having been illegitimate, un-
wanted, and generally suffering from the same neglect and mistreatment
typical of the families in the study. Betty was seven months old when her

pregnant mother discovered that her husband was married to another

woman. She threw him out and soon married someone else, but mean-

while she put Betty in foster care. Betty went through a succession of fos-
ter homes and a year in a convent school before returning home at age

five to her mother, step-father, and two new babies. Raymond’s mother
died of diphtheria when he was fifteen months old. He lived with his

aunt for nine months, then a teenage sister whose care of him was
described as “extremely casual”^ and who would often go out and lock

him up alone in the house without food. Norman’s mother had tubercu-
losis and spent much of the time when he was two to five in the hospital.

He was looked after by a variety of people during that period. And so on.

Except for their weird detachment, the surface personalities ot the

affectionless thieves were quite varied. Some gave the impression that
they might easily develop into desperate and dangerous criminals; some
were apathetic loners, some energetic and active, and some of the latter
aggressive and bullying. Only one was overtly cruel. Kenneth, “although
full of hard luck stories, was alert and resourceful.”^^ A few seemed
54 What Do Children Need?

unnaturally eager to please. Charles washed himself frequently and was

very concerned about his appearance.
Some of the affectionless children, if separated from their mothers

early enough, when they were still only a few weeks old, did very well
with a mother substitute thereafter. These children were traumatized
only later when they were ripped out of their second home, sometimes in
order to be returned to their first. In contrasting the various impacts of

different separations, Bowlhy noted that for the separation to be a cause

of affectionless character, the child must be old enough — usually six

months — to have established a firm emotional tie with his mother figure
before being sent away.
There was much variety among the individual cases the separations —
occurring at different ages, for different reasons, from families with differ^
ent emotional climates, and ending in different ways. But each of them
had what Bowlhy called a “remarkable lack of warmth or feeling for any^
one.”^^ They were markedly different from almost all the other children
in this respect. They tended to be solitary and unresponsive, impervious
to punishment, indifferent to kindness. Foster parents typically com'
plained that they could not reach these children
— “We never seem to

get near her” was a typical lament. The children wandered away from
home, as if the very concept of home did not matter much to them.
(Kenneth stayed away for a week once, sleeping in empty houses, after a

mild altercation.) They were frequently truant from school. Many lied

brazenly and had no real friendships. In a short interview they often

made a good impression. But Bowlhy quickly learned that their amiabih
ity was superficial and signified no authentic openness. He believed they

formed the hard core of repeat offenders in the criminal population. (He
need only have looked as far as John Dillinger, America’s most famous
desperado of that era, to have found an example of a hard'boiled career
criminal whose childhood was very similar to that ot the youngsters he
observed.) But beneath the antisocial attitudes Bowlhy saw a profound
and unreachable depression, as if, when they lost their loving universe, a
switch turned off in them that could not he turned back on. “Behind the
mask of indifference,” he wrote, “is bottomless misery and behind the
apparent callousness despair.”*^
The attitudes of these haunted children toward material things gave
uncanny voice to their inner torments, and Bowlhy now turned his

attention to the symbolic meaning of their behavior. In doing so, he

would attempt to show that although the children’s fantasies were full ot
 Psychopaths in the Making 55

distortions, the distortions were nevertheless stimulated by real experi'

ences that were beyond their capacity to work through.

Although all stole repeatedly and compulsively, the thefts were often
pointless, and two of the boys would destroy any presents given to them.
Albert stole a bicycle despite having one of his own. Charles stole “a tin
of baking powder, his father’s razors, army discharge badge, and a bul-

let. Roy stole toys from shops and rabbits, which he killed. Nansi stole
money from her teacher’s bag, from the landlady’s pocket, from a Sal-
vation Army box, which she pried open with a knife. Several of them
robbed the psychiatrist during play, making off with one or more of the
toys, even if them seemed to
they had already been given one. None of

have any sense of the meaning of what they had done. They could not
say why they stole, and they seemed impervious to the wrong they had
done to others.
A child who’s been separated from his mother, Bowlby argued, not
only craves her love but also the symbols of her love. And so, typically,

the young thieves stole milk, food, or money to buy food. These are the
things they naturally associate with gratification. Often they stole from
their mothers, like Norman K., who broke into his mother’s money box
— a tragically perfect symbolic act, suggesting envy of what the mother
has and is refusing to bestow on him, that Klein would have loved. Such
wild and pathetic cravings and their ability to endure into adult life was
epitomized for Bowlby by an adult patient he had at the time who took
her morning tea from a baby bottle. He predicted that such longings,

incorporated into the personalities of the affectionless children, would

eventually be acted out sexually as well, and that many of the affection-

less girls would become prostitutes, a supposition he supported with

figures drawn from a League of Nations survey showing that a high pro-
portion of prostitutes suffered severe early separations.
All children are subject to cravings and hostilities directed at their
parents. But usually such impulses are checked as the young child devel-
ops a superego, a watchful inner monitor of right and wrong that some
analysts believe first comes into being as the child realizes that the hated
mother is the same person as the loved mother and wants to protect her
from his rage and hostility. Bowlby suggested that the affectionless

thieves lacked superego controls because their loving feelings had either
never had the opportunity to develop or had been swamped by rage.
Some children were removed from their mother at a time when their
love for her was only beginning to coalesce (typically in the second halt
56 What Do Children Need?

of the first year). Others, having established loving ties, were ripped out
of their homes and somehow “expected to settle down cheerfully with

strangers.” Such children not only hated their foster mothers hut hated
their real mothers as well, and would typically have nothing to do with

them if they visited and icily avoided them for long periods when and if
they returned home.
To hate one’s mother, Bowlby said (echoing Klein), is to conjure up a
picture of her that is hateful and malicious. “Phantasy, horn of rage,” he
said, “distorts the picture of the real mother. A kindly mother who has to
put her child in hospital, a frustrating yet well-meaning mother and a
really unkind mother can, by this process, alike come to be regarded as

malicious and hostile figures.”

Normally Ihe wrote], when such phantasies arise in children,

they are soon corrected to some degree by contact with the

real mother, who, whatever her shortcomings, is never so bad
as the bad mother which the child pictures to himself when
he is in a rage. . . . But where a child does not see his mother
for many months there is no opportunity for this correction of

phantasy hy reality-testing to operate. Extravagant phantasies

of the kind described then become so entrenched that, when
the child returns to the real situation, he can see it in no
terms but those of his phantasies.*"^

People who grow up hating their parents invariably hate themselves as

well. If we see them as malicious beings, how can we ourselves, who are

so identified with them, he any better? What’s more, to have such hateful
feelings toward someone we feel we should love makes us feel all the

more despicable. Thus, little Leo, who sported a devil-may-care attitude,

also suffered pangs of suicidal depression. “1 know I’m a wicked hoy,” he
told his mother on one occasion. “You had better murder me, then 1 will
^
he out of your life.”*

The adults who tried to work with these children were repeatedly frus-

trated and infuriated hy their hard-hoiledness — a quality that Bowlhy

attributed to their determination never to he hurt again. Sometimes the
indifference to others was only skin-deep. But they all presented them-
selves as utterly uninterested in affection or hostility, thereby disarming
everyone they met of the power to affect them.
By this process Bowlhy believed the affectionless thieves had become
 Psychopaths in the Making 57

locked in a painful isolation that would imprison them forever, their

hunger for love and their rage at its absence only showing itself in erup'

tions of meaningless sex, theft, and aggression. Were they utterly beyond
reach? Not necessarily. But the task of establishing the love and trust

necessary for a healing connection had become woefully complicated

and perhaps beyond the capacity of the average parent to cope with.
Bowlby saw the which these children were
tragic internal prison in

trapped as an extreme manifestation of what other children suffer on a

smaller scale. But in these extreme cases and in the separations that
seemed to contribute so profoundly to their cause, he had the wedge he
was seeking to open the debate on the impact of parenting and early
experience on emotional development.

Written mainly in 1940, Bowlby’s paper “Forty^four Juvenile Thieves:

Their Characters and Home-Life,” was finally published in the Inter^

national Journal of Psycho-Analysis in 1944, after he came out of the army.

The paper, Bowlby believed, benefited considerably from the influence
of a group of army psychologists who had helped turn him into a research
scientist during the war. In 1946 it was republished as a monograph,
which gave it a second life and a wider readership.
The paper shows all the trademarks that characterize his later writing.
It is beautifully and simply written. Aside from a few technical terms, vir^

tually any intelligent person could read and understand it and be moved
by it. His prose is clear and direct. He uses metaphors, many of them
drawn from medicine, that forcefully drive home his themes. Thus:

“Theft, like rheumatic fever, is a disease of childhood and adolescence,

and, as in rheumatic fever, attacks in later life are frequently in the nature

of recurrences.”^^ He shows an unusual concern with matters outside the

realm of psychiatry — in particular, with social problems and social poh
icy; and he clearly believes that his work can he of service to more than
psychiatric patients. In dealing with the contributions of others, he does

not stir up fights or waste the reader’s time with lengthy refutations of
established views — no attacks, no bitterness, no pointed statements, no
attempt to aggrandize himself in any way. He just states the tacts as he
sees them and keeps his feelings to himself regarding whatever inadequa-
cies may exist in the theories he’s building upon or implicitly disputing.

When disagreeing, he does so plainly and quickly, without rancor or

innuendo, and then moves on to an elaboration of his own view. His
adaptation of Melanie Klein’s material on child fantasy shows what will
58 What Do Children Need?

be a characteristic Bowlhy trademark — an unhesitant acceptance of

whatever he finds useful, regardless of its source, and a willingness to gen^

eroLisly credit even those whom he dislikes or who have denounced his

views. His voice is upstanding, sensible, and authoritative.

Further research would eventually establish that separations, even
long separations, do not in themselves cause a child to become a psycho^
pathic personality. It’s the depriving separation that’s so calamitous, where
the child never has a chance to develop a true attachment; where there’s
no alternative mother figure to take up where the first mother left off

and perhaps to keep her memory alive; or where there are, early on, a

series of short-term mother figures and thus repeated losses, all of which
cause a bitterness and mistrust to develop and the shutting down in the

child of his natural tendency to reach out for love and connection. In all

this, Bowlby’s thinking would be largely validated; although his failure to

make the distinction between types of separation would cause some
enduring misunderstandings.^^ Later research would also find that where
delinquency is a factor, separation is not the only cause; it is often pre-
ceded by distortions in the parent-child relationship, often with the
father.^® Meanwhile Bowlby’s broad point, about the danger ot early

depriving separations, not only in terms ot the suffering it causes but in

its disturbing impact on character formation, had been made in a power-
ful way. His study would help inspire a revolution in child psychiatry,
especially after it came to the attention of an important official at the
newly founded United Nations.
 5

Call TO Arms:
The World Health Report

In 1948 the Social Commission of the United Nations decided to com-

mission a study of the problems and needs of homeless children, a major
concern in postwar Europe. Ronald Hargreaves, chief of the Mental
Health Section of the World Health Organization in Geneva, was eager
to get the WHO involved by reporting on the psychiatric aspects of the
issue. Hargreaves, who’d met Bowlby in the army during the war, had
been impressed with his monograph on the young thieves. Indeed, he

had already proposed that Bowlby do a report for the WHO

on juvenile
delinquency, which Bowlby rejected. He now approached Bowlby again,
this time suggesting a psychiatric report on homeless children.
Bowlby
promptly agreed. In 1950 he moved to Geneva for six months to investi'
gate and write a report that would have ramifications in psychiatry, adop^
tion procedures, and ordinary home life around the world.
Bowlby spent weeks gathering data on the Continent
six in France, —
Sweden, Switzerland, Holland —
and then five weeks in the United
States, using the opportunity to communicate with social workers
and

child psychiatrists (including David Levy, William Goldtarb, Milton

Senn, and Rene Spitz) and to become acquainted with the literature. He
60 What Do Children Need?

found an unexpected gold mine: Levy writing about adopted children

who were deceitful and eerily detached; Bender reporting on psycho^
pathic'like children who had been in a series of foster care and adoptive
homes; Bakwin, Goldfarb, and Spitz warning about the psychiatric dam-
age done to institutionalized babies; and a number of European clinicians
with similar, if less conclusive, findings. Bowlby quickly assimilated this

and a related mass of material and six months later produced Maternal
Care and Mental Health which finally united and gave a single voice

clear, eloquent, commanding — to the solitary figures who had been urg'

ing modern society to see the suffering of its children.

Bowlby began by emphasizing the similarity of findings. Mainly un^
aware of one another’s work, they unanimously found the same symp'
toms in children who’d been deprived of their mothers — the superficial

relationships, the poverty of feeling for others, the inaccessibility, the

lack of emotional response, the often pointless deceitfulness and theft,

and the inability to concentrate in school. “So similar are the observa^
tions and the conclusions — even the very words — that each might have
written the others’ papers.”^
Beyond the startling consistency was the fact that the researchers had
approached the problem with different methods, which proved comple-
mentary to one another. Bakwin and Spitz directly observed the suffering
of infants separated from their mothers. Bender’s and Bowlby’s studies
had been retrospective, examining children who were already quite dis-

turbed and attempting to discover what experiences they shared in com-

mon. The problem with such studies, however, is that they are in a sense
playing v/ith loaded dice — they examine only disturbed chileiren and say
nothing about other children who may have had identical misfortunes
yet remained undamaged. This shortcoming, however, was made up for

by Goldfarb, who followed the development of a group of normal infants

who had been given up in infancy by their mothers. One group was raised
in an institution until approximately three and a half, when they were
placed in foster care. Babies in the other group were placed in foster
homes as soon as their mothers gave them up. On each test of develop-
ment — general intelligence, visual memory, concept formation, language
ability, school adjustment — the figures revealed the damage done to
institutionalized children and the benefits of foster care. Most dramatic
was the impact on the capacity to form relationships, with thirteen of the
fifteen institutionalized children incapable of forming deep or lasting
connections, a problem that did not exist for the children in foster care.
 Call to Arms 61

Wartime experience in England rounded out the picture. In Hamp'

stead Dorothy Burlingham and Anna Freud ran a residential nursery for
children whose parents’ lives had heen disrupted by the war, some of the
mothers living in shelters because their homes had been bombed out in
the blitz. The two analysts found that only infants in the first months of
life adjusted easily to such an environment. They struggled to make the

separation for the other young children in their care easier by doing it in

slow stages, but while this process worked for children over three or four
years of age, for children between one and a half to two and a half years,

it proved a catastrophe. A typical case was Bobby, a twenty-four^month'

old boy who, despite being looked after by a mother substitute and vis-
ited daily by his mother during the first week, started deteriorating when
her visits tapered off, becoming listless, incontinent, aggressive, and un-

willing to eat. Another little boy, named Reggie, who had been in the

nurseries almost continuously since he was five months old, formed

passionate relationships to two young nurses who took care of

him at different periods. The second attachment was suddenly
broken at two years and eight months when his “own” nurse
married. He was completely lost and desperate after her depar-

ture, and refused to look at her when she visited him a fort-

night later. He turned his head to the other side when she

spoke to him, but stared at the door, which had closed behind
her, after she had left the room. In the evening in bed he sat
up and said: “My very own Mary- Ann! But I don’t like her.”^

Other were equally poignant. Three-year-old Patrick, whose

stories

mother was working in a munitions factory, kept himself from crying

because his mother said she would not visit if he cried. In place of crying,
he began a ritual that consisted of nodding his head whenever anyone
looked at him and repeating that his mother would soon come for him,
that she would dress him and take him home. If the listener accepted his
story he seemed fine, hut if he was contradicted he hurst into tears.

Within became more monotonous and compulsive,

a few days the ritual

Patrick constantly attempting to reassure himself with his story. As he

continued, it became more elaborated: His mother would come for him,
she would put on his coat and leggings, she would zip up the zipper, she
would put on his pixie hat. When someone asked him whether he could
stop repeating this again and again, he tried to be a good boy and cooper-
62 What Do Children Need?

ate. But now, instead of speaking the words, he mouthed them and
simultaneously began enacting the ritual by putting on the imaginary
clothes. Within a few days this was further streamlined and reduced to
something of a tic. “While the other children were mostly busy with
their toys, playing games, making music, etc., Patrick, totally unintet'

ested, would stand somewhere in a corner, moving his hands and lips

with an absolutely tragic expression on his face.’’^

In this case we see the way in which the mother’s psychological prob'
lems — her unwillingness to tolerate Patrick’s crying — complicated what
would in any case have been a difficult grief reaction and pushed his

behavior into a symptom, a chronic and compulsive tic that was dissoci'

ated from his feelings and from the situation in which those feelings orig'
inated.
Other children who seemed to have adapted well to the nursery fell

apart when they left, often becoming hostile toward their mothers, de^
manding and jealous, throwing temper tantrums and developing a
strange emotional detachment, sometimes accompanied by a monotO'
nous rocking of the body and in extreme cases by head banging.
Once again language function was singled out as suffering in the
absence of a mother figure, and Burlingham and Freud found that “when
children are home on visits . . . they sometimes gain in speech in one
or two weeks what they would have taken three months to gain in the

nursery.’’^ Only when the children had a substitute mother in the nursery

did they seem to open up again to learning.^

After reading the reports of the wartime nursery, Bowlby spoke to

Anna Freud about the experience. She said that nothing they did could
prevent the severe deterioration of the very young children placed in
their care. She and Burlingham felt the problems inherent in the situa^
tion were so grave and would take so many workers to correct that they

concluded, “It would be preferred to arrange tor each helper to take a

couple of children home with her and close the nursery.”^
All this proved an opportunity for Bowlby to argue that the mother'
infant relationship is an extremely important one, that it was not a pleaS'
ant amenity for the child but an absolute necessity, and that significant
early separations are perilous to the child and ultimately to society as

well. Many professionals were resistant to accepting these findings. But

Bowlby forcefully argued that they and their policies must change. He
insisted that families, especially poor families, needed greater assistance if
 Call to Arms 63

they were to stay intact. He advocated that large numbers of people be

trained in marriage and child guidance and in work with parents of the
very young. He said the large outlays of funds required would be far less

than the later costs of institutional care and delinquency.

While noting that certain researchers were hopeful that, given extra'

ordinary care, maternally deprived youngsters were capable of being

rehabilitated,Bowlby warned that others had found such children dam'
aged beyond repair.® He also acknowledged that some children do seem
to escape the ravages of early separation. But, in a typical Bowlby flout'
ish, he added, “the same is true of the consumption of tuherculat'infected
milk or exposure to the virus of infantile paralysis. In both these cases a
sufficient proportion of children is so severely damaged that no one
^
would dream of intentionally exposing a child to such hazards.
At this time social workers in many countries typically separated chib

dren from their mothers because of unsatisfactory home conditions

which often meant untidiness or material want or because the mother —

was unwed. To Bowlby this was madness. He argued that those responsi'
ble for child welfare had been so preoccupied with physical health and,

often, physical appearance, that they had at times taken expensive steps

“to convert a physically neglected but psychologically welhprovided

child into a physically welbprovided but emotionally starved one.”^^ He
believed that, unless a mother (or a foster mother) was cruel or abusive,
an untidy, disorganized, and unwed mother was often better equipped to
emotionally care for a child than a tidy and welhrun institution. In one
of the many moving portions of the report, informed no doubt by his
recent experiences as a father, he wrote:

The mothering of a child is not something which can be

arranged by roster; it is human relationship which alters
a live

the characters of both partners. The provision ot a proper diet

calls for more than calories and vitamins: we need to enjoy

our food if it is to do us good. In the same way the provision of

mothering cannot be considered in terms of hours per day but
only in terms of the enjoyment of each other’s company
which mother and child obtain.
Such enjoyment and close identification of feeling is only

possible for either party if the relationship is continuous

Just as the baby needs to feel that he belongs to his mother.

64 What Do Children Need?

the mother needs to feel that she belongs to her child and it is

only when she has the satisfaction of this feeling that it is easy
for her to devote herself to him. . .

It is for these reasons that the mother^love which a young

child needs is so easily provided within the family, and is so
very very difficult to provide outside it. The services which
mothers and fathers habitually render their children are so
taken for granted that their magnitude is forgotten. In no
other relationship do human beings place themselves so unre-
servedly and so continuously at the disposal of others. This
holds true even of bad parents — a fact far too easily forgotten

by their critics, especially critics who have never had the care
of children of their own. It must never be forgotten that even
the bad parent who neglects her child is none the less provid-
ing much for him. Except in the worst cases, she is giving him
food and shelter, comforting him in distress, teaching him
simple skills, and above all is providing him with that conti-
nuity of human care on which his sense of security rests. He
may be ill-fed and ill-sheltered, he may be very dirty and suf-

fering from disease, he may be ill-treated, but, unless his par-

ents have wholly rejected him, he is secure in the knowledge

that there is someone to whom he is of value and who will

strive, even though inadequately, to provide for him until

such time as he can fend for himself.

Bowlhy admonished governments, social agencies, and the public for

their failure to appreciate the central value of maternal care, as impor-

tant for mental health in infancy and childhood “as are vitamins and
proteins for physical health.” (This analogy, originally Levy’s, would he
associated with Bowlhy ’s name for decades.) He warned that mothers
should neither he sentimentalized nor punished. “At one time the puni-
tive attitude took the form of removing the baby from his mother as a

punishment for her sins. Nowadays this punitive attitude seems to lead
in the opposite direction and to insist that she should take the responsi-
bility for caring for what she has so irresponsibly produced, thereby fore-
closing the chance for a happy adoption.”^
Reporting data, synthesizing meanings, raising critical questions,
Bowlhy left barely a stone unturned. In emergencies, he said, children
are ideally left with relatives, and he urged that funds he provided so that
 Call to Arms 65

poor families could accomplish this when a relative lived far away. He
reviewed the care of maladjusted and sick children, the proper way to
run residential nurseries when there was no other recourse but to institU'
tionalize a child, the causes of family failure in Western societies and

how it could he reversed.

He warned that when children are neglected, “as happens in every

country of the Western world today, they grow up to reproduce them^

... are a
selves.” Again he used a medical metaphor: “Deprived children

source of social infection as real and serious as are carriers of diphtheria

and typhoid.” Although it would be decades before research would
prove his point, Bowlby declared that the psychopathic parent who was
often found to be the cause of child neglect is usually the grown-up

maternally deprived affectionless child.

In numerous areas — from adoption policy to the impact of short-term

separations — Bowlby cited the lack of adequate research and suggested

where might proceed. Given his concerns and the directions they
it

would eventually take, one recommendation stands out with special res-
onance to anyone reading his report more than forty years later. The
partial forms of maternal deprivation, due sometimes to ignorance but
more often to unconscious hostility on the part of the mother deriving
from experiences in her own childhood, could well form the subject of

another report.
The impact of the World Health report can barely be overstated. Most
immediate was the effect on public policy, particularly adoption, social

work, and hospital practices worldwide. It heightened interest in Spitz’s

work, helping to spread his film, which also had great impact. It caused
new thinking about the causes and prevention of delinquency and the
training of young women for motherhood. It stimulated an extraordinary
aroused
flood of research, including experimental work with animals.
It

new thinking about the nature of the infant-mother bond. And it made
Bowlby a household name in Britain and a champion of child care work-
ers in many parts of the globe. It also placed him at the center of a huge

and stormy debate that in one form or another would swirl around him

for the remainder of his life. The idea that early experiences have serious

consequences and psychosocial development was, amaz-

for intellectual

ingly, controversial in itself, but the controversy was

only heightened by

the fact that many of Bowlby ’s claims were exaggerated

by others and

quoted out of context.

The WHO report was followed two years later by the publication ot a
66 What Do Children Need!

paperback abridgment. The paperback was the inspiration of Margery

Fry, an educator, a juvenile court magistrate, and a member of the es'

teemed Fry family (Roger Fry, the painter, was her brother). Called Child
Care and the Growth of Love, the book quickly became an international
best'Seller, ultimately reaching a half a million in sales. The book led to
a worldwide change in outlook toward the needs of small children and
widespread improvement in institutional care, its impact compared to
the exposure by Elizabeth Fry (an earlier member of the Fry clan) of the
horrendous conditions in early-nineteenth century prisons. The issue

of mother^love and maternal deprivation had arrived in a big way, and so

had Bowlby.
 6

First Battlefield:
"A Two-Year-Old
Goes to Hospital”

little boy was just 10 iTiontbs old when he had gastroenteritis and
was taken to hospital 1 accompanied him there but on entry he was
me without even time for a goodbye kiss (this being
firmly taken from
partings only
explained as the best course of action, as long drawn out
”
made matters worse)
Thus wrote a British mother as recently as 1964. She continued:

He fretted terribly and missed being played with, and every

time we visited him he was crying, and even before he saw me
coming down the corridor. 1 was allowed to give him his sup^
per in the 1 hr. nightly visit and to change his nappy if neceS'
sary.

I visited him thinking I could be natural with him but

time when
found myself upset too, especially at the leaving
his cot, tears
the bell went. He would stand at the end ot
streaming down his face and wouldn’t even wave
bye-bye. I

knew they were doing all they could hut how 1 wondered all

he had settled down to sleep or play and if he cried

evening if
68 What Do Children Need?

all day. This lasted the whole of the two weeks he was there
and on coming home with him he was very mistrustful of me
leaving him even to go into the next room.^

When children are hospitalized today, parents are generally accorded

unlimited visiting privileges. Fifty years ago they were often not allowed
to visit at all, and later, in the 1950s and much of the 1960s, visiting
hours were typically restricted to one hour a week. The children’s wards
in a hospital were tight little ships in which the intrusion of parents was
unwelcome and, indeed, deemed unhealthy. The “happy children’s ward”
was part of the mythology of the era that hospital staff sought eagerly to
promote. In the case of older children, the myth and the reality did not
always clash. But for the very young, hospital visits, even short ones,
were an emotional catastrophe. In the early fifties, the BBC’s Christmas
programming invariably included a visit to hospitalized children. When
the microphone was brought to older children, they happily consented to
wish their families a Merry Christmas over the air. But toddlers stared
silently and then burst into tears. This happened so consistently that
soon the little ones were omitted from this part of the program.^
As early as 1943, Harry Edelston, a psychiatrist at Leeds, charged that

children were emotionally damaged by their hospital stays. His case stud'
ies were largely ignored. Ten years later, despite Bowlby’s World Health
report, despite the protest of L. A. Parry, whose 1947 Lancet article vigor-
ously protested the restrictions on parental visits to hospitalized children,
despite a 1948 editorial in The Nursing Times advocating hospital reform,
Edelston wrote that many of his colleagues still refused to believe in
“hospitalization trauma.”^
The pediatricians and nurses who ran the children’s wards regarded
visitors as nuisances who disrupted hospital routines and as carriers of
infection (although why parents should be more efficient germ carriers
than hospital staff or student trainees was never explained). It was well
known that children who had settled down after a few days on the ward
typically began to cry and became unmanageable again after parental vis-
its. The staff did not welcome such eruptions.
According to child psychiatrist Edward Mason, who began working at
Children’s Hospital in Boston in 1951, head nurses in some wards had
immense power over which parents could visit, regardless of hospital pol-
icy. In many countries of the Southern Hemisphere, where medical facil-
ities were typically less advanced and where family ties were extremely
 First Battlefield 69

powerful, mothers had routinely slept at the foot of the bed when their

children were hospitalized. But as hospitals improved technically, they

began emulating northern practices.^

The and children endured because of such poli^

suffering that parents

cies was great, especially when the child was very young and
incapable of

understanding his mother’s disappearance for days on end, incapable of

anticipating the next visit. To him, the abandonment seemed, on
each

parting, to be permanent. The feelings engendered in small children

might (very loosely) be compared to an older child’s learning that his

entire family had been killed in a car crash or that everyone he loved on

earth had intentionally abandoned him — and then having his rage, his

sense of loss, and his feelings of being unwanted redoubled when the pet'

petrators returned for a surprise visit but refused to stay or take him with
them.
One can imagine how such a state ot affairs, going on for months or

years, would finally exhaust one’s ability to hope, lock one into a rela^

tivelypermanent hardness, and cement one s anger against those who

had formerly been so loved. Simultaneously, one s hunger for attachment
would continue to incite jealousy and aggression and show itself in diS'
torted forms, like theft. But even when such painful separations lasted for

only a few days, they presented a problem, as these parents testified.

• I left John [twenty'One months oldj, on admittance, hap-

pily playing about the ward, quite fearless in his relationship
with the Sister and the other children. He came home with-
drawn and with no confidence at all. He would not go to his
Daddy and I could not leave him at all, because he screamed

and sobbed. He was frightened of being touched by me in any

way (undressed etc.) and he cowered, literally, before neigh'
bors and friends whom he knew hitherto.^
• When we arrived each day, as soon as she was beginning to

recover that is, we used to see through the observation win'

dow that she was sitting in the cot listlessly staring at nothing
in particular. VC^hen we went in she cried for a moment and
then brightened up for the rest of our stay. As we prepared to
trying
leave she would stand at the end of the cot, desperately
to climb out (something she never attempts at home) purple
with rage and screaming with grief. We left her like that every

day. .
70 What Do Children Need!

• When it was time for me to go, the child went nearly fran-

tic, the Sister told me that she sobbed night and day, even in
her sleep. ... it was absolute torture for both of us when the
time came for me to leave. I couldn’t sleep knowing that my
baby just layed and sobbed and only sleeping when her crying
had exhausted her.

When she came home (after only 5 days) I thought all would
he well, instead she wouldn’t let me put her down to do any-
thing, she wouldn’t go to anyone (not even her father), she

was terrified I was going to leave her, and when I took her
upstairs to bed, and she saw her cot, she almost went hysteri-
cal. In the end I had to go to bed and have her with me, even
then she didn’t sleep for fear I would leave her.^

This last child’s behavior on returning home would prove typical,

although in many cases the clinging was preceded by rejecting the par-
ent. The biologist Konrad Lorenz, whose theories about animal behavior
would soon play a huge role in Bowlby’s intellectual development,
described a vivid example from his own life. Lorenz had left his daughter
when she was three and did not see her again for almost a year. “She
loved me very dearly and when we met again she seemed somewhat
reserved, but she greeted me very politely and nicely and was very glad to
see me. Then when I had left the room, very tactfully so that I could not
hear her, she said to my wife, ‘But our former Daddy was so much more
beautiful.’”®

What could be at work in the mind of a little girl who experiences

such abandonment, aside from anger and rejection? It would seem that
the child, to protect herself against further heartbreak, dissociates from
her emotional investment in the father. He becomes like a figure in a

wax museum, a weak imitation that can arouse no emotion beyond dis-

appointment that he is not the real thing and perhaps angry resistance if

he tries to prove that he is. The rejecting behavior of children who had
been hospitalized against their will by parents whose job it is to protect
them from the intrusions of strangers was often more severe.
In most cases, however, rejection was quickly followed by clinging.
The range of symptoms in this stage included timidity, lost confidence,
pronounced fear of strangers, violent outbursts, often aimed at the
parents, nightmares and nighttime fits, refusal to sleep alone, a furious

jealousy over mother’s attention (“She hated me even to speak to any-

 First Battlefield 71

one else”). Parents often felt they were getting back a different child
altogether, even though these children had all suffered only very brief

separations compared to Bowlby’s affectionless thieves. “It has com-

pletely changed her,” one mother wrote. “I could hardly believe it was

the same wee darling boy,” wrote another. “He had lost so much weight,

his face was pinched and haggard as if he’d been so miserable, and he

could do nothing but hang on to me, and hug me tight. I ll never forget

his first words to me — ‘Mummy, 1 thought you were never coming back
for me.’”^

Even a child of four, old enough to understand that her parents would

he hack soon for another brief visit, able to look forward to that and
know, despite her painful loneliness, that her misery would be temporary,
showed how different a short period of time can seem to a person that
age. Hospitalized for five days, one four-year-old, who had been visited by

her parents daily, later told a neighbor that her parents had taken her to
the hospital and left her there “for weeks and weeks.” It was not uncom-

mon for the small child to feel that being sent to the hospital was a form

of punishment. “I’ll be a good girl,” one child pleaded with her parents,
“please don’t send me.”^^
Most of the parents who wrote these letters appeared to be patient and
devoted. But dealing with a changed and rejecting child was trying. At
one time [the child’s] attitude distressed me so much that 1 found myself

wishing for another baby on whom to shower this surfeit of rejected

mother love. Fortunately for him we did not have another, and now a

year later 1 can say that we are quite normal again.” Another mother
wrote: “I think I got off lightly by loving him back to security instead ot
smacking him into a deeper sense of insecurity as I knew exasperated par-

ents too often do.”^^ volumes about how a trauma like

Her insight speaks

early separation, motivated by urgent need and surrounded

by good

intentions, can worm its way into the fabric of the mother-child
bond,

and through the tormented psychology of one working on the uncompre-

hending hurt and frustration of the other, destroy it.

In 1946, after reentering civilian life, Bowlhy took a job at the Tavistock

Clinic in London, which was going through a major reorganization^^

and

quickly became deputy director, in charge of the children’s

department.

Here, he immediately instituted therapy with both children and

parents

and pioneered seeing whole families together a mode ot treatment he —

remained committed to for the rest of his life.^"^ Child psychiatry was still
72 What Do Children Need?

a new field, and the government was eager to have services set up
throughout the country. By the fall of 1946 Bowlby had the clinical
services and training programs for new therapists operating. In 1948 he
received his first research grants to launch a study of the effects of early
separation.
That year he hired James Robertson, a young social worker, to observe
children who were being sent off to the hospital. Robertson had unique
credentials, having worked in Burlingham and Freud’s wartime nursery.
He proved to be an extraordinarily sensitive observer of children and
their feeling states, and in short order his passion about their treatment

in hospitals was at least as great as Bowlby ’s own.

On 11 February 1948 I went in all innocence [Robertson

wrote many years later] to the short'Stay children’s ward at the

Central Middlesex Hospital, the parent hospital of the TaviS'

tock clinic. This was solely for the convenience of observing a
batch of children in one location, and not because of any
awareness on my part that young children in hospital pre^
sented a special problem. The consultant pediatrician and his
ward sister, for their part, welcomed me into their bright and
The nurses who moved briskly about smiled at the
tidy ward.

young man in an ordinary suit who had come among them.

The atmosphere was of orderliness and of everything being
under control. I was told that this was a happy children’s
ward. But within a few hours I grasped the essentials of a prob'
lem of distress that was not being acknowledged in the hospi-
tal professions. It determined the direction of my professional
life for many years to come.^^*

In social background, outward manner, and life experience, Robertson

was Bowlby’s opposite. Angry about social injustice, deeply sensitive to
and sometimes rankled by class distinctions, and touchy about attribu-
tion and credit, Robertson also had an open, rounded face, a ready smile,
and a gift for gab that might have belonged to an affable TV host. In a

*R()bertson may have forgotten when he wrote of his “innocence” here that his own daughter had
been hospitalized for six days in 1945 when she was only thirteen months old, and that the enforced
separation had been a hitter experience for the parents.
“I offered to scrub the floors or do any task

in could he with her,” Joyce Robertson later recalled, “hut they wouldn’t have
the hospital so that 1

me.” (Milton Senn’s Interview with James and Joyce Robertson, October 25, 1977.)
 First Battlefield 73

beautiful documentary film, Where's Mommy; The Robertson Studies of

Mother^Child Separation, that Edward Mason made about Robertson,

Bowlby is, by comparison, so dryly upper crust and in control of his feeh
ings that the two men seem to come from different planetsd^

One of seven children from a poor Scottish family, Robertson grew up

in Clydesdale, near Glasgow, where housing conditions were among the
worst in Europe. The family lived in a cold water flat and shared an out-
house with two other families. From age 14 to twenty^eight, Robertson
worked in the steel mills to help support his family and read voraciously
at night. At twenty^eight he attended a one^year school for the higher
education of working men and women. When war broke out, Robertson,
a conscientious objector, and his soon^to^be wife, Joyce, went to London
to help out with the families of the dock workers, whose neighborhoods
were being heavily bombed in the blitz. Shortly thereafter they were

hired by Anna Freud to work in the Flampstead nursery around the cor-
ner from the Freud home where Sigmund Freud had died a year earlier.

Robertson had been disturbed by the evacuation of trainloads of chib

dren to remote parts of Britain (a process he had participated in), the diS'
tress of the children having been apparent to him. He was therefore

attracted to the nursery’s practice of keeping the children near enough

for the mother to visit and Burlingham and Freud’s philosophy that
to

the children were put at greater risk by being separated from their moth'

ers than from being near falling bombs.

Robertson became the handyman on the property, tending to boilers
and carpentry as well as organizing the fire^watching services, Joyce one
of the caretakers, tending to the babies from 6 a.m. when they awoke

until they were tucked in at night. In the meantime, with bombs falling

in the background, the young couple attended staff meetings and semi'
nars and learned psychoanalytic theory. By the end of the war, Robertson
was doing social work with the families, helping to determine how the
children would be resettled.
By the time Robertson, now a qualified social worker and training in
psychoanalysis, went to work for Bowlby, he, his pregnant wife, and their
first child were living in a hostel for homeless
people, where they

remained for two and a half years. “Surely very tew psychoanalysts,’’

Robertson later said, “had done the major part of their training in such
strange domestic conditions.”^ ^ Robertson now joined Bowlby’s sanatO'

rium study, an ambitious project designed to follow up children, aged one

these chib
to three, who had been hospitalized. His job was to observe
74 What Do Children Need!

dren as they were being admitted and to record their reactions. He also

observed children in a fever hospital and a residential nursery.

At the Central Middlesex Hospital Robertson saw desolate children
silently sitting on their cots. He saw a kindhearted nurse rebuked for
stopping to show concern for a silent toddler. He saw a child who was
seriously ill in an isolated area attended to by a special nurse; the nurse
sat quietly behind the child’s cot so that the sight of her would not cause
him to make exceptional demands. On Sunday when the parents came
the silent ward rang with anguished cries. When Robertson expressed
his concern to the hospital staff he was told that the children’s distress

was inevitable and that they all soon settled down. “Doubts 1 expressed
were smilingly put aside as those of a sentimental psychologist. But if 1

followed some of the children back home I found invariably that the
‘settled’ state in the hospital ward had given way to difficult behavior
clinging to the mother, temper tantrums, disturbed sleep, bed-wetting,
regression and aggression particularly against the mother as if blaming
her.”^^

The hospital staff had an explanation for this, too: The mothers were
simply less competent than the nurses in caring for children.

In the course of the next two years, Robertson also observed children
in a long-term ward, most of whom suffered from pulmonary tuberculosis,
still a scourge at that time and regularly requiring long hospital stays,

sometimes up to four years. Here he was able to observe the process that
Bowlhy and the other maternal deprivation pioneers only knew by its
end product. These children typically went through three stages of emo-
tional reaction to their separation. Robertson called the first stage
“protest” — crying, clinging, screaming after parents had left. In this
stage the child is frightened and confused and urgently looking for his
lost mother, turning anxiously toward any sound that might herald her
approach. This is gradually succeeded by “despair.” The child becomes
listless, loses interest in his surroundings, turns away from food, sheds
only occasional tears. He desperately wants his mother, hut, Robertson
said, he is losing hope of getting her back. Robertson had already seen
these two stages in the short-term wards. But in the children requiring
long hospitalization, he observed a third stage that he first called
“detachment.” The child now seems to awaken somewhat to the possi-
bilities of ward life. He interacts more. He smiles.He eats. But he has
awakened a different child. He no longer seems to know his mother
 First Battlefield 75

when she comes, to cry — or even to appear concerned when she

leaves. Gradually, he seems to become indifferent to maternal care alto-

gether.

On Sundays these “detached” young children seemed indiffer^

ent to the arrival of their parents. There was no crying, no
clinging, no turning away. They were more interested in what

their parents had brought than in the parents themselves.

They searched bags and stuffed chocolate into their mouths.

My impression was that feeling for their parents had died

because of the passage of time and repeated disappointment of
the wish to be taken home and had been replaced by hunger
for sweet things which did not disappoint. “Don’t he greedy,”
perplexed parents would say.^^

Among the children Robertson observed during this period was Mary,
a cuddly child of two and a half years. When she was discharged
a year

later, she seemed no longer to need her mother, squirmed

out of her

embrace, and then stood looking at her “as if not knowing what affection

was about.” Robertson, who had grown attached to Mary, was able to fob
low her for a number of years, seeing her turn into a self-centered girl,

ruthless in her relations with her older brothers, and, as she grew
older,

increasingly solitary. Although Mary did well in school, Robertson was

pained to find that, at twelve, she was unpopular with classmates and dis-

liked by teachers and staff. He tried to enlist the compassion ot the adults

around her, earnestly explaining the ordeal Mary had suffered, but to no
avail. “Disappointed though 1 was, 1 admitted to myselt that 1 would not

wish a son of mine to marry Mary.”^^

As he became more engrossed in his work, Robertson grew increas-

ingly infuriated with the unwillingness of professionals, outside of

Bowlby’s team at the Tavistock, to believe what he was saying. The most
painful blow came in 1951 when Robertson presented his findings to a

group of pediatricians. Nervously aware of his low status, meager creden-

tials, and poor, working-class roots, Robertson
was crushed when the

only doctor he thought he could count on as an ally. Sir James Spence, a

pioneer in the humane treatment of hospitalized children who had

privileges to many
already extended visiting rights and even overnight
parents, bitingly attacked him. “What is wrong with emotional upset?”
76 What Do Children Need?

Spence said. “This year we are celebrating the centenary of the birth of

Wordsworth, the great Lakeland poet. He suffered from emotional upset,

yet look at the poems he produced.
Bowlhy would long recall the day in 1951 when Robertson told him
that since nobody would accept his evidence he was going to make a
film. It would he a primitive film, made with a Bell and Howell spring'
wound 16mm camera — hand held and relying on window light. A child
subject was selected by sticking a pin into the fout'page waiting list for

elective surgery at the Central Middlesex Hospital. Because there was

some concern that Robertson might only show the child when distressed,

he and Bowlhy worked out a system whereby the toddler who’d been
selected would be filmed during the same forty'ininute periods twice
each day, as shown by the clock, which was always in the frame. The
child’s name was Laura. She was two years, five months old.

“When I saw Laura,’’ Robertson later recalled, “I got a shock. For three
years I had been talking to pediatricians about the states of acute distress
little children invariably got into in pediatric wards, how when they lost

their parents, when they were ill and in pain, these little children
shouted, screamed, rattled their cots, their noses ran, their eyes streamed
with tears, how they tried to climb out of their cots to get at the mother
who had disappeared. IBut Laura] was what we call in Britain ‘a little

madam,’ a child of parents who were older than usual in having their first

child and who had reared her with love but with strictness. At two years
and five months, she was very well behaved, a very controlled little girl

who, 1 saw immediately, was going to be the one child in a hundred who
was not going to demonstrate what I had been shouting my head off for

about three years.

Robertson was tempted to go hack and ask for another child, hut he
knew that that would destroy any hope of his being perceived as ohjeC'
tive. He proceeded.
The Laura we see in the film, who is now known to child develop'
ment experts throughout the world, is an exceptionally beautiful little

girl with straight blond hair pulled hack on top with a ribbon. She
proved to he as selLcontained as Robertson feared. She never wailed in a
prolonged and uninhibited manner. But her feelings continually broke
through her composure, which she then struggled to reassert. On the first

day, after the routine hath, she ran naked to the door and tried to escape.

Her expression, formerly bright, became dull, and she periodically sobbed
 First Battlefield 77

quietly to herself, clutching her teddy or a favored blanket. Several times

she quietly asked, “Where’s my mummy?” She later looked more and
more tense and miserable, despite heroic efforts to keep from crying.
Visiting was normally restricted to once a week, but because Robert-
son wanted to observe how her attitude toward her parents would change
over the eight-day stay, at his request Laura’s parents were allowed to

visit every other day. On the second day of her hospitalization, her par-
ents visited for the first time. On seeing them, she immediately burst into

tears and pleaded with them to take her home. In one poignant moment,
we see her looking into her father’s face in a heartbreaking expression of
supplication, as she clutches his tie — something she frequently did when
distressed — and pleads pathetically, “Don’t go. Daddy.”

In each successive visit the parents were greeted by a grayer, colder

response, until she seemed frozen. On the fifth day she looked resentful
and unsmiling when her mother appeared, and wiped away her mother’s
kiss. Her lip trembled when her mother said she was going home. When

her mother waved good-bye on the seventh day Laura stood motionless
with her head bent.
When her mother came to fetch her on the eighth day, Laura was
apprehensive, as if not allowing herself to believe that she might really
be going home. She greeted her mother shaking with sobs. When at last

they prepared to leave, Laura scoured the ward for her belongings,
including a tattered book her mother suggested she leave behind. She
seemed determined to leave nothing of herself there. As Laura marched
out of the hospital, clutching her belongings, she refused to take her
mother’s hand.
Two days after returning home, Laura was a bright-eyed child again,
“as though a lamp had lit up inside her,” thought Robertson at the

time.^^ But like other children who experienced brief hospital, stays, she

went through a period of anxiety and irritability after she returned, sleep-

ing poorly, soiling herself, refusing to let her mother out of her sight,

throwing violent temper tantrums, often just after seeking an embrace.

Five months after her surgery, Laura suffered something of a relapse.

She was staying with her grandmother, someone she knew quite well,

while her mother was in the hospital for five weeks because of complica-
tions following childbirth. For some reason, which was never made clear,
her father thought it better not to visit her during that period, which
must have contributed considerably to the stress of separation. When
Laura’s mother called the grandmother’s house after returning from the
78 What Do Children Need!

hospital, Laura was excited to hear her voice and eager to go home. She
went home in a state of excited expectation, hut when her mother

greeted her at the door, her excitement disappeared and she seemed to go
blank. After a pause she said, “But I want my mummy.” And for the next

two days she continued in that vein, refusing to accept her mother,

although she accepted her father instantaneously and without apparent

problem.
A month later, when she accidentally walked in on a portion of
Robertson’s film, which he was screening for her parents, she burst into
tearsand cried to her mother, “Where was you all that time?”^^
In November 1952, James Robertson and John Bowlby went to a
packed meeting of the Royal Society of Medicine in London to present
the film. The film, which cost about eighty dollars to make, was lowTey.

Far from showing a distraught, hysterical little girl, it presented a child

who remained fairly well in control throughout, her emotions revealing
themselves mainly through little signs, like the twitching of her fingers

or the narrowing of her eyes. Still, it was powerful, more powerful per-
haps than the documentary of a more uninhibited child might have
been. Although Laura was self-contained and Robertson’s documentary
caught only a few episodes of weeping, it nevertheless proved powerful

evidence of Bowlby’s theories on the dangers of early separations because

of the poignant drama of the child’s struggle with distress. Like Spitz’s
film of a few years earlier, it forced people to see that little children do

indeed suffer a great deal. The audience, some three hundred pediatri-
cians, nurses, and administrators, was outraged.
“I was immediately assailed for lack of integrity,” Robertson said. “1

had produced an untrue record. 1 had slandered the professions. 1 had

tricked the hospital; 1 had chosen an atypical child of atypical parents; 1

had . . . filmed selectively and 1 had edited it dishonestly. People stood up

and said that their children’s wards were not like that, two-year-olds were
all happy, no parents ever complained, etc., etc.”^^^ Many demanded that

the film he banned.

Some wards to which Robertson had had access were now closed to

him. “Pediatricians who had welcomed me into the wards from that day
on did not welcome me. Some were personally hostile and walked across
the street when they met me.”^' Attempts were made to discredit him. It
was said that Laura was never filmed when she was happy. The Central
Middlesex staff charged that Robertson had interfered with nurses trying
to minister to Laura. A child health expert protested to Bowlhy, saying
 First Battlefield 79

that Robertson had failed to he objective, and citing as evidence his pan^
ning from Laura’s emotionless face to her twitching fingers. As Rob'
ertson toured with the film, the reaction was similar throughout Britain.
On two occasions, the chairman of the proceedings abandoned his neu'
tral role to join the attacks on the guest. A Glasgow nurse told him,

“Yon may be true of a wee English lassie, but it’s not true of a Scottish
bairn!”^^ Professor Spence, inexplicably, was as dismissive as before, but
attempted to placate Robertson with the reassurance that he preferred
Robertson’s simple, misguided views to the “more dangerous attitudes” of
Bowlhy.^^ Resistance in the medical profession was so stiff that Bowlby
and Robertson temporarily withheld A Two^Year-Old Goes to Hospital

from general release to allow the turmoil to subside.

“The medicals, many of them, were in a dilemma because Dr. Bowlby
was a respected medical and one of themselves,” Robertson recalled,
“whereas was a nonmedical and an intruder. Dr. Bowlby and I, of
I

course, were in complete agreement about the problem of the pediatric

ward and about the truth of my film, hut some of the medicals began to
try and split us. It began to get around that Dr. Bowlby — decent chap
and one of us — had modified which of course he hadn’t, and
his views,

that it was this nuisance, Robertson, this interloper, who went on plug'
ging the totally untrue, psychological, sentimental stuff about the dan'
gers of early hospitalization.”^'^

The reactions of the British analysts, who had seen an earlier screen'

ing of the film, was divided. Anna Freud was largely supportive, but the
Kleinians rejected the hypothesis that the child’s distress had anything to
do with separation from the mother. Wilfred Bion, a prominent Kleinian
who had served with Bowlby in the army, argued that Robertson and
Bowlby had missed the point and that Laura was probably upset because
her mother was pregnant.
Surprisingly, the reviews that soon began appearing in the medical lit'

erature were uniformly positive. And younger nurses and pediatricians

frequently buttonholed Robertson to tell him that they agreed with him
and that they would do things differently if they were in charge. A few

higher'Lips were also sympathetic. The medical superintendent of Cen'

tral Middlesex said, “Well, I am sorry it is my hospital. But accept that I

it is an objective record which should be shown widely. Such reaC'

tions gave Robertson the courage to persist.

In 1953 the World Health Organization made Robertson a short'term

consultant and sent him on a six'week tour with the film to the United
80 What Do Children Need!

States. There he was commended for having done an admirable job of

documenting conditions in England but informed that his film had no
relevance to the United States, where children were less cosseted and

thus better able to withstand separations. Even where the problem of

separation was understood, Robertson saw little interest in the obvious

solution — allowing mothers onto the wards.

In 1955 pediatrician^psychiatrist Ered Stone returned to Britain after
three years of training in the United States. “I arrived in this hospital,”

he told historian Alice Smuts, “one month after Jimmy Robertson had
given his first presentation of his film A Two-Year-Old Goes to Hospital in

Scotland. Can you imagine the atmosphere I arrived in? And I was
greeted with howls of rage and despair and aggression — and, of course,

‘You’re one of us; you know that this is all rubbish.’ And, ‘We’re going to

get you a big research grant to disprove all this Bowlhy nonsense.’”
At the time, visiting arrangements for parents at the Royal Hospital
for Sick Children in Glasgow were two halThour sessions per week. “I

would drive up the hill here to come to an afternoon clinic, and there
would be two hundred parents queueing up in the rain to get in for their

half hour’s visiting.” Eventually, one of Stone’s pediatric colleagues sug'

gested a pilot study. The colleague had jurisdiction over two pediatric
wards and was willing to allocate one of them for unrestricted parental
visits for six months to prove that it wouldn’t work. Stone readily agreed.
When the nursing staff got wind of the plan, however, they threat'
ened to resign en masse. Stone suggested that meetings be set up
between his staff and the nurses, twice a week tor three weeks, before

anything further was done. “And I think it’s true to say that never has so

much distress been aired in so short a time in our department. The

aggression after the first meetings was unbelievable, truly unbelievable.”
But gradually, with the exception of one nurse, who did, in fact, resign,

the aggressive posture gave way to tears.

“They said, ‘You don’t understand what you are asking of us. You’re
really meaning to say that any time of the day, no matter what we’re
doing, no matter what state of chaos we’re in, a parent can just walk in

and see how we’re neglecting these poor kids, that they’re lying there in

their own excreta, that they’re screaming because they have to have an
injection, that somebody’s forgotten to get their diet and we’ve got to do
the whole procedure all over again — you want this?’ And, of course, at

that point we had to say, ‘But what on earth makes you think that we’re
 First Battlefield 81

criticizing what you’re doing? You’re doing an impossible job remarkably

well.’ And, of course, then the tears came, as you can imagine.”
The nurses had other causes of anxiety, too. “They talked about the
mothers who were ungrateful and critical and dirty and lice'tidden — you
can imagine. Well, they all agreed that there should be a pilot experi'
ment to see whether it would work.”
The experiment, however, was never actually concluded. In fact,

Stone says, “1 never heard any more about the issue at all. Nobody ever

came back to me and said, ‘The six months are up.’ Nobody ever

reported that it had been a success or a failure; all I knew was that some^

how I heard that two wards were doing it, four wards were doing it, the

whole hospital was doing it. And since then we’ve had almost unre^
stricted visiting in the whole hospital.

Another early convert was Dermod MacCarthy, a pediatrician at the

Amersham General Hospital outside London. MacCarthy had seen the

initial screening of Robertson’s film and had felt that it had slandered
pediatrics. He complained about it afterward to his ward sister. Ivy

Morris, who responded, “But what Mr. Robertson says is true.” Later, he
wrote, “I was angry, but after the film I really heard children crying for
the first time.”^® MacCarthy’s recognition that he had repressed all

awareness of the young patients’ suffering was a typical experience tor

many who view Robertson’s film. An elderly nurse told Bowlby: “This
film brings back to me the first child I ever nursed in hospital. This child

was a little boy. He grieved for his mother and it simply broke my heart.

After that I never saw grief again until I saw this film.”^^

MacCarthy soon opened his ward to unrestricted visiting and encour'

aged mothers of children under five to sleep in. Staff found that their
anxiety about mothers’ getting in the way, mothers stuffing their chiL
dren with sweets, and the like were ilLfounded. On the contrary, moth'

ers were usually welcome as valued additions to the team.

The changed policy gave Robertson the opportunity for his next film,

Going to Hospital with Mother. The film documents the fiveMay hospital'

ization of twenty'month'old Sally, whose mother is with her the whole

time. Though she gets upset at times, Sally never shows the despair or
rejection of her mother that was typical of children admitted on their

own. We see the comfort she gains from her mother’s presence in numer'
ous situations — getting her face forcefully washed, being examined by a

strange doctor, getting an injection, waking fitfully in an unfamiliar

place. Sally never becomes fretful or withdrawn, and upon returning
82 What Do Children Need?

hcMne, displayed none of the behavioral changes characteristic of other

returning children. When released in 1958, Going to Hospital with Mother

was well received by the profession, in marked contrast to its predecessor.

This did not, however, herald the advent of major changes. Change
came slowly, and restricted visiting lasted in various forms through the
next three decades.
“Pediatricians had little dynamic understanding of child develop'
ment,” Robertson said later. “It would be said of my film, ‘Yes, that is a

pleasing film; the child was contented with the mother by her side, but
these changes are too difficult and too costly to make. They are not, in

fact, wholly necessary. Some more visiting will be enough.

On the Continent, resistance was even stronger. It was not until the
late 1970s that pediatricians there began considering the adjustments
that were slowly taking hold in Britain and the United States.
In 1957 the British government established the Committee on the
Welfare of Children in Hospital, chaired by the distinguished surgeon Sir
Harry Platt. Two years later the committee accepted all of Robertson’s

recommendations, and the minister of health announced that they were

henceforth ministry policy. He urged that all hospitals work toward
implementing the recommendations. Robertson, meanwhile, had shaped
recommendations into a book. Young Children in Hospital, which was
quickly translated into nine languages. In the coming years he wrote
influential articles for British newspapers, appeared on BBC'TV, and
continued his travels. In the early 1960s, he spoke on BBC radio, and in

a hold stroke told parents that “if they decided to stay quietly by the cot

of their sick child they could not be evicted.” Several parents wrote to
say that they had tried it and that it had worked. In some cases hostility

on the part of hospital personnel eventually gave way to sympathy and

even the offer of a cup of tea.^^^

In the mid' 1960s Robertson proceeded to make a series of films called

Young Children in Brief Separation, which included an unusual and dra'

matic, not to mention lahot'intensive, procedure: In four cases he and

Joyce became the foster parents of small children whose mothers were
hospitalized. They wanted to see what effects separations might have if

they took place under near ideal circumstances — that is, without the
potentially confounding variables of illness, a strange hospital setting,
and inadequate substitute care.
 First Battlefield 83

In each case, the children were healthy and happily attached to their
parents with no previous experience of separation. They were allowed to
become familiar with the Robertson home before moving in, and Joyce
found out all she could about their likes, dislikes, routines, and diets.
Fathers were encouraged to visit, to dine with the children, to bathe

them and put them to bed. In the case of the two'and-a'half-year'old

Kate, Joyce visited the family a month before the separation and the

child formed a strong attachment to her. Only gradually did Kate show
signs of anxiety after moving in with the substitute family. In one
poignant play episode, Kate re-creates her own family apartment, show-

ing that she loves her parents and wants them back but the play parents

do not seem to want her. “Throw them in the dustbin!” she cries. She
demands that they be banished from the room but soon cries to have
them back.'^^

For Jane the anxiety was somewhat worse. She displays it in the film
with excessive unnatural laughter and with an exaggerated smile that
seems designed to force smiles from the foster family. When her father

visits, she becomes remote but then clings to him and cries when he tries

to leave. She becomes distressed when she and Joyce walk past her home,
which is nearby. The Robertsons believed that, at seventeen months,

Jane had a difficult time both assimilating Joyce as a new person when
Joyce made her advance visits and retaining a clear image of her mother
once away. In the end, she became extraordinarily attached to Joyce, and
remained so even after she went home —
where she had to share her
mother with a new baby. Although the Robertsons were concerned
about subtle changes in her development, they believed that Jane’s expe-
rience with the foster care was nevertheless largely positive. Despite her
evident troubles, she functioned well enough in the Robertson home to
learn new words. When her mother retrieved her after ten days, Jane
went to her instantly, inviting her back into a maternal role."^^

Of the four children, Thomas, the oldest, suffered the greatest conflict

of loyalties and had the most difficulty accepting Joyce’s care. In the film
he displays and ambivalence by mixing his affection toward
his anxiety

her with aggression. Fie also avoids looking at his mothers picture. The
Robertsons considered such emotional difficulties to be a state of man-

ageable anxiety,”"^"^ which the children were able to live with and, for the

most part, work through. Narrating the aftermath of her caring tor little

Lucy, Joyce Robertson recounts that at Lucy’s request she visited the tarn-
84 What Do Children Need?

ily several times after the reunion. “My last visit was an outing to the
park. Lucy invents a game which plays out the separation and reunion.
She walks me away then runs hack to her mother. Then she reverses the

game. But she never returns to me.’’"^^

The Robertsons believed that “no matter how good the substitute

mothering, separation from the mother remains a hazard for the young
child because of the discontinuity in their relationship.’’"^^ But each of
these four toddlers was able to form a bond with Joyce, and after a sepa^

ration of ten to twenty^seven days, reunite happily with his mother. The
Robertsons were thus convinced that, if a substitute mother is available,

the child will reach out to her and enter into a relationship that will
both reduce the child’s stress “and sustain his experience of responsive
mothering'type care.’’"^^ On the basis of such evidence they urged that
foster care be the rule for children in brief separations and that hospitals
have foster mothers on staff to be assigned to unaccompanied children.
In 1969 Robertson released John. The seventeen^month^old John was
taken to a residential nursery for nine days while his mother was having
a new baby. The nursery was well staffed and amply provided with toys.
In the first day four different nurses attend to him, all of them kindly and
cheerful, all of them apparently appealing to him. He is unable to keep
any one of them by him, however, as this is not the way the system
works. His tentative efforts at directing his love toward a new person are

dashed as she abandons him and is eventually replaced by another. Most

of the other toddlers, residents almost since birth, are accustomed to this
life. They are noisy and aggressive, quick to defend themselves and to
snatch things they want, unperturbed when a nurse leaves the room and
unsmiling when she returns. John, initially interested in the new chiL

dren, is soon perplexed by them and doesn’t know what to do when they
pinch or smack him.
On the second day, John is still coping well but unable to get a nurse
to pay attention to him, often because the nurses are responding to more
clamorous children. His distress building after the third day, he becomes
gradually more isolated and moves about aimlessly. The nurses, accuS'

tomed to intermittent crying, miss his signals of despair. He seeks com^

fort from a large teddy hear. When *
his father visits, he is slow to respond
but fights furiously to leave with him. On the fourth day a nurse who had
comforted him two days earlier returns hut he is no longer interested in

her. His crying loses its edge of protest, becoming weak and pitiful. By
the time the nurses recognize the seriousness of his condition, it is too
 First Battlefield 85

late to help, especially since none of them is available for the continuous

attachment he needs. By the fourth day he is refusing food and by the

eighth he hasn’t eaten for several days. When his mother at last appears

to take him home, he begins to wail and throw himself about. He avoids

looking at her and refuses to allow her to comfort him. When their eyes

meet at last, he looks at her in a way she’s never seen before.

“The making of the ‘John’ film” Robertson later said, “was a very, very

distressing experience. We couldn’t intervene any more than Joyce giv'

ing an occasional cuddle to the child. We couldn’t take him over.

Otherwise we’d have been thrown out of the institution.” On returning

home, John showed many of the signs of insecurity that had become
familiar to Robertson, and they lasted for months. “This experience,”
Robertson said, “left John with a profound anger against his parents and
a deep sense of insecurity.” He feared the child would be permanently

damaged by it."^®

The reception of John was almost as tumultuous as the Laura film had
been seventeen years earlier. Again professionals and government offi'

dais were dealing with a phenomenon that had been invisible to them
before. Many reacted with outrage, but the period of resistance was
shorter, and reviewers were almost uniform in their praise: “A horrifying

film which forces us to look at what despair is for a young child” The
British Journal of Psychiatric Social Work. “No words can convey John’s
stress reactions as powerfully as the camera does. The impact of John’s

hour by hour increasing misery and deterioration becomes almost un^

bearable” The Journal of Child Psychotherapy. The Lancet called the film
a landmark and said it would do for residential care what Laura had done
for pediatrics. Indeed, it soon helped put an end to the use of residential
nurseries in Britain and promote foster care as the preferred alternative.

Robertson had become by now something of a one-man institution,

invited to give lectures and show his films around the world. For twenty
years he needed a full-time secretary-assistant, much of whose time was

devoted to the deluge of mail and telephone inquiries that poured into
Robertson’s little room at the Tavistock. In 1961 he oversaw the creation
of the National Association for the Welfare of Children in Hospital.
Robertson eventually quit the organization, angered by what he consid-
ered to be a slight, but it continues his work to this day.
Both he and Joyce had become markedly alienated from Bowlby as

well. Part of the subtext of the making of the series on Young Children in
86 What Do Children Need!

Brief Separation had, in fact, been a theoretical dispute with Bowlby.

Bowlby had held that separations from the mother figure caused distress
and despair in most children regardless of how they were treated during
the separation, a point to which Anna Freud had strongly objected.
When Robertson, who remained devoted to Freud, proposed the new
series of films to investigate this issue,Bowlby apparently displayed a
lack of interest, which struck Robertson as “very like resistance. There
was also speculation that he and Joyce resented Bowlby s appropriat-
ing or getting credit for James’s ideas. At times their distaste for him
seemed oddly impassioned. But, like it or not, their careers remained tied

to his name.
Bowlby, true to form, continued to credit Robertson for his theoretical
and practical accomplishments and to speak highly of him at every turn.
In his 1973 book on separation, he acknowledges that the Robertsons

foster study caused him to modify his views on the traumatic conse-
quences of separation, “in which insufficient weight was given to the
influence of skilled care from a familiar substitute.” Although he was
no doubt hurt and confused by their continued rejection, he gave no
hint of it in his public utterances (and apparently not in private either).
His blurb on the back of Robertson’s last book, which appeared in 1989,

the year after Robertson died, is a fitting epitaph: “James Robertson was a
remarkable person who achieved great things. His sensitive observations

and brilliant filming made history, and the courage with which he dis-

seminated — often in the face of ignorant and prejudiced criticism

what were then very unpopular findings, was legendary. He will always

be remembered as the man who revolutionized children’s hospitals,

though he accomplished much else besides. am personally deeply grate-
1

ful for all that he did.”^^

Bowlhy’s career had meanwhile taken some new and surprising turns.
Always quick to grasp the theoretical implications of new data, Bowlby

had right from the begun including Robertson’s formulations in his

start

own writings. The stages of protest, despair, and detachment immedi-

ately became part of the standard language of maternal deprivation. But

as early as 1951, Bowlby, wholly convinced of the truth of Robertson’s

observations — and believing that they called fundamental psychoana-

lytic assumptions into question^^ — was already off in another direction.

 7

Of Goslings and Babies:

The Birth of Attachment Theory

When a duckling is hatched it attaches itself to the first moving object it

sees. Almost invariably that will be its mother; although if a human sci'

entist elbows his way into first view, the duckling will become hopelessly

attached to him and follow him everywhere. Other instinctive behaviors

can similarly be distorted, or fail to develop at all, depending upon what

the young animal encounters or fails to encounter in its environment.
We know this and many other facts about the bonding behavior of birds
and mammals thanks to the work of ethologists like Konrad Lorenz and
Niko Tinbergen.
Bowlby first heard of Lorenz, now considered the father of modern
ethology, from Norman Hotoph, a young psychologist at the London
School of Economics. Hotoph mentioned Lorenz’s work on imprinting,
the process by which early bonding takes place, and Bowlby, absorbed in
questions regarding the infant-mother bond, managed to obtain Lorenz’s

1935 article on the subject, “The Companion in the Bird’s World.’’^ The
qualities of imprinting — a strong bond, which occurs quickly and firmly
and is unrelated to feeding — had great appeal for Bowlby. Indeed, after

reading Lorenz, it seemed that Bowlby had become imprinted himself.

88 What Do Children Need!

In the summer of that year, John and Ursula Bowlby took their vaca^
tion in Scotland near Ursulas father, who was an ornithologist. Julian

Huxley, the biologist, was visiting at the time. “I inquired tentatively,

Bowlby said, “did he know anything about Lorenz and ethology? This led
to an enthusiastic commendation from Huxley who knew him very well
and who had done work of his own in this field.
Since Darwin’s time zoologists had been cataloguing the instinctual
behavior of animals as a means of classifying them. Darwin, an ethologist
before the name was invented, devotes a chapter of The Origin of Species
to animal behaviors, noting that each species has its own peculiar reper-

toire of instinctual behavioral patterns, which are as fundamental to the

species as its The problem with the term “instinct,” however,
anatomy.^
as it came to he used by many biologists, is that it implied a rigid, inher-
ited behavior, performed identically by every member ot the species.
Those who believed that animals were governed strictly by instincts
reduced them essentially to machines that were programmed to behave
in certain ways from birth.

But as naturalists began to study animal behavior in the wild, a new

view, closer to Darwin’s meaning, began to coalesce. Lorenz and Tin-
bergen moved away from the idea of rigid instincts, favoring instead

“species-specific behavior.” A species-specific behavior (a bird’s peculiar

song or nesting activity) is instinctive, but it is not preordained. In order

to come into being it must encounter certain responses in the environ-

ment. Thus, the song a chaffinch will sing is limited to a certain range

and quality by its genetic predisposition. But it must hear the song of an
adult chaffinch if its song is to develop at all. What’s more, it must hear
the adult song within a certain sensitive period when it is receptive to

this particular kind of learning. The same is true ot imprinting. It a duck

or gosling does not see a moving object within a relatively hriet time

after hatching, normal bonding will not take place.

The interplay between instinctual behavior patterns and events in the

environment was found to exist throughout the animal world, with the
young of each species prepared to react to significant symbols and events.
Ethologists found that the mating behavior of the male stickleback Hsh is

elicited by a shape resembling a pregnant female; the young herring gull

opens its mouth wide in response to a red spot similar to the one found
on the beak of adults of the species. Behavior involving mating and par-
enthood could in many species be modified by learning, turned ott when
no longer needed, or strung together in long sequences to produce com-
 Of Goslings and Babies 89

plex social interactions that climax in copulation or that, when taken as

a whole, comprise caring for the young.

To Bowlhy was immediately plain that the discoveries of the etholo'
it

gists must apply in some form to human beings, that people, too, must

have bonding behaviors and intergenerational cues, that they, too, must
he prewired for some sort of relational experience, and that with them,
too, nature’s intentions could go awry if the environment failed them as

it had with his young thieves. Intellectually, his encounter with ethology

was the breakthrough of his career and the beginning of a lifelong love

affair with Darwin, whose thinking formed the foundation upon which
ethology was built.

“At that time,” Bowlhy recalled, “the conventional wisdom was that
infants are only interested in mothers because mothers feed them.” Anna
Freud, without irony, called this “cupboard-love,” a term that delighted
Bowlhy and that he used to great derisive effect for many years."^ “1 was
profoundly unimpressed by that,” he said. “I just didn’t think it was true,

I knew it wasn’t true. There was a lot of fancy talk about breast-feeding
and bottle-feeding and so on; 1 regarded it all as rubbish. It was com-
pletely contrary to my clinical experience. There were very loving moth-
ers who had bottle-fed their babies and some very rejecting mothers
whom met I in the clinic, women who were obviously very hostile, who
had breast-fed their babies. And it seemed to me that the feeding vari-

able was totally irrelevant, or almost totally irrelevant. So 1 was unim-

pressed with the conventional wisdom, but 1 had nothing particular to
^
put in its place.”

He needed something. Critics were already insisting that the children

in the maternal deprivation studies Bowlhy had made famous were not
suffering from the loss of their mother, were not even suffering from the
loss of a mothering figure, but simply from a lack of adequate stimulation.
Behaviorists in particular were making this point, and few of the studies
were strong enough empirically to prove otherwise. Many of the babies
who had suffered lengthy separations from their mothers had been placed
in institutions where they were left in near total isolation for much of

the day, with no one and nothing to play with. Behaviorists, therefore,
were able to look at Bowlhy ’s data and fit it comfortably into their own
theories, where the love of a mothering figure was completely unneces-
sary to healthy emotional development. Most people who worked with

children had a hard time buying that idea. It was hard to believe that all

a child needed was enough stimulation and that it didn’t matter where it
90 What Do Children Need?

came from or whether there was a single person with whom the infant

could form a meaningful relationship. Such thinking seemed to be so

utterly opposed to what intuition tells us about a child’s needs. But sci'

ence cannot run on intuition. It needs a theory grounded in data. The

ethologists provided that. And Bowlhy responded to it with enthusiasm:
“1 mean talk about eureka. They were brilliant, first'class scientists, hrih

liant observers, and studying family relationships in other species — rela-

tionships which were obviously analogous with that of human beings

and doing it so frightfully well. We were fumbling around in the dark;
they were already in brilliant sunshine.”^
In addition to improved strategies of investigation, ethology gave
Bowlhy an explanation: Separations from mother were disastrous devel-
opmentally because they thwarted an instinctual need. It’s not just a nice
thing to have someone hilling and cooing over you, snuggling you, and
adoringly attending to your every need. It is a built-in necessity, and the
baby’s efforts to obtain it, like the parents’ eagerness to give it, are bio-

logically programmed.
Bowlhy would soon declare that sucking, clinging, following, crying,

and smiling are all part of the child’s instinctual repertoire, whose goal is

to keep the mother close by. He saw the child’s smile as a “social releaser”
that elicited maternal care. And, like Ronald Fairbairn, he abandoned
the Freudian notion of drives, based on hidden forces like libido and
aggression, that accumulate within us and crave discharge. Instead he
saw an array of innate behavior patterns — relationship-seeking patterns
like babbling, looking, and listening — that are enriched and developed
by the responses they receive from the environment.^
Bowlhy introduced the formal term “attachment” to describe the
infant-mother bond. Unlike bonding, which suggested an instantaneous
event (and which years later was briefly and erroneously considered crit-

ical to a healthy mother-infant relationship®), attachment suggested a

complex, developing process. Indeed, to Bowlhy attachment was much
closer to the idea of love, if not identical with it. He now proceeded to
define a new series of developmental stages based on the maternal bond.
During the first year the child is gradually able to display a complete

range of “attachment behaviors,” protesting mother’s departure, greeting

her returns, clinging when frightened, following when able. Such actions
are instinctual and rooted in the biological fact that proximity to
mother, because it is essential to survival, is satisfying. The formation,
maintenance, and renewal of that proximity begets feelings of love, secu-
 Of Goslings and Babies 91

rity, and joy. A lasting or untimely disruption brings on anxiety, grief,

and depression.

In much of Freud’s writing, the tie to the mother is curiously absent. He

believed that after a brief love affair with the breast, the infant moves
quickly into a stage of primary narcissism in which his main love object
is his own body. Only much later, after the age of two, does the child

become focused on the opposite-sex parent as an object of desire and

love. The relationship of the child to its mother during the first two years

of its life was a nonissue in classical Freudian thought.

Even when he did consider the power of early bonds, Freud was
strangely fixated on father^love. It is not until 1938, just before his death,
that Freud spoke with some feeling about the power of the early tie to
the mother, describing it as “unique, without parallel, laid down unalter^

ably for a whole lifetime, as the first and strongest love-object and as the

prototype of all later love relations — for both sexes.

^

But the idea that motherdove has such critical and lasting impact did

not begin to take hold in orthodox psychoanalytic circles until Anna

Freud began to move away from Freud’s father'Centered universe. Ah
though she never married or had children of her own, she had had many
maternal relationships, beginning with her nephew Ernst, whose mother,
Anna’s Sophie, had died, and continuing with the children of
sister

Dorothy Burlingham, her close friend and collaborator with whom she
lived (platonically by all accounts) for decades. She was also, of course, a

child analyst, her work with children much more focused on concerns
with mastering reality than on Melanie Klein’s arena, which was primi'
tive fantasy life. She came to see the mother as playing numerous
vital

roles in the child’s psychological development: the person whose rela'

tionship with the child forms the prototype for future relationships; an
auxiliary ego, who helps the child cope with difficult situations; and the

social legislator who inducts the child into the world of rules and stan-

dards (a modification of her father’s view that paternal authority forms

the basis of the superego). She believed that the loving attachment
to

the mother was ultimately more important to psychological development

than the instinctual repression her father had emphasized.
Anna Freud’s nursery work in Vienna and then London during the war
s emo-
further convinced her of the mother’s central role in the child
tional development. These little children much preferred
sleeping by

their mothers’ side in dismal air raid shelters

—
“a state ot bliss to which
92 What Do Children Need?

they all desired to return”^ ^

— to the creature comforts of the Hampstead
nursery. She discovered that children’s responses to the war crises

reflected their mother’s responses — calm if the mother was calm, dis^

traught if she was distraught. And it was plain to her that separation from
the mother, not the father, was traumatic for the children in her care, a
discovery that forced a new attitude among her analytic colleagues.
Anna Freud was a strong supporter of Bowlby’s and Robertson’s work
on separation. She urged that mothers be admitted with their hospitah

ized children because otherwise, battered by illness, strange circum'

stances, and medical procedures, a child’s ego resources — its ability to

cope with the pressures of reality — are quickly overwhelmed.

By the 1950s most analysts were at least informally agreed that the

relationship with mother is of critical psychological importance, that the

child’s first relationships are key foundations of his personality, and that
his tie to his mother is, by the end of the first year, very strong. Among
the classical Freudians, Erik Erikson was perhaps most eloquent on the
subject. He believed that the infant’s trust in the care he receives enables
him first to achieve the bodily regulation he needs — in feeding, sleeps

ing, eliminating. Later it enables him to let his mother out of sight

“without undue anxiety or rage.” In the end it “forms the basis in the
child for a sense of identity” and of being “all right.
But neither Erikson nor anyone else had a convincing theory to
explain why any of this should be so. Eour major theories were in vogue
at the time, the most popular continuing to be that attachment develops
because the mother satisfies the baby’s physiological needs, mainly for
nourishment and warmth. Others argued that the child has a built-in
need to suckle at the mother’s breast, and eventually it learns that there

is a mother attached to the breast, and so relates to her as well. A third

theory, advocated by the Hungarian analyst Imre Herman, presumed an

inborn need to cling to artother human being. The fourth was that
infants crave a return to the womb and that this craving naturally draws
them to seek mother and her embrace. (Bowlhy could barely conceal his

disdain for this concept: “It is difficult to imagine what survival value
such a desire might have, and 1 am not aware that any has been sug'
gested.”*^)

Empirical researchers, like Charlotte Biihler, the great Berlin psychoL

ogist with whom Spitz had trained, had long noted that babies in their

first weeks respond to the human voice and face in a totally unique
way.*"^ Klein had commented on the infant’s tendency to gaze at her
 Of Goslings and Babies 93

mother’s face, and become engaged, through facial expressions, in what

appeared to be “a loving conversation between mother and baby.”^^ But
none of this had worked its way into developmental thinking. Anna
Freud had stuck by her father, arguing that love begins in the stomach
and only gradually moves outward from there, and most analysts, even if

the evidence of their eyes told them otherwise, followed suit. Thus,

Budapest'trained analyst Therese Benedek, a brilliant and sensitive

developmental thinker, believed the infant needed to be smiled at,
picked up, talked to, etc.”; the American analyst Margaret Ribble, a

major voice against maternal deprivation, believed that infants had an

innate need for contact with the mother; Spitz believed in the impor^
tance of skin contact with the mother and decried its decline through
much of the Western world. But all three held that such needs were built
first upon feeding. This idea had become so entrenched, it was hard to
think in any other way.^^
Bowlby’s theory had something in common with the sucking and
clinging theories, but the views that were closest to his were expressed by
a few object relations and interpersonal pioneers. Fairbairn, who saw our
biological drives as aspects of our relational needs, once illustrated this

point with an anecdote from a psychoanalytic session in which his

patient cried, “You are always talking about my wanting this or that

desire satisfied; but what I really want is a father.”^

Michael Balint, a psychoanalyst of the Budapest school who later set'

tied in England and with his wife, Alice, believed that the wish to
he

loved, totally and unconditionally, is a primary need of human beings

from the time they are born. If they don’t get it, they spend the rest of

their lives searching for it. Balint saw this problem as rampant in civi-

lized countries because the close mothet'child tie is severed at too early

an age.^® Balint’s analyst, Sandor Ferenczi, had been rejected by Freud

toward the end of his career for trying to give patients the love they’d

lacked in their early years (going so far in his abandonment of analytic

neutrality as to hug patients and let them sit on his lap). Like Ferenczi,
Balint believed that the deprivation of parental love underlies much
neurotic phenomena, and he constantly saw patients trying to wrench
from their analysts what their parents had failed to give them.
And then there was Winnicott. Flis thinking was also close to
Bowlby’s, but their relationship was perhaps the trickiest. Winnicott
had

come to psychoanalysis from pediatrics, and his intuitive gift for working

with children was considered uncanny and inspirational by many. At the

94 What Do Children Need!

same time, his popular advice on child rearing made him the British

equivalent of Dr. Spock, with a cult following of adoring women. Win^

nicott, too, had taken strong some of them predating Bowlhy’s,
positions,

on both the centrality of the infant^mother bond and on the critical

importance of the quality of mothering. Like Bowlby and Fairbairn, he
believed that the baby had a builtdn need to form an emotional relation-
ship with the mother.^^ Unlike Klein, he was aware of the damage moth-
ers could do, having treated thousands of mothers and children at the
outpatient clinic at the Paddington Green Children’s Hospital. He saw
the way the mother held the baby, both in the physical and psychologi-
cal sense, as extremely important and had written that “the earliest anxi-
ety is related to being insecurely held.”^^ He and Bowlby had co-signed a
letter in 1939 warning about the psychological dangers of evacuating
young children from the cities during the war. Although strongly influ-
enced by Klein, whom he considered a genius — he counted himself a

Kleinian for many years and was apparently preoccupied with her ap-
provals^ — Winnicott could not accept some of her main concepts (the
idea of innate envy particularly infuriated him).
While his interest in Bowlby seemed minimal and his statements

about him (mainly in private) were mixed, their views were so similar
that years later Bowlby would repeatedly say that he felt that Winnicott

stated in poetic terms many of the same things that he, Bowlby, was try-
ing to express scientifically. But, while Bowlby concentrated on the envi-
ronment and on the scientific underpinnings of intimate relationships,
Winnicott focused on the child’s subjective experience. He was in a

sense the ground on which Bowlby and Klein met.

Winnicott’s attention to parental impact was also evident in his con-
cept of the “false self.” — grandiose, compulsively
A false self pleasing,

precociously mature and competent, — does not develop on

etc. its own.
It is a response, Winnicott believed, to the mother’s psychic intrusions.
Winnicott’s delicate sensitivity to the child’s needs, the intricacies of the
mother-infant relationship, the ways in which a child forms a sense of
self in the context of that relationship, and the numerous subtle avenues
via which this process could go awry not only went a long way, in
Charles Rycroft’s words, toward resurrecting “the traumatic theory of
neurosis, hut would eventually contribute to a reputation that today
places Winnicott higher than almost any other British analyst.
Although Winnicott was considered something of a maverick in the

1950s (both Freud and Klein considered him a lightweight and neither
 Of Goslings and Babies 95

would allow her students to study with him on a regular basis^^), he was

not a threatening one, partly because he was not inclined to build theory
at the expense of those to whom he felt some sense of loyalty. If Bowlby
showed an implacable masculine spirit and a willingness to let the chips

tall where they may in the single-minded pursuit of his idea, Winnicotts

approach was essentially more feminine and maternal,^”^ an inclination

he readily noted in himself and which no doubt contributed to his clini-
cal genius. As a result, his theoretical statements sometimes had a pecu-

liar, ohfuscatory quality about them, as he tried to maintain allegiances

and build bridges, a tendency that Bowlby, for one, found maddening.^®

It was a cardinal tenet of Kleinian thinking, for instance, that the baby’s
first relationship is with the breast. Winnicott appeared
to go along with

this and yet differ with it at the same time. “When it is said that the first

the breast,” he wrote, “the word ‘breast’ is used, I believe, to

object is

stand for the technique of mothering as well as for the actual flesh. It is

not impossible for a mother to be a good enough mother (in my way of

putting it) with a bottle for the actual feeding.”^^ This makes good sense,

but it is not what Klein had in mind.

By the late 1950s Bowlby was ready to present the main themes of his
new synthesis, and he did so in three papers addressed to his analytic col-

leagues, the first and most important of which was called “The Nature of
the Child’s Tie to His Mother.” If we look at things the way the etholo-
gists do, Bowlby said, there is no point any longer in thinking in terms of

psychological drives, or energies that accumulate and require discharge.

Freud, in keeping with the scientific thinking of his day, assumed
that in
self-preservation
order to explain the motivation behind a sexual or
behavior, one must postulate sexual and self-preservation drives.
Bowlby

said this way of thinking was now obsolete. You don’t

need an eating

drive to explain eating. You don’t need a seeing drive in order to explain

the existence of the eye or phenomenon of sight. The eye works the way
leading
itdoes because natural selection favored physiological variations
to improved vision. Mating and parenting behavior
work the way they

do because natural selection favored certain behavioral responses

in

these realms.
What’s more, the trigger for many instinctual behaviors dwells as fre-
initiated or
quently outside the organism as within. Thus eating can be
body; but
terminated only by certain chemical signals from within the
by the spread of another bird s tail, the
mating behavior may be elicited
96 What Do Children Need?

color of its beak, or the song it sings. Between parent and child, between
mating adults, certain actions naturally elicit certain responses.

The primary purpose of many ot the infant’s and young child’s instinC'

trial responses, in man as in other animals, is to ensure proximity to the

adult, which is necessary to survival. Sucking, clinging, following, crying,

and smiling — and perhaps cooing and babbling as well — are all instinc'

tual responses that eventually coalesce to form the broad mosaic of

attachment behavior. Initially unrelated to one another, in the course of

the first year of life they gradually become integrated and directed toward
a single purpose.

Crying and smiling do not actively attach the child to the parent;
rather they attach the parent to the child. Bowlby noted the abundant
evidence from animal studies that shows the mother responding immedi-
ately to the bleat, call, or cry of its young. (Bowlby observed that ah
though the human baby’s crying response is often terminated by food,
babies frequently cry when they are not hungry and will stop when they
are picked up, rocked, or spoken to.) The smile, Bowlby wrote, “has a
comparable though more agreeable effect.’’ Social smiling, which shows
up at around six weeks, is activated by the human face — or, in the early
days, by a simple gestalt composed of two dots — and it is a powerful
social releaser of maternal behavior, beguiling and enslaving the mother.
“Can we doubt that the more and better an infant smiles the better is he
loved and cared for?’’^^

(By advancing this idea, that parents have instinctual responses to

their young and that those responses are elicited by such things as smiles
and cries, Bowlby was, indirectly, advancing an explanation for why
nurses and other medical professionals could fail to hear or see the agony
of hospitalized bahies or deceive themselves into believing that the “set'
tied’’ child who no longer cried for his mother was not depressed or suT
fering. To use Bowlby ’s language, the infants’ cries activated the nurse’s

own attachment systems, making them feel compelled to offer maternal

care. But because they could not possibly mother all the babies in their
unit, they were placed in a biologically and psychologically untenable
position, as if constantly having to listen to alarms and sirens and not
being allowed to respond. Repression, rationalization, denial, and other
defenses were the only recourse, the only refuge from a searing heart.)
From the hahy’s point of view, Bowlby said, none of its instinctual
attachment responses is more primary than the other, and so it is miS'
 Of Goslings and Babies 97

taken to give preeminence to sucking or feeding, or to assume attach'

ment follows strictly from their satisfaction. Clinging, to take one exam'

pie, which is most primates, sometimes shows

essential to the survival of

extraordinary tenacity. In some species it is maintained both day

and
If an
night in the first weeks of life and clearly has no relation to feeding.
chimpanzee kept from clinging to its mother, it will cry plain'
infant is

The slightly older infant chimp will throw a temper tantrum. In

tively.

human hahies it is most pronounced when the child is afraid, when it is

about to be put to bed, or after a separation from the mother.
The very young infant relies mainly on responses that will evoke
maternal behavior — that is, crying and smiling. But as it grows older, it

uses clingingand following to assure proximity. Following with the eyes

fob
shows up by the third month. Later, as it is able to crawl, then walk,
lowing is activated if the child is tired, hungry, in pain, or afraid.

Following reaches its many would assume, when the child is

peak, not, as

most helpless and dependent, but when he is becoming more indepen'

dent, toward the end of the second year, because now he
is able to crawl

or walk into dangerous situations but not able yet to

fend for himself in
attach'
an emergency. ^X^ith data like this, Bowlby was able to argue that
ment is quite separate from dependency based on feeding and not caused
by it at all.

Bowlby saw sucking as another important element in the attachment

repertoire, contributing to the child s sense of felt security. He disap'
“In my
proved of the modern tendency to wean children at six months.
life need a
experience most infants through much of the second year of
great deal of sucking and thrive on milk from a bottle at bedtime. He
insisted, however, that breast'feeding is not the big deal its been made

out to be, nor are the early months of infancy so much

more important

than the later ones. “It is my impression that fully as many

psychological

disturbances, including the most severe, can date from

the second year

of life when clinging and following are at their

peak as from the early

months when they are rudimentary.

and fatigue, Bowlby added, will all cause a child
Anxiety, fear, illness,

to increase these attachment behaviors.

Bowlby argued that neurotic disturbances arise out of distortions of

the attachment system. The very sorts of maternal behaviors that cause a
mother— minor rejections, for instance,
child to feel hostile toward the
or short separations — may also he expected to intensify following and
98 What Do Children Need?

clinging; hence the neurotically dependent behavior of some children

and adults. When the mother figure is massively rejecting or totally
absent, attachment behaviors may fail to mature or become totally re-

pressed — the condition that Bowlby himself had seen in the affection-

less thieves and assumed to be at work in adult psychopaths.

The attachment system does not exist for the purpose of creating love

or for healthy emotional functioning in later life. Each of the instinctual

responses Bowlby describes is present because of its survival value.
Without them the child would die, especially the child that was born on
the primitive savannahs where people first evolved. None of these
survival behaviors is directed at the mother per se. In the early months
individuals are vague presences. He argued, however, that the mother be-
comes the key player not because the baby is one day able to grasp that
there is a person attached to the breast, but because she responds to and
becomes the object of all the various attachment behaviors.
Before the baby is able to grasp the idea of a person, its suckling may
relate only to the breast or the bottle, the smiling to any human face,
and it is enough that the other attachment behaviors get responded to,
regardless of where the response comes from. But gradually, as mother
comes into view as an individual, their mutual history of dovetailed
responses coalesces into the single phenomenon of love. The once sepa-
rate, unrelated, instinctual patterns are now elements of a more complex
system, attachment behavior, with a single direction and single purpose.
And now there is a person to whom the baby feels attached and toward
whom he consciously directs these behaviors. By a certain age, differing
broadly among babies but probably six or seven months at the latest,

“good mothering from any kind woman ceases to satisfy him — only his
mother will do.”^^ Subsequent studies of adoption in the first year tended
to bear Bowlby out on this point, with children under six months of age
tending to adjust to a new mother with much less distress.

Bowlby notes that throughout the animal world, species-specific

behaviors, whether on the part of the child, the parent, or the mate, tend

which remains a tenet of attachment thinking to this day

*1 find this idea, —
that “it is only during
the second 6 months that these proximity and interaction promoting behaviors are integrated into a
coherent system, organized around a particular figure or figures” (Bretherton, 1985, p. 7) difficult —
to accept. My own son showed a clear preference for my wife and me from earliest infancy, and
made hy students at the Tavistock Clinic in recent years make the same finding
infant observations
Hopkins, personal communication, February 5, 1993.) See also Mary Ainsworth’s study of
(Juliet
Canda children in Chapter 10. However, as the adoption studies mentioned above indicate, the
strength and quality of the infant’s attachment does seem to go through some sort of important
change in the latter part of the first year, at least in many infants.
 99
Of Goslings and Babies

to he focused on one individual (or in the case of the parent, one group
fact as so impot'
of individuals, the children). Bowlhy sees this neglected
will live on in furious psy'
tant, he gives it a name, ^^monotropy, which

chological dehates for the next three decades.

helieve to
By monotropy Bowlhy did not mean, as many of his critics
this day, that the child becomes attached to only
one figure. Rather that

there is* a hierarchy, with one figure, the chief caregiver, usually the

most people live or seek to live

mother, at the head. Even in adult life,

hy this hierarchical design, choosing one person who is their central

one person who is loved above all others and whose

attachment figure,

presence most insures a feeling of security. When we are at our lowest,

due to illness or emotional distress, this is the person we most want

nearby.
In childhood the primary caregiver is crucial to healthy development.
are damaging in
He believes that maternal deprivation and separation
early years largely because the child’s precious
central figure has been
the
removed, causing his attachment life and all its emotional and develop-
Bowlby argued that, given this
mental derivatives to become disrupted.
natural predisposition, threats to abandon a child who misbehaves,
all of
threats to send him away, or threats by a mother to kill herself
which he found to be distressingly common — are terrifying to the child,
child or adolescent
generating unbearable levels of anxiety. In the older
the personality as to
they also arouse anger, which can so dominate
become Years later, another researcher reported the case
dysfunctional.
couldn’t stand to
of an adolescent who, after killing his mother,
cried, “I

this a fitting illustration of his

have her leave me.”^^ Bowlby thought
theory.
true mourning, another
Bowlby believed that babies were capable of
despite the evidence of
controversial point with his analytic colleagues,
Heinicke,
Anna Freud’s wartime nurseries and a 1956 study by Christoph
team. In 1958 Peter
an American psychologist who worked on Bowlby ’s
Marris published Widows and Their Families. Bowlby
saw that the widow’s
similar to what had been
reactions to loss that Marris described were
observed in infants. The widow’s pattern typically included rage, which
might be directed almost anywhere — at a doctor, at herself, at the lost

The rage, Marris believed, while it lasted, gave the woman

husband.
sink into depression.^^ Here, in grown
courage. Only afterward did she
observed in small
adults, was the protest and despair that Robertson had
Bowlby that babies mourned like adults when they lost
children, proof to
100 What Do Children Need?

a precious other and that this mourning of the loss of an attachment fig-

ure remained critical throughout life.

In healthy mourning, the individual gradually reorganizes. He allows

himself to separate from the lost person and begins to reach out to form
new attachments. Pathological mourning, in contrast, either never lets

up or is disavowed and continues to control the person unconsciously.

For the small child who loses his first attachment figure and is not given
a new one in time, like the institutionalized children Spitz filmed or the

affectionless thieves Bowlby treated, pathological mourning is the only

possible course, with results that distort the development of the child’s

personality.

The detachment that Robertson had observed in young children

who’d suffered long hospitalizations and that Bowlby saw as the root of

affectionless psychopathy was one type of pathological mourning. It was

dangerous because it sealed off the personality not only from despair but
from love and other experiences that could disconfirm his feelings of
worthlessness, guilt, and bitter mistrust, thus precluding any working
through of the grief. In less severe cases, a prolonged and depriving sepa^
ration in infancy will not lead to such profound character disorder. But
the pathological mourning process will show up when the child faces
separations or losses later in life and, as many classical analytic case stud'

ies had shown, predispose him not only to separation anxiety but to
depressive episodes as well.^^
Bowlby now made an effort to interpret the sequence of reactions to
separation discovered by Robertson and to integrate them into his deveh
oping theory. If protest, despair, and detachment are the three primary
responses of the young child to separation from his mother, then we can
see in these reactions, he says, the basis for key emotional processes that
govern our psychology. Protest is an embodiment of separation anxiety,
despair is an indication of mourning, detachment is a form of defense.
In addition, Bowlhy argued that separation anxiety, which had long
been a subject of debate in analysis — and which had been attributed to
all sorts of psychic processes, from Otto Rank’s idea that it was a recapit'
ulation of the trauma of birth to Melanie Klein’s proposition that the
child assumed that the absent mother had somehow been eaten up or
destroyed by the child itself — was a natural, evolutionary'based child'
hood reaction. It was an intrinsic fear, like fear of the dark, loud noises,

isolation, and strange things and people. Separation anxiety, like

 101
Of Goslings and Babies

stranger anxiety, however, only becomes pronounced for many children

in the second half of the first year. Bowlby explained this by noting that

as the infant’s attachment system becomes more organized around a par^

ticular mothering figure, so does his tendency to withdraw from or
become upset by strange situations. He believed that separation anxi'

ety was particularly problematic for a child because in many cases attach-

ment behavior was activated, but because the attachment figure was not

present, it could not be shut off. He also saw such attachment frustration
as the source of the hostility he had seen in so many young children who
had either been separated from or neglected by their mothers.

Just as Freud, beginning with hysteria, outlined an entire theory of

human psychology and psychopathology, Bowlby, building from his work
on separation, was in the process of doing something similar.
Unlike
stand on.
Freud, however, he had both Freud s and Darwin s shoulders to
but
Bowlby’s theory was bold, full of common sense, speculative
grounded in science, and threatening to the older views. His papers were
course,
loaded with calls for research. Bowlby was unusual in this, as, of
he knew, and he must have also known that such calls, aside from
repre-

noses of his
senting his true belief in what was needed, also tweaked the
analytic colleagues. He was almost daring them: Here’s what
I believe;

there’s what you believe; let’s see which position is home out by the data.

He could afford to state modestly that much of what he was saying was

speculative and needed to be empirically confirmed; he was

confident it

would be. Meanwhile, in advocating research, he was also saying, in

game with which 1 am
effect, “Let’s play a quite familiar and which you

have never bothered to learn.” Few psychoanalysts were also researchers,

indeed it was quite rare and researchers to speak the same

for clinicians

language let alone be united in one person. Meanwhile, research

on early
separation undertaken by Robertson, Heinicke, and Rudolph
Schaffer,

support to
former and current members of his team, were already lending
his theories.
Throughout his papers, Bowlby made every effort to align himself with

Freud and to offer citations from Freud to establish that he

was plowing

fields that Freud had sanctioned, even if Freud

had never plowed them
himself. He makes complimentary statements about the contributions of

Freud and Dorothy Burlingham and quotes from them

approvingly
Anna
makes
where he feels in tune with their findings. On the other hand, he
no hones about declaring certain views obsolete: Dependency is irrele-
102 What Do Children Need?

vant to attachment and attachment behaviors. Mother and breast should

no longer be equated nor should good mother and good feeding. It is

wrong to speak of the earliest phase as oral or the relationship to the

mother at that time as anaclitic, a popular term in psychoanalysis, coined
by Spitz, and meaning simply that babydove is built on the baby depen^
dency. With these brief statements at the end of “The Child’s Tie,’’ he
attacked sensitive elements of analytic orthodoxy, be it Freudian or
Kleinian. There was something for everyone to be upset about.
 8
"What’s the Use
TO Psychoanalyze a Goose?"
Turmoil, Hostility, and Debate

“All of us find security in being with people we know well and are apt to

feel anxious and insecure in a crowd of strangers. Particularly in times of

crisis or distress do we seek our closest friends and The need for
relatives.

companionship and the comfort it brings is a very deep need in human

nature — and one we share with many of the higher animals.’’^

Thus began John Bowlby’s first (and, as it would turn out, only) offi'

dal advice to parents. A small pamphlet, called Can 1 Leave My BabyL

it

was published in 1958 by the National Association of Mental Health.

He continued:

This need for companionship is even stronger in young crea-

tures than in grown-ups. Whether it is a brood of ducklings on

a pond, twin lambs in a meadow, or a human toddler around

the house, the young are quickly distressed if they get lost and

scamper to get close to their mothers as soon as anything hap-

pens which frightens them.^

In his new pamphlet, Bowlby continued his assault on the stift'upper-

under'
lip philosophy by which he had been reared and helped parents
104 What Do Children Need?

stand the implications of his and Robertson’s work on early separation.

He also wanted to offer some solace to mothers who were confused by

the abundance of conflicting advice or who, because of their exposure to
Bowlhy’s or Robertson’s previous work, as filtered through the popular
press, dreaded leaving their babies for an evening at the theater.

Bowlby, who by now had spent many years running a mother’s group

at a welhbaby clinic from which he’d learned a great deal about mothers’
experiences,^ sympathized with parents regarding the relentless demands
of young children, which seemed to grow worse rather than abate as the

infant developed into a toddler. “If they are frightened and upset, they

cling to their parents like leeches,’’ he said. “All parents find this irksome
and at times wish to be free of it. Some become alarmed lest it go on for^

ever.’’"^ But he insisted that this process was natural and that if it was
accepted, the demands would decrease and a secure child would emerge.
“When you have a baby,” he said, quoting an old adage, “you have five
years of hard labor ahead of you. If you don’t get it over at the beginning,
you’ve got it coming to you later.”^

Bowlby was not in the least opposed to a livehn nanny, as long as she
was a good one, as long as it was recognized that she may become the
true mother figure in the child’s eyes, and — this he emphasized — as long
as she stayed. For by the time the child reaches six months of age, his pri'

mary caretaker becomes a crucial and irreplaceable person who must be

there fairly continuously for the next two and a half years. In a statement
that seems particularly poignant in light of his own childhood experi'
ence, he warned, “For a child to be looked after entirely by a loving
nanny and then for her to leave when he is two or three, or even four or
five, can be almost as tragic as the loss of a mother.”^
Bowlby favored separation in small doses. “It is an excellent plan,” he
said, “to accustom babies and small children to being cared for now and
then by someone else.”^ It will prepare the child for future emergency
separations and give the mother some breathing space. He told mothers
that they should not worry about being away from their children for an
evening or even, occasionally, for a long weekend. But until the age of
three, longer separations are perilous and will usually have lasting ramifi'

cations in the form of insecure behavior. He thus opposed hospitalizing

young children, advised seeking outpatient alternatives wherever possi-

ble, rooming with one’s child if hospitalization is unavoidable, and mak-

ing a fuss if this is resisted.
 Turmoil, Hostility, and Debate 105

If the mother must be hospitalized, he urged that the child be taken

care of hy a familiar person, ideally the father or a favorite aunt or
granny, but always someone who can serve as a fulhtime caretaker. But,
upon return home, Bowlby cautioned, “Don’t expect a cheerful welcome
and a warm hug. Expect instead a rather bemused little person who keeps
his distance and hardly knows what to make of things. He may even run

away from you or claim that you are not his mother at all. ^ When the

initial rejection is followed hy night waking, clinginess, temper tantrums

over the slightest separation, don’t blame grandma for spoiling the baby
while mother was away, Bowlby said. And don’t try to correct such
annoying behavior with firmness. The only corrective is “a lot of love

and reassurance.”^
Much of this was immensely controversial at the time, when the insti-

tutionalization,wartime evacuation, and sending of young children to

boarding schools were consecrated practices. But, unique in the history
of psychoanalysis, it had a rich body of research supporting it. In the
vastness of psychoanalytic thinking, Bowlby’s thesis might be considered
small; but it was strong. It would remain controversial as increasing num^
hers of women entered the work force.
But was in his discussion of parents and their roles that Bowlby may
it

have seeded the ground for an even greater furor. In Bowlby’s view,
advanced for 1958 but relatively unchanged in later years, the fathers

role is mainly as a support for mother. The father becomes increasingly

important to children as they get older, but because he is away so much
during the infant’s and toddler’s waking day, he rates a poor second when
they awaken in the middle of the night and seek reassurance. He urged
fathers not to be too upset by that and to recognize that their role, keep'

ing their wives happy, is a part of child care, too. To Bowlby, a nonstop

worker himself, whose work was his life, and whose rare displays of teixi'

per were occasioned by the intrusions of his children,^^ it perhaps seemed

inconceivable that a father could be more intimately involved, so that

his presence, too, would be a source of security.

The mother, on the other hand, has a tougher and more exacting job,
“perhaps the most skilled in the world.”^^ She may he exhausted, over^
whelmed, and short-tempered, hut she gets great satisfaction from know-

ing how and irreplaceable she is. Meanwhile, she must never
crucial

forget that the job cannot be skimped without lasting damage to the
child. “One cannot ever really give back to a child the love and atten-
106 What Do Children Need?

tion he needed and did not receive when he was small.With understand-
ing and affection, and perhaps skilled help, one can go a long way
towards it, sometimes a very long way, but it will never be quite the
”12
same.
Such pronouncements set teeth on edge. Among psychologists. Bowl-

by’s insistence that what was missed early on could never be tully re-

placed was controversial and remains so to this day. There is evidence

that sensitive periods do exist for human children, when, for instance,

language skills must be acquired, and the same may be true for the ability

tobecome meaningfully attached and thus able to give and receive love.
But we are much more flexible in this respect than lower animals and
much better able to make up our losses, a fact that Bowlhy would come
to recognize.
Some women, meanwhile, believed that Bowlhy was determined to
chain mothers to the home, and in this they were partially correct. He
fully welcomed women into the professions — as the many women who
worked for and with him have attested — hut he expected them to face

up to their maternal responsibilities when and if they had children and

take a leave of absence. He did believe in a gradual return to work for

those who wanted it, beginning on a part-time basis, after the child

reached the age ot three; and, as an employer, he went out of his way to
make such arrangements possible. But such fine points were hardly
acknowledged in the storm that soon swirled around him.
Bowlhy ’s advice to parents and his recommendations regarding social

policies affecting families and young children were less controversial

among analysts, even seen as progressive. Indeed, he was having more
positive impact on social policy and child care practices in England and
around the world than any other analyst in history. For analysts it was
the underlying theory that irritated them.

Both Melanie Klein and Anna Freud, now the rival doyennes of British
psychoanalysis, found the analytic-ethological concoction Bowlhy was
brewing, with his constant references to birds and beasts, distasteful, and
they let their followers know it. Analytic critics charged him, among
many other things, with gross siijiplification of theory, with assuming
that all pathology resulted from disturbances of the mother-infant bond
(when it was well known that early medical and other environmental
traumas could equally he at fault), and with overlooking the infant’s ahil-
 Turmoil, Hostility, and Debate 107

ity todevelop a negative concept of the mother on wholly irrational

grounds —
such as when she fails to relieve his suffering despite her best
efforts, or when the arrival of a new sibling brings forth intolerable feeh

ings of abandonment, rage, and guilt.

“An infant can’t follow its mother; it isn’t a duckling,” said the

Kleinian analyst Susanna Isaacs. “The human baby is very helpless for a

prolonged period of time. And during this period it forms an internal

image of the mother; this would be a mixture of memory, of fantasies, of

the baby’s response to the actual reality of the mother.”^

To Bowlby was the cornerstone upon which fantasy was built. If
reality

a fantasy becomes terribly distorted —

as he believed it did with his affeC'

tionless thieves, so that only a hateful picture of the mother is retained,

regardless, in some cases, of how unfair it may be — that internal picture

does not come about for nothing. It is based on an experience — in the

case of the young thieves, early separation, hostile treatment, neglect,

abuse, or some combination of the four. No matter how distorted intan-
tile fantasies may be, negative internal images of mother will not grow

out of an essentially positive, secure experience. But in the Kleinian

view, fantasy did not depend on such links to reality. Using familiar
Kleinian imagery, Isaacs argued that “babies are very capable ot perpetU'
ating a distorted view of the mother, of introjecting a ‘bad mother’
where

none exists. It’s their helplessness; helplessness and fantasy go together.

And even the most perfect mother will frustrate the baby once in a
while, keeping him waiting. This makes the baby angry, and the emO'
tions of infants are terribly intense — its so difficult for them to assess

what’s happening to them.” Whereas the distortions of some babies tend

toward the negative, other babies distort in a more positive direction,
enabling them, according to Isaacs, to “introject a good internal picture
—
when they haven’t had a mother at all just different people looking
after them.”^"^ Such thinking, needless to say, was anathema to Bowlby.
The Kleinians, like the Freudians, protested that Bowlby ignored
intrapsychic processes — and that it is just these processes that distiiv

guish people from fish, birds, and rodents. As the leading Kleinian
dogs,

theorist of her day, Hanna Segal, recently said (quoting from Ogden
Nash): “What’s the use to psychoanalyze a goose?”^^
tact that the
The debate over such issues was very bitter, despite the
participants were largely in the same camp, all of them psychoanalysts
who accepted basic analytic principles, with Bowlby and Anna Freud
108 What Do Children Need!

in the forefront of the fight against maternal deprivation and Bowlhy

openly indebted to Klein for her views on early relatedness, fantasy,
mourning, and depression.
But the debate was not taking place under normal conditions. The
forties and fifties were an adventurous, incestuous, and, as some believe,

a shameful time in British psychoanalysis.^^ The Anna Freudian and

Kleinian camps were at such dagger points that it seemed they would
split the Society apart, and, indeed, they would have if a “gentlemen’s

agreement’’ had not been worked out after the war that enabled the two
groups to coexist. If Freud was a less combative personality than Klein,
she was no less rigid in matters of theory and was, besides, a somewhat
cool and aloof personality, ever conscious of the mantle of royal legiti'

macy she’d inherited. Intricate issues of loyalty and betrayal surrounded

them both, and both were dogmatic, almost theocratic, in their views.
(“Anna Freud worshipped at the shrine of St. Sigmund, Klein at the
shrine of St. Melanie,” Bowlby said.^^) “These were not people you could
disagree with,” says Juliet Hopkins of the Tavistock staff. “If you dis'

agreed, you were wrong.”

Anna Freud had been analyzed by her father, a practice that is not
only considered grossly unethical today but violates commonsense stric'

tures against boundary violation and incestuous intrusion. That she was
never able to form a significant attachment to another man or a sexual

connection with anyone, male or female, suggests the degree to which

she remained stuck in her father’s emotional orbit. As if not to be out-
done by the founder, Klein had, ir\ turn, analyzed her own children with
horrendous results. Winnicott, like Bowlby, had been analyzed by Joan
Riviere and supervised by Klein. Klein asked him to analyze her son,

which he did (from 1935 to 1939), while strenuously resisting her efforts

to supervise the case. (She did, however, succeed in supervising the

analysis of her grandson by Marion Milner.) Later, as Winnicott and

Klein delicately dueled over theory and as Klein continued to reject him
(even at one point refusing to allow his paper to be included in a special

double issue of the International Journal of Psycho-Analysis in honor of her

seventieth birthday because he refused to revise it in such a way as to
more fully incorporate her ideas), his wife, Clare, also an analyst, went
into treatment with Klein, which could only have been loaded with
unconscious meanings and motives for all three of them.^*^ Paula Hei-
mann, initially a devoted follower of Klein’s, found it necessary to pub-
licly dissolve their connection after disagreeing with her mentor on a
 Turmoil, Hostility, and Debate 109

matter of theory.^^ Scientific meetings of the Society were often charac'

terized by grotesque bickering and attacks, with the younger candidates,
who were desperate for patient referrals from their superiors, often smirk'
ing to one another as someone from the opposing camp presented his

views. Meanwhile, Anna Freud and Melanie Klein sat in the front row
“with barricades of people between them.”^^ In 1962 Charles Rycroft, a
brilliantand sensible analyst from the independent group, became so fed
up that he stopped attending meetings of the Society and a decade later
quit altogether.^"^
If it was a shameful time, it was also in many ways a fertile time, with
British analysis experiencing a ferment that would not hit the States for

another decade. Major questions were being asked not only about theory,
but practice, and the fact that the Society had not split meant that all

factions had to listen to some way with opposing views.

and contend in

Winnicott was extending the treatment hour for some patients beyond
the standard fifty minutes, and, like some of his peers, was providing a
more supportive, maternal stance for the patients who seemed to need it.
The once monolithic analytic neutrality, which in practice often felt

cold and distant, was now being questioned.

Meanwhile, analysts were seeing more and more of the type of
patients mentioned earlier who did not fit the standard concept of neu'
rosis. Instead of suffering from internal conflicts and being haunted hy
gyilt — but otherwise having a pretty clear sense of who they were — they
had trouble in relationships, often felt empty, detached, or worthless.

Such patients, once considered inappropriate for analysis, were now

being treated en masse, and the sources of their disturbances were invari'
ably traced not to oedipal conflicts hut to early, preoedipal, experiences,
often related to faulty parenting.
It was into this agitated context that Bowlby — a Kleinian apostate

stepped with his new pronouncements.

It isunderstandable that Bowlby ’s views would not have been popular
with his peers. It was not just his rejection of certain sacred psychoana-
lytic axioms. His ideas about ethology and prevention, because
they

tended to bypass intrapsychic processes, seemed vaguely unanalytic if not

downright anti^analytic. To some it seemed, as Isaacs and Segal implied,
that he was reducing a human child to a goose, with predictable, preset

responses,and ignoring not only the richness and diversity of individual

psychology hut the ways in which an irrational inner world ot guilt, fear,
shame, identification, projection, and paranoia can serve as the distort'
no What Do Children Need?

ing lens through which both the child and the adult perceive and experi-

ence reality.

Bowlhy was a nuts-and-holts man. He liked hard data and research-

able concepts. He was proud that his forty-four juvenile thieves paper

was the first psychoanalytic article to include statistics. He was unlike

almost any other analyst in that respect. Although he was stirred by

poetry and quoted it in his books, he saw himself as a scientist, one who
would help (perhaps force) psychoanalysis to live up to its scientific pre-

tensions. He preferred to work in the daylight where facts were facts and
a man could feel in command of himself and his destiny; he was more
wary of the darkness, with its shadowy ambiguities, where faulty and
nonsensical notions, products of an overly active imagination, could gain
credence and where disturbing feelings could not be easily warded off.
Even some of those who saw value in his work (like the American ana-
lyst Otto Kernberg^^) would fault him for failing to build a better bridge

to thinkers like Klein, for whom the shadows were the main thing. But

that was not where Bowlhy wanted to he.

Evolutionary ideas spoke powerfully to Bowlhy about the emotional

needs of children. It was as if he had found a key he had been searching
for all of his intellectual life, a key that would help him build a theory

that would serve to avenge the wrongs done to him as a child by making
it intellectually untenable for those wrongs to persist. But he went
beyond that, demonstrating the necessity of intimate attachments
throughout the life cycle and the critical importance to every human
being of separation, loss, and mourning. In a vast work of reinterpreta-

tion (and stunning mental exertion), he plowed through the entire ana-
lytic canon demonstrating how early losses — especially the mother or
father’s dying when the patient was a young child — had been overlooked
in one case study after another, resulting in fallacious assumptions and
theory building. Although he was rarely acknowledged or cited at the
time, his work on loss sparked in the 1960s the first flowering of analytic
interest in this subject.

In his 1980 hook Loss, the third volume of the attachment trilogy that

he would come to write, Bowlhy summarized the conditions that he con-

sidered essential prerequisites for healthy mourning of a lost parent by a
child or adolescent:

[F|irst, that he should have enjoyed a reasonably secure rela-

tionship with his parents prior to the loss; secondly that ... he
 Turmoil, Hostility, and Debate III

be given prompt and accurate information about what has

happened, he allowed to ask all sorts of questions and have

them answered as honestly as possible, and be a participant in

family grieving including whatever funeral rites are decided

on; and, thirdly, that he has the comforting presence of his
surviving parent, or if that is not possible of a known and
trusted substitute, and an assurance that that relationship will

continue.

Bowlhy was able to demonstrate convincingly the successful coping of

those rare children who had these conditions: They could be sad, they

could talk about it, they could have their pain, perhaps for a long time,
without having to dissociate from it and become prey to damaging
unconscious processes. He was also able to demonstrate the disturbing
consequences — the primitive defenses, the failure of subsequent deveh
opment, the formation of phobias and other hysterical symptoms, the
depression, the extreme mental disorders, the antisocial behavior, the
irresolvable anger toward the living parent, the inability to form new
relationships — for the many who did not have these protective condi-

tions. Bowlhy was a master at describing the environmental prerequisites

to healthy emotional development and the psychological damage that

followed their absence. But when it came to what went on in the psyche
of the child, the path of the child’s hatreds and aggressions and how they

might get psychologically fixed in fantasy, or the torturous course of the

child’s identifications with his parents, living or dead, he was less cre^

atively involved and ultimately less interesting. Even in the realm of

early loss, it would he left to later analysts,^® who built on his work, to

describe in detail the complex fantasies and disturbing identifications

that bereaved children are burdened with and how they can be helped

with therapeutic attention.

At times Bowlhy veered toward the utopian. One could almost assume
by reading him that when the right conditions prevailed, mankind could
finally master the bedeviling qualities that have always made the human

animal such a problem to himself.

Bowlhy did show interest in intrapsychic processes, more than his

detractors give him credit for. His work on what he called “defensive
exclusion” explored aspects of repression and dissociation. And he cer^

tainly put a great deal of effort into demonstrating how the child builds

up an “internal working model” of self and other based on his experi-

112 What Do Children Need!

ences with the intimate people in his life (see Chapter 15). But if the

inner world of Bowlhy’s child, buttressed as it was by research with lower

animals, was more scientifically substantiated, it was less complex, less

rich, less tormented and passionate than that conceived by some of his
peers, especially the Kleinians, who would gain ground in England in the

coming decades for the very reason that the seething they saw in the
child — greed, aggression, envy, sexual possessiveness — was so helpful in

understanding the seething that analysts saw in themselves and their

patients.

But, again, that was not Bowlby’s terrain. He was averse to any theory

that could not be verified empirically, and beyond that 1 think he deeply
disliked viewing the young child as Klein did, churning with wild fan-
tasiesand naturally inclined toward hatred and violence, as well as love
and the wish for reparation. In the end, he narrowed his focus to the
things that mattered most to him and did great work in that realm.
Largely because of what he bravely rejected — drive theory, the death

instinct— and because of what he

partly quietly, perhaps avoidantly,
omitted — deeper look
a fantasy — few of
at life his colleagues could see

that what mattered to Bowlby mattered to them, indeed that he was

onto something of huge consequence. Sadly, to this day senior people in
the field can he found who believe that Bowlby was attacking psycho^
analysis and that his goal was to destroy it.^^

Bowlby always remained committed to psychoanalysis as a discipline

that was concerned with certain questions — the role of the unconscious
in emotional life, the meaning of dreams and fantasy, the psychosocial
development of the child, the source of neurotic distortions. He felt that

Freud had asked many of the most important questions. But he argued

that every other academic discipline is defined by its realm of interest,

not by a particular theory that must be adhered to.^^ Other analysts had
struggled mightily with this problem, often twisting and contorting
themselves in order to adapt new ideas to an overarching theory that
often seemed at odds with the facts. Bowlhy’s ideas would not adapt, or
perhaps more accurately, he was not the type to make a strenuous and
circuitous adaptation when he saw a simpler, more direct route before

him. This hluntness proved threatening to his colleagues.

«

Sometimes in science an overarching theory is recognized to he an “as

if’’ proposition, a metaphor, or useful fiction. It lasts until fresh knowL
edge makes retaining the old metaphor more trouble than junking it and
adjusting to something new. Those who devised the view ot the universe
 Turmoil, Hostility, and Debate 113

with the earth at its center originally saw it as just this sort of useful fie-

tion. It was a starting point for ordering one’s ideas and building a coS'

mology.^^ But eventually it became a religious tenet, and those who

opposed it were burned at the stake.

In psychoanalysis an allegiance to theory can represent an especially

complex set of emotional commitments. As in other fields, it means an
allegiance to things one has said in the past, to the way of thinking to

which one has become accustomed, and to certain people with whom
one has taken sides, including one’s mentors. It may in addition entail a

tie to one’s own analyst whom one may love or venerate. It has often

meant an allegiance to Freud, the worshipped founder, to whom one per-

haps paid obeisance in all one’s writings. And, most important, adher-
ence to a particular theory also represents a loyalty to oneself and to the

way one has worked with patients. To threaten an analyst’s overarching

theory — or “metapsychology” — therefore, is sometimes tantamount to

threatening the very meaning of his work and his life.

Bowlby knew Anna Freud quite well and they were always on cordial

terms. They were linked by a common concern for the treatment of small
children, and Robertson, her loyal prote;aage;aa, worked for Bowlby. But

she nevertheless considered herself the guardian of the flame, and she
disliked heresy. Freud didn’t go into outright attack when Bowlby eluci'

dated his new theory; that was not her style. Besides, she believed “Dr.
Bowlby is too valuable a person to get lost to psychoanalysis.”^^ But “to
say she cold-shouldered it,” Bowlby recalled, “would be an understate-
ment. She banned it.”^^

The key Freudian assault came after Bowlby’s third paper, the one on
mourning. The paper was published in Psychoanalytic Study of the Child, a
highly respected annual, edited by the top people in the classical camp,
including Freud herself. Without informing Bowlby or giving him a

chance to reply, his paper was followed by three rejoinders — by Anna

Freud, Max Schur, the analyst who had also been Freud’s physician, and
Rene Spitz, all of whom essentially dismissed Bowlby by arguing that
what was valid in his thinking was not new and what was new was not
valid.

On the whole, Freud’s response was essentially defensive, using the

royal “we” to enunciate where Bowlby was correct and where he had
gone She argued that very young children were not capable of true
astray.

mourning, that Bowlby was too focused on environmental events in

themselves (“as analysts ... we do not deal with happenings in the exter-
114 What Do Children Need?

nal world as such but with their repercussions in the mind”^"^), that he
misused and misunderstood analytic terms, and that he had naively diS'
counted the pleasure principle. Strangely, she enunciated so many areas
of agreement that the uninitiated might miss the true purpose of her
exercise. Spitz, who saw himself as something of a rival of Bowlby pre^
cisely because of the similarity of their views, was somewhat scolding
about Bowlby’s lapses — including his failure to consider the develop^

mental stages young children go through and his misunderstanding of

Spitz’s own work. Schur was more openly disdainful, especially of

Bowlby’s failure to appreciate the infant’s “orality.” Whatever validity

their objections may have had, in the annual’s fifteen years of publican
tion no replies like this had appeared before and none have appeared in

the thirty'odd years since. Clearly Bowlby represented a special sort of

threat.

The hostility that the Kleinians directed at Bowlby’s ideas was more
severe and not made any easier for him by the fact that they had become
the dominant influence at the Tavistock Clinic where Bowlby was still

officially in charge. Trainees at the clinic, who were supervised by both

Bowlby and by a Kleinian analyst, sometimes felt, as one put it, as if they

were sitting “between two warring parents.

Klein’s followers exhibited neither Anna Freud’s frozen cordiality nor

the gentlemanly restraint that Bowlby always maintained despite his

obvious and sometimes disagreeable selBcertainty. They were particularly
hard on Bowlby when he read his papers at the scientific meetings of the
Society. In 1957, a few days before he presented it, Bowlby gave a copy of

“The Nature of the Child’s Tie to His Mother,” his key paper on ethology
and attachment formation, to a colleague whom he had previously con^
sidered a friend. The colleague turned out to be one of Klein’s followers,
and apparently distributed the paper among them.^^ Their vociferous
opposition when Bowlby presented it struck him as having been orga-
nized in advance, with each speaker arriving at the platform with a dif-
ferent element of his theory to demolish. “Almost all the principal
“1
people around Klein at that time had a bash,” Bowlby later recalled.
don’t think she spoke, hut it was very much in character for her to get

her disciples to do her dirty work for her.”^^ Bowlby’s colleague and pre-
sumed friend was among those who joined in the attack.
Sadly, if not surprisingly, Joan Riviere, Bowlby’s former analyst, was
also among those who protested, and Winnicott, who was then president
of the Society, wrote to thank her, expressing concern that to accept
 Turmoil, Hostility, and Debate 115

Bowlhy would mean to reject much of what Freud had fought for.^^ He
also wrote to Anna Freud reporting that although he personally came out
well in Bowlhy ’s new papers, they left him with a feeling of “revulsion”

nonetheless."^^ A year later, when Bowlhy read his second paper, “Sep^

aration Anxiety,”"^ ^ the response was even more heated, and two extra
meetings had to be held to complete the discussion.
Bowlhy, of course, had his defenders, including David Malan, then a
young analyst, who was so upset by the hostile reactions expressed at the
“1
meetings that he rose to Bowlby’s defense. Afterward, he recalled,
went away somewhat terrified about the consequences of being so pas-
sionate a supporter of John’s position — would 1 be excommunicated at

once?” (Winnicott wrote to thank him, too!)"^^

Bowlby turned fifty in 1957. In certain respects, he was a force to be

reckoned with. He was not only prominent because of his work on
maternal deprivation, which had become an international cause celebre
in social work, child psychiatry, hospital pediatric units, adoption agen^

cies, and ordinary homes, but his analytic papers were powerful, persua-
sive, clearly grounded in a thorough knowledge of psychoanalytic theory,

and, unusual for analysis, based on scientific studies, most of them, of

course, from outside the analytic realm.
Through all this, Bowlby had been active in the British Psycho^

Analytic Society, and he was on good terms with many of his critics,

having kept his hurt feelings to himself. He’d been training secretary
from 1944 to 1947, and was deputy president under Winnicott from 1957
to 1961. He had helped keep the Society together after the war when the
split between the followers of Anna Freud and Melanie Klein threatened
to tear the organization apart. It was at that time that a middle, or

independent, group formed, composed of analysts aligned with neither

Freud nor Klein, many of whom, as Bowlby liked to joke, still considered
the center of the universe to be London, not Vienna. “Although in the
natural course of things,” he later said, “1 might well have been elected
the next president, it had become clear that there would be too much
opposition and, with my agreement, those supporting me did not propose

me.”"^"^ Indeed, after the hellish response to his attachment papers,

Bowlby stopped attending meetings of the society. Unread, uncited, and
unseen, he became the nonperson of psychoanalysis and was lost to his
peers for the better part of the next three decades.
By the late 1950s Bowlby was having trouble on other fronts as well.
116 What Do Children Need:'

Misunderstanding of his work was so widespread that mothers were

warned never to leave their young children, and social workers and chil-
dren’s officerswere keeping children in emotionally destructive homes
rather than risk separation from the motherd^ Bowlby himself was forced
to admit in his careful way that “some of the workers who first drew
attention to the dangers of maternal deprivation resulting from separa-
tion have tended on occasion to overstate their case.’’"^^

Problems persisted on the research front as well. Bowlby ’s work on

maternal deprivation continued to stimulate an immense amount of
study, research psychologists replicating and reexamining the studies

upon which Bowlby had relied. Some reported successful replications."^^

But others were punching holes in much of the data. Doubts, for
instance, were raised about the emotional, physical, and intellectual fit-
ness of the children who suffered as a result of maternal deprivation (“In

many cases the institutional population is a substandard one,’’ one critic

said"^^). Since many of the institutions in the early maternal deprivation

studies gave the children precious little in the way of tactile or visual

stimulation, not to mention a caretaker to love and belong to, opponents

were able to maintain that good institutions would create happier,
healthier children and might even, as one would later write, “facilitate
proper social stimulation.”"^^ What’s more, evidence was produced to
show that some children did not seem to suffer any damage if deprived of

maternal care. If some children survived such deprivation intact, perhaps

the real problem of those who didn’t lay within the children them-
selves.^^ (Of course, it could be argued with equal logic that since some
people do not get tuberculosis after drinking infected milk, something is

wrong with those who do.) In 1956, Bowlby published his report on the
sanatorium study for which Robertson had been hired. Far from being
able to solidly contradict the nay sayers, this study suffered from its own
problems with research design, enabling Bowlby to only weakly support
the premises he had hoped to prove.
The fact that children reacted in a variety of ways to a separation from
their mothers — some becoming clingy and others affectionless — fueled

skepticism that both reactions could be caused by the same phenome-

non. Even those who accepted the idea that an early deprivation could
he damaging found it difficult to believe that problems would not disap-
pear once conditions improved for the child. (One pair of researchers

claimed that the mental retardation which seemed to befall severely

deprived infants spontaneously reversed itself when the children got
 Turmoil, Hostility, and Debate 117

older, even without improved conditions.) The idea that severe or fre-

quent separations from the mother may, in Bowlby’s words, be a “fore-

most cause of delinquent character formation” was taken as particularly

hard to swallow.
Bowlby was, meanwhile, subject to frequent misreadings, with some
critics assuming that he neglected to consider the contribution of dis-

torted parent-child relationships or that he warned of the certainty of

damage rather than the The misunderstandings may have been so
risk.

widespread because much was circulated by word of mouth or picked up

from inaccurate secondary sources. Thus the British psychologist R. G.
Andry could write with apparent innocence, “The concept of ‘maternal
deprivation,’ as brought out clearly by Dr.John Bowlby, seems to imply
that one of the most dangerous pathogenic factors in child development
is the harm that may be done if a child has been deprived of the natural

mother’s love through separation.”^^ Bowlby, of course, was talking not

about the child in general but about the very young child, not about sep-
aration in general but about prolonged separation, and not about the
natural mother but the mother figure. Errors such as these set Bowlby up
as a convenient straw man.^^^
The big blow came in 1955, when psychologist Samuel Pinneau pro-
duced a memorable review of Spitz’s work on hospitalism, including his

now famous film on depressed institutionalized infants. Pinneau noted

inconsistencies in the number of children Spitz reported studying; incon-

sistent and even contradictory descriptions of the children, the parents,

and the institutions themselves; and failure to account for cultural,
racial, and socioeconomic differences in the two populations of infants

he studied. He pointed out that the children in the Mexican foundling

home were available for adoption, which would have created a sampling

bias, since presumably the healthier babies were constantly being

removed. Pinneau questioned the validity of the developmental
Finally,

scale Spitz used to evaluate the children. From a scientific point of view

the critique was devastating. The president of the New York State
Psychological Association described the job Pinneau did on Spitz as “a
kind of hydrogen bomb perfection of destructive criticism; not a para-

graph is left standing for miles around.”^^

Although Spitz remained convinced of the truth of what he report-
ed, he was unable to defend himself against this shattering critique,

because in terms of scientific method his study was rickety at best. As his

chief protege, the American developmental psychiatrist Robert Emde,

118 What Do Children Need?

acknowledges, Spitz’s 1946 paper would not be accepted, as is, in a pro^

fessional journal today. As a sensitive clinician observing children who

looked sad and lonely and ultimately hopeless, and who made him feel

despairing when he was with them. Spitz could justifiably suggest that

their emotional and physical decline was caused by the lack of maternal
care.^^ But those who believed that the children were suffering from
genetic deficiencies, malnutrition, birth complications, or simply a lack
of stimulation, could counter with equal justification that Spitz s conclu-
sions were completely unproved. Psychologist C. W. Eriksen, reviewing

the debate in 1957, wrote, “Spitz, in this reviewer’s opinion fails ade-

quately to meet the questions that have been raised. While most of us

will continue to believe in the importance of mothering during infancy,

we must recognize that this belief has more the characteristics of a faith

and less the basis of demonstrated fact.”^^

Flaws were found in other maternal deprivation studies as well. In
some cases the institutions were not named or described; in others the
ages of the children when the separations from mother occurred were
omitted. While Bowlby might insist that the overall body of work sus-
tained his thesis despite such flaws, many in the field, especially those
whose theoretical inclinations lay elsewhere, were now able to dismiss
the whole Bowlby-driven brouhaha about mother-love as so much sound
and fury signifying nothing. A review of the literature by Lawrence
easier concluded that none of the studies Bowlby relied on offered “satis-

factory evidence that maternal deprivation is harmful for the young

infant.” Asserting that “the human organism does not need maternal

love in order to function normally,” Casler argued that a child needs an

adequate range of stimulation and experience and, although these are

normally provided by people, they need not be.^^ He suggested that
Bowlhy’s thesis on the central importance of mother-love had gained
such wide currency less because of convincing evidence than because of
the need of psychologists to believe in it.

Bowlby’s ethological theory, meanwhile, was strictly speculative. True,

it was built on empirical data, but it was data based on animals that were
distant from man. He had made an imaginative leap, and many rejected
it with little trouble. But in the midst of all this,
«
two developments were
taking place, in Africa and the United States, that would tip the balance
in Bowlby’s favor.
 9
Monkey Love:
Warm, Secure, Continuous

In 1958 the psychologist Harry Harlow reported a series of experiments

that every college student now learns about in introductory psych. Har-

low, a researcher and theorist in animal learning at the University of

Wisconsin in Madison and the president of the American Psychological

Association at the time of his now classic report, had been losing many
of his rhesus monkeys to disease. To overcome this he had decided to
separate sixty of his infant monkeys from their mothers six to twelve

hours after birth and raise them in total, germ-free isolation. They were
fed with tiny bottles and they thrived.
But Harlow did notice some curious developments. The infant mon-
keys became ardently attached to the folded gauze diapers that were used
to cover their cage floors. Much like children who become attached to
blankets or soft toys, the monkeys clung to their cloth pads and protested
violently if an effort was made to remove them.^ Other baby monkeys,
raised on a bare wire mesh cage floor, had a difficult time surviving their

first five days. Strangely, they fared better if a wire mesh cone was placed
in the cage and quite well if the cone was covered in terry cloth.^

Such phenomena aroused Harlow’s curiosity and naturally brought to

120 What Do Children Need!

mind Spitz’s study of hospitalism. Harlow now decided to attempt a repli'

cation of Spitz, this time with monkeys. His purpose was not just to
understand the conditions under which withering away arises in infants

and how it can be abated but also, more broadly, the nature of affectional

ties and how they are reflected in biological conditions.

Until this time love had not received much attention from scientists.

Academic psychology was worlds apart from psychoanalysis and its con'
cern for emotions. Its realm was ruled by behaviorists, heirs of Watson
and Pavlov, who considered behavior the only legitimate study for psy'
chology. As a result, Harlow said, all we knew about love was based
on casual observation, intuition, and informed guesswork. Regarding
mother'love, sociologists and psychologists were in accordance with psy'
choanalysts: The baby loves the mother because she feeds it. Harlow
found this implausible.

“It is entirely reasonable,” Harlow said, using the language of behav'

iorism, “to believe that the mother through association with food may
become a secondary'teinforcing agent, but this is an inadequate mecha'
nism to account for the persistence of the infant'maternal ties.”^ This
was an important point. If, for instance, a monkey learns that a poker

chip can be used to obtain a banana, the chip will become a secondary
reinforcer, and he will respond to it with some of the same enthusiasm
with which he responds to food. But if the poker chip will no longer get
him a banana, after a while he will lose interest in it. But, as Harlow saw
it, human affection does not diminish when such associations cease. It

lasts a lifetime. Harlow similarly could not accept the psychoanalytic

emphasis on the breast and the infant’s need to suckle as explanations for
haby'love.
Harlow believed he had in rhesus monkeys the perfect vehicle to test

some of these questions. Aside from being more mature at birth and
growing more quickly, the rhesus infant was almost identical to the
human infant in its responses related to feeding, physical contact, audi'
tory and visual exploration, learning capability, and so on. (The similar'

ity would cost many of these little monkeys dearly, for some of the
experiments that followed were cruel, intentionally duplicating the con'
centration camp'like miasma Spitz had observed in children.) Harlow
now devised an experiment that thirty'Six years later remains one of the
benchmark studies in the field. He separated eight tiny rhesus macaques
from their mothers and raised them in cages where they were entirely
 Monkey Love 121

alone except for access to two contraptions he called “surrogate moth-

ers.” One of the contraptions was essentially a block of wood, softened
with a coating of sponge rubber, and covered with cotton terry. It had a
circular face with large eyes and a light bulb behind it to generate

warmth. The other surrogate was made only of wire mesh but also had a
face and a bulb. For four of the monkeys the cloth-covered mother was
fitted with a feeding nipple. For the other four the wire mesh mother had
the nipple. But regardless of which surrogate mother did the feeding, the
infant monkeys spent virtually all their time, some sixteen to eighteen

hours a day, clinging to the cloth mother. The monkeys’ affectional ties

to their cloth mothers were sustained even after long separations. And
when the infant monkeys were placed in a strange situation, a room
filled with a variety of stimuli known to arouse monkey interest, they

always rushed initially to the cloth mother when she was available, clung

to her until their fear dissipated, and rubbed their bodies against her.

After several sessions like this, they began to use the cloth mother as a
base for explorations.
One of the experimental monkeys was born prematurely, before the
faces for the surrogate mothers had been constructed. This monkey was
thus forced to live with a faceless mother whose head consisted of only a
blank circle of wood. This did not seem to impede the little macaque’s
attachment. But after six months it was given two new cloth surrogates,
one rocking and one stationary, both with completely detailed faces.

“To our surprise,” Flarlow said, “the animal would compulsively rotate
both faces 180 degrees so that it viewed only a round, smooth face and
never the painted, ornamented face. Furthermore, it would do this as

long as the patience of the experimenter in reorienting the faces per-

sisted. The monkey showed no sign of fear or anxiety but showed unlim-

ited persistence.”"^

The monkey’s indomitable preference for the familiar face to gaze at is

indeed reminiscent of human love and some of its corollaries like home-
sickness. It also offered support for a view which Bowlby had expressed,
that part of the human baby’s affectional tie to the mother was its search

for the mother’s face.

Harlow’s studies dealt the first scientific blow to the belief that affec-

tional ties were based on nursing: For rhesus monkeys, at least, cuddly

contact proved far more important — a fact that brought great joy to the

Bowlby camp. But more than that they showed how important it is for
122 What Do Children Need?

the infant to have someone to be attached to, and a particular someone,

as the baby macaque with the blank'faced mother seemed to demon'
strate. The surrogate mothers were woefully inadequate. They offered

security hut they were utterly passive, did no teaching, and did not relate

to the babies, leaving them bereft of all the emotional skills that children

naturally develop by being with their mothers. This relational vacuum

would haunt Harlow’s monkeys in later life, when future experiments

found that they had difficulty relating to peers and more difficulty yet in

raising children. And yet the surrogate mothers meant the world to these

little monkeys. They seemed to love the cloth'covered mother dearly,

despite the fact that she did so little for them, not even feed them.
Harlow was so impressed with the results of his study and the support
it gave to the theories of the British analyst John Bowlby, of whom few in

America had heard, that he wasted no time in generalizing to other

members of the animal kingdom. Indeed, he became rhapsodic on the

subject of motherdove, composing several bits of verse, like this one,

accompanied by a picture of a baby hippopotamus with its mother:

The Hippopotamus

This is the skin some babies feel

Replete with hippo love appeal.

Each contact, cuddle, push, and shove

Elicits tons of baby love.^

In September 1957 Bowlhy went to California for a year as a fellow at

the Center for the Advanced Study in the Behavioral Sciences in

Stanford. Having been alerted to Harlow’s work by ethologist Robert
Hinde, with whom, by now, Bowlby had been sharing ideas for several

years, in April of 1958 he attended an American Psychological AssO'

ciation conference in Monterey, where Harlow gave his paper. “I heard
him speak and I saw his films, which had a very powerful effect on me.”^
On his way hack to England the following June, Bowlhy visited Harlow
for two days in Madison. The two were natural allies, and in the coming
decades Bowlhy and Harlow would remain intensely aware of each
other’s work. Each would inspire so much related research on both mon-
keys and children that the fields of ethology, comparative psychology,
child development, child psychiatry, and psychoanalysis would become
entwined in ways never imagined before.
 Monkey Love 123

* * *

Then in 1961, Ronald Hargreaves, the chief of the Mental Health

Section of the World Health Organization, who had commissioned
Bowlby’s original 1951 report on maternal deprivation, decided to put
out a sequel. asked authors in several disciplines, from pediatrician
He
Dane Prugh to anthropologist Margaret Mead, to consider all the new
research that had become available on the subject since Bowlbys bomb-
shell a decade earlier. Bowlby asked Mary Ainsworth, an American
col-

league who had worked on his staff, to stand in for him. Ainsworth
produced a coherent statement of Bowlby’s and her views. For
brilliantly

the first time in one place she clarified many of the misunderstandings,
successfully repudiated oft-repeated criticisms, and smoothed out some
of

Bowlby’s own apparent inconsistencies and dubious hunches.

Ainsworth broke the debate down into its constituent parts. She

noted that the catch-all phrase “maternal deprivation” was actually com-
posed of three different dimensions —
the lack of maternal care (insuffi-

ciency), distortion of maternal care (neglect or mistreatment), and

discontinuity in maternal care (separations, or the child’s being given to

one mother figure and then another) —
and that these three dimensions
were frequently confounded, making it difficult to study any one of them
alone. Carefully sifting through dozens of studies, she assessed what
they

had to say about the effects of each of these conditions, and, in doing so,
she was able to disentangle many apparent contradictions.

To accomplish this Ainsworth had to chop through a forest of con-

flicting data and make sense of a hodgepodge of variables. The studies

predepriva-
focused on children of different ages, different IQs, different
experiences, and
tion histories, different degrees and lengths of depriving
conse-
they used a variety of measures to assess the emotional and social
quences of these experiences. The result was a massive contusion
in

which it seemed that everyone could find something to support his point

of view. Ainsworth found that many of the studies that seemed to dis-

pj-QYg one or another of Bowlby s hypotheses suffered from unreliable

measurements, sloppy research design, or a misunderstanding on the

disqualifying
authors’ part of the phenomena they were studying, thus
them from a voice in the debate. One research team, for example, took
institution they
the friendliness of maternally deprived children in the
fact that
were studying to be a positive sign, apparently oblivious to the
clinicians since Levy and Bender had cited indiscriminate affectionate-
ness as one of the chief symptoms of psychopathic or affectionless chil-
124 What Do Children Need?

dren. Flaws like this, based on carelessness or incomplete knowledge,

abounded, and Ainsworth sniffed them out, one after another, like a

bloodhound.^
Ainsworth did a better job of defending Spitz than Spitz had done
himself, citing evidence from a French study, whose statistics and meth'
ods could not be faulted, that supported his contention about the severe
deterioration of seriously deprived infants.
After more than sixty pages of sorting through all the data, making

sense of what seemed a hopeless complexity, updating, clarifying, and

indicating where more research was needed, Ainsworth was able to con-

clude — and to conclude convincingly — that Bowlby’s 1951 assertions

were essentially sound. It was a tour-de-force performance, which many
considered to he the outstanding chapter in the new volume, and it won
some who had withheld judgment until now. Bowlby could
the respect of
have had no doubts about what she had done for him.
Meanwhile, Flarlow and Hinde were both reporting a stream of new
experimental results with monkeys, which, if not exactly proving
Bowlby’s theories, at least gave them considerably more credence, espe-
monkey studies variables that confounded the research
cially since in the

on separation among human children could be much more carefully con-

trolled. The monkeys Flarlow had isolated for six months in early life

showed persistent abnormalities into adulthood, particularly in social

and sexual behavior, and proved to be abusive, even murderous, parents.®

Hinde, meanwhile, found that infant monkeys suffered even from short
separations from their mothers and that their distress caused by a thir-
teen-day separation was more pronounced than that caused by a six-day.
The separated monkeys exhibited protest and depression and tended to
he dingier after reunion, just like the children Robertson and others had
observed.^ Interestingly, the young monkeys’ disturbance was greatest if

there had been tension between mother and child before. Five months
after a separation experience signs of stress remained, the young monkeys
being more timid than controls when placed near strange objects or in
strange situations.
Within a few years Harlow and his wife, Margaret, were also reporting
on the development of normal monkey attachment in a paper called
“Learning to Love.” “The outstanding quality of the good primate
mother’s behavior during this time,” they wrote of the first months of
life, “is total or near total acceptance of her infant — the infant can do
no wrong — and she anxiously supervises its beginning sallies beyond her
 Monkey Love 125

arm’s reach.” The monkey mother demonstrates “total, tender, loving

care. She either does not punish her infant or at most punishes it with

complete gentility.” She handles her baby’s physical needs, provides

“physical support and intimate physical contact, which seems to be

important to the development of childhood security.” And she protects

her child from all threats.^ Coming on top of the Ainsworth review,
^

such monkey studies, not to mention analogous studies of rats, dogs, and
other mammals, quieted the opposition and placed attachment theory on
a more secure footing. “Thereafter,” Bowlby said, nothing more was
heard of the inherent implausibility of our hypotheses; and criticism
became more constructive.”^^

But even if attachment theory had gained plausibility, nothing defini'

tive couldbe said about the nature of human attachments based on mon-
key experiments. And given the restrictions on what a researcher could
do with human babies, a more conclusive statement on the infant-
mother bond seemed hopelessly out of reach. Unbeknownst to Bowlby,
Hinde, and Harlow, however, Mary Ainsworth was about to change that.
 If

if
I

*
 PART II

Breakthrough:
The Assessment
OF Parenting Style
 Ainsworth in Uganda

Born Mary Dinsworth Salter in Glendale, Ohio, in 1913, Ainsworth

grew up in Toronto, the oldest of three daughters in a model middle^class

home. Her childhood was suffused with propriety, physical security, and
educational opportunity. When she decided at three that she
wanted to
mother enthusiastically purchased the requisite primers and
read, her

flash cards. But there were troubling emotional currents in the home,
and Ainsworth came to feel that in some fundamental way her
upbring'

ing had failed her, that it accounted for the nagging

doubts and hesitan^

cies she had about herself. Such troubling questions about her own home
life would help set the course of her career.
she entered the University of Toronto in 1929 at
sixteen,
When
Ainsworth encountered her first mentor, ^^(/illiam Blatz, whose abnormal
security theory, and,
psychology class consisted almost entirely of his
troubled by insecurity, she was drawn to it. Blatz had
made emotional
security and the ways in which it is conferred or
withheld the corner^

stone of his personality theory. was impressed with his idea that the
“1

child derives security from being near his parents,”

Ainsworth told me.

“And that that security enables him to move out to explore his world, to
130 The Assessment of Parenting Style

learn about it, and to acquire the skills to master what he encounters out
there. 1 don’t remember if he called that ‘using the parent as a secure

base from which to explore the world,’ but that is how 1 finally came to

phrase it.”^

Toronto’s psychology department was imbued with a messianic spirit,

which Ainsworth quickly and joyfully shared and retains to this day: that

the science of psychology could be used to fundamentally improve the

quality of human life. She became a psychology major, did her doctoral

dissertation on Blatz’s security theory, and went on in 1939 to become

a lecturer at the university, before doing a four^year stint as an army
major in charge of personnel selection during the Second World War. In
1946 she returned to the university, where she and Blatz cO'directed a
team assessing security in various aspects of adult life. She also began
training as a diagnostician during those years and quickly became such
an expert on the Rorschach, the projective ink blot test, that she co-

authored a volume with Bruno Klopfer, the leading Rorschach inter-

preter of the day.
Blessed with a quick mind and a keen eye, Ainsworth was a brilliant

and eager researcher. But she had neither the hunger nor the disposition
of a scientist on the make. Although she could be intellectually tough
and, with her students, rigorous and demanding, she was nevertheless
capable of discounting herself. In 1950, when she married the younger
Len Ainsworth, who had recently completed his master’s degree in psy-
chology, she readily dropped her work in favor of his education. “It didn’t
seem like a good idea for Len to remain at the U of T” where his new —
wife was on the faculty
— “for his Ph.D., so we went to England. He got

admitted to University College, London, and I went along.”^

She assumed she would find work but was not able to get placed

through the few letters she wrote before leaving. “Once in London, I

looked up Edith Mercer, a woman that I had gotten to know in the army,

my opposite numher in the British forces. One day she phoned me up

and said, ‘You know, there is a job advertised in the Times/ It was
for someone to assist in research on the effect on personality and devel-
opment of separation from the mother in early childhood; and they
wanted somebody familiar with .development research and projective
techniques.”^
The ad had been placed hy Bowlby. As Robertson’s work had gotten
under way, Bowlby needed a researcher who could analyze the raw data
 Ainsworth in Uganda 131

with experience
that Robertson was bringing in as well as a clinician
using psychological tests. When he interviewed her, Bowlby recognized
for his team, having not
at once that Ainsworth represented a perfect fit

only the requisite but interests that coincided with his to a remark^
skills
to be one of the
able degree. Answering his classified ad would prove
most significant acts of her career.

Ainsworth was a scientist of Bowlby s caliber — brilliant, subtle, an

demonstrated and
independent thinker, sensitive to what empirical data
what it But their personalities, at least in terms of outward mea^
did not.
sures of confidence and extroversion, were markedly different. Bowlby
in his cot'
was a leader with a commanding presence and a conviction
opposition. Ainsworth,
rectness that only grew stronger in the face of
socially and
although tough-minded intellectually, was more retiring
would soon
prone to self-doubts when encountering rejection, which she
face on a grand scale.
she
Ainsworth did not have destiny writ large in her features,
If

quickly recognized that Bowlby did. She was not

only taken with his
apparently secure personal-
ideas, but struck, too, by his formidable and

ity. “He made no bones about the fact that he was single-handedly fight-

ing the analytic establishment, that it pained him some, but that he was
was a long time before felt any
convinced he was on the right
1
track. It

sense of getting close to him or being a friend. But 1 had no difficulty

even though he’s

whatsoever making him into a surrogate father figure—
professional marriage
not much older than I.”^ It was the beginning of a
that would prove as fruitful and enduring as
any in the history of psy-

chology.
During her three and a half years with Bowlby,
Ainsworth got caught
baby need a mother fig-
up in the issues that concerned him. Why does a
a prolonged
ure to develop normally? Why is it so adversely affected by
separation from her after the relationship has
been established? Why
separation, is the
when they are reunited, especially after a depriving
is not recovered tor a
child so deeply upset that the original relationship
about herself. She wanted
long time, if ever? She had parallel questions
own insecurity, which was clearly not based on any
to understand her
separations or losses but which she nevertheless suspected had something
to do with her and her connection with a rather self-absorbed
early years

and distant mother.^ And so the question formed, What goes on between
year that lies behind all this? This ques-
the mother and child in the first
132 The Assessment of Parenting Style

tion, brewing in a sense since Blatz, became so important to her now that

she was determined that when she had her own project, it would be a
study of the mother-child relationship in the first year of life.

Ainsworth’s clinical background had been entirely the realm of psy-

chological assessment. She was not a therapist and, unlike Bowlbys fel-

low analysts, was therefore not absorbed in the details of patient work.
As a personality and developmental psychologist, she was more con-
cerned with how a child progresses cognitively and emotionally, ques-
tions that analysts were also concerned with but had never studied
scientifically. To her, therefore, the ethological model, and particularly
the way ethologists had closely observed family relations in other species,

proved a powerful attraction. Perhaps human infants and families could

be observed the same way.

Robertson’s method of observing children had an especially strong
impact on her. “It was Jimmy’s work that I most admired,’’ she later said.

“In studying separation he got acquainted with the families before the
child was separated; he did observations of their behavior during separa-
tion, and followed them when they came home. And made I up my mind
that whenever I went elsewhere and could start a project, it would be a
study of this sort — direct observation in the natural environment.’’^

That is what she did in Uganda.

In 1954 Len Ainsworth got a job at Makerere College in Uganda (now

the University of East Africa). Although Mary, again, had nothing lined
up for herself there, she soon launched one of the pioneering studies in

modern infant research. With no lab, with meager institutional support,

with no help collecting or analyzing the data, with no one but her inter-
preter, she rounded up twenty-eight unweaned babies from several vil-

lages near Kampala and began observing them in the home, using the
careful, naturalistic techniques that Lorenz and Tinbergen had applied to

goslings and stickleback fish.

When Ainsworth arrived in Uganda in 1954, infant research was it-

self in its infancy. Babies had of course been studied, especially by such
pioneer psychologists as Mary Shirley and Charlotte Biihler, and the
neonate was just now coming under renewed attention. But in the previ-

ous two decades child development research had focused almost exclu-
sively on older children, mainly four and up,^ with the exception of

Swiss psychologist Jean Piaget, who had given considerable attention to

cognitive development in tbe first three years of life. Emotional and rela-
 Ainsworth in Uganda 133

tional development during those crucial years had been neglected. In

1961 Rudolph Schaffer and Peggy Emerson, who did one of the impor^

tant early studies of attachment, could summarize the

hard knowledge

about infant social development in one long sentence: “We

now know
overtures by a
that at the age of 4 weeks the infant will react to social
appears; that
reduction of bodily activity; that at 6 weeks the first smile
from 2 months on the mother will be visually recognized; that at 3

months the infant may vocalize in reply to others’ speech; and that after

8 months he no longer smile indiscriminately at all and sundry.”

will
going
Longitudinal studies, in which infants were carefully observed
meager
through the stages that brought them from newborns, with very
one^year^olds were
relational abilities, to highly attached, eagerly related
unheard The processes and variables involved were largely a matter of
of.

conjecture.
Kampala, in the province of Buganda,
The Ainsworths settled in

which for centuries before colonization had been a powerful and inde-
pendent kingdom with a highly developed and complex society. The

families Ainsworth studied lived in six villages about fifteen miles out'

side the capital, in a thickly populated

agricultural area. The houses,

each of which were surrounded by about an acre of land, where the fam^
ily grew itsown vegetables and fruit, were rectangular dwellings with
corrugated
several rooms. The floors were earthen, the roofs thatched or
metal. They had no electricity or running water. All the families were

either Christian or Muslim, all the schooLage children attended school.

None of them spoke English.

The East African Institute of Social Research, located on the Make'

campus, gave Ainsworth a modest grant to support her
work as well
rere
to the standard
as the companionship of fellow researchers. According
practice, new researchers spent a month in a Ganda village acquainting
followed suit.
themselves with the customs and language, and Ainsworth
to engage in
She picked up enough Luganda during this short period
standard small talk, to understand much of what
was said to her, and to
told the chiefs she
speak to the local chiefs about her purposes. She
needed a sample of unweaned babies for her study. After overcoming

their initial suspicions, they arranged meetings with local villagers,

where Ainsworth gave her speech again.

Ainsworth offered them tranS'
In return for the mothers’ cooperation,
portation to the nearest clinic, not just for mother
and baby but tor all
in those days. Eventually,
the children in the family, a valuable service
134 The Assessment of Parenting Style

she also provided dried skim milk to the families at wholesale cost.
Volunteers were few at first, but in the end, Ainsworth had to turn sub'
jects away. Her final sample consisted of twenty^eight babies from
twenty'three families (there were two sets of twins and two polygamous
households). With her interpreter, Katie Kibuka, Ainsworth visited each
home for two hours every two weeks for nine months. Mrs. Kibuka, a
Ganda woman who’d been educated in the United States, conducted
most of the interviewing of the mothers while Ainsworth observed.
Their visits were more like social calls than scientific interviews. They
made frequent shorter visits as well, sometimes to deliver medicine or
to render some other service, sometimes just to preserve good working
relations. In short order, Ainsworth was immersed in the lives of her

subjects.

Ainsworth’s original idea was to get to the bottom of the debate that
raged around early separation. Was separation from the mother harmful,
per se? Or was the real problem the deprivation of maternal care that
usually accompanied separation? Ainsworth had heard that the Ganda
separated infants from their mothers at the time of weaning, usually to be
reared by grandmothers. This ancient custom, whose reported purpose
was to encourage the baby to “forget the breast,’’ appeared to offer
Ainsworth what is known as an “experiment of opportunity.” For here
she would certainly find separation in its purest form, without the con^
founding variable of deprivation. She assumed she would also find a
group of modernized families who did not follow such practices with
whom she could make a comparison. She further hoped to contrast the

effects of abrupt weaning, which was reputedly the custom among the
Ganda, with gradual weaning, which she expected to see among those
who had abandoned ancient customs. But Ainsworth had not even fin-
ished collecting her sample when she realized that the expected experi'
ment of opportunity would not materialize. The European assumptions
about Ganda child^rearing practices proved inaccurate; abrupt weaning
was not a common practice after all, and very few children would be sep'
arated from their mothers when taken off the breast.

Ainsworth had already begun collecting data, however. And she was
observing everything: breast feeding, thumb sucking, bathing, general
cleanliness, bowel and bladder training, bed wetting, soiling, sleeping

arrangements, carrying and holding the baby, sharing of mothering, maS'

turhation, distress, anger, crying, as well as how the babies responded to
discipline, beatings, medical care. She was keeping tabs on the total
 Ainsworth in Uganda 135

amount of care given. She was tracking developmental milestones, like

sitting, crawling, standing, and walking. And all this data, never before

collected by a social scientist, was pointing her in a new direction: She

was observing attachment in the making.
The baby is born and can be readily given to a foster parent without

apparent harm. But sometime during the first year all that changes, such

that even brief separations from the mother figure are a cause of distress.

“How,” Ainsworth asked, “does this attachment develop? What factors

development or delay or prevent it? What are the

facilitate this
criteria

which enable us to judge that an attachment has been formed?”^

When Ainsworth first arrived in Uganda she still adhered to the tradi^

tional view that attachment was born of feeding. But as she

observed the

families, herimmediate impression was that Bowlby had been right. The
baby is not a passive recipient creature who becomes attached to
his

mother because she satisfies his needs. These were very active babies.
They went after what they wanted. I began to see certain behaviors that
indicated that the baby was becoming attached, and 1 was
able to list

There was, for instance, the

them in chronological order of appearance.

differential stopping of crying. The mother picked up the baby, the baby

would stop crying, but if somebody else tried to pick him up at that

point, he would continue to cry. Differential smiling. Differential vocah

izations. began to see different situations where attachment to the

1

mother could be spotted; and you could differentiate an attachment fig'

ure from some other person, even a familiar person.”
For each child she now paid careful attention to a
growing list of
In addition
behavioral patterns that she saw as typical of attached babies.
to differential crying, smiling, and vocalizations,
she included, crying

when the mother leaves; following mother; showing concern for

the face in
mother’s whereabouts; scrambling over the mother; burying
mother’s lap; using the mother as a safe haven when in a
strange situa-

flying to mother when frightened; and greeting

her through smih
tion;

ing, crowing, clapping, lifting the arms, and general excitement. She
behaviors and
noted the extent to which each baby demonstrated these
the age at which they first appeared. One baby, Samwendi, showed dit'

ferential crying at nine weeks. “He was restless,” Ainsworth wrote,

“when held him, despite my efforts to quiet him. When Mrs.

Kibuka
1

position, and finally

held him he was restless also, despite several shifts of
he cried violently. He stopped crying as soon as his mother took him.

Other examples from her notes:

136 The Assessment of Parenting Style

• William (twentyAour weeks) began to crawl and to try to

follow his mother whenever she left the room, letting out an
immediate cryd^
• Muhamidi (thirty^three weeks) was brought into the room
by a young girl and put down on the floor. He immediately
crept to his mother, smiling and crowing with pleasure,
• Nakiku (twenty^eight weeks) sat near her mother on the
floor, bouncing about, smiling, and gurgling, and then crawled
to her mother and buried her face in her mother’s lap.”^"^

Having been interested all her professional life in the subject of secu'

rity, and having been indoctrinated early on in Blatz’s security theory,

which included the idea that the parents provide a secure grounding

from which the child feels he can safely make excursions into the world,
Ainsworth was sensitive to how the Ganda babies used their mothers in
this respect.Again and again she observed that once they were able to
crawl the children would make small excursions away from their moth-
ers, always maintaining consciousness of her whereabouts and returning

to her periodically, either physically or with a smile. Using the phrase

that would become a cornerstone of attachment thinking, she wrote,

“The mother seems to provide a secure base from which these excursions
can be made without anxiety.’’^^

If a child is attached, and secure in his attachment, she decided, he

did not need to be glued to his mother. “He can even leave the room on
his own initiative and his aplomb in so doing is sometimes in sharp con-
trast to his consternation when his secure base gets up and moves off.’’*^

It was just that consternation which demonstrated that exploratory

behavior was an aspect of attachment dynamics, that the one could
barely proceed without the other.
She witnessed one child after another using the mother this way.
Some started exploring later than others, some took longer than others
to wander off mother’s lap when strangers were present. A brave baby
might gain enough reassurance by making eye contact with his mother
and exchanging smiles from across the room, but most would return for

contact before going off on another jaunt. One little boy, Paulo, clam-

ored away when his sisters left the house, but followed them only as far

as the door, where he stood, holding on to the doorjamb, unwilling to

go any farther. Paulo spent much of his time playing away from his
mother but protested vehemently if she left the room. Information such
 Ainsworth in Uganda ^37

as this, so familiar to sensitive parents, had never before been scientific

cally observed.

When some did whenever Ainsworth and

a child felt threatened, as

Kibuka entered the house, he would stay close to his mother, and
exploratory behavior would cease. In strange situations, like the clinic
where the children were taken periodically, they would often fly to their
mothers when separated.
Ainsworth’s observations led her to hypothesize five phases in the
development of attachment. The first is an undiscriminating phase, in
which the newborn baby has almost no social responses and then later

responds readily to almost anyone. A phase of differential responsiveness

follows in which the baby shows signs of knowing and preferring his

mother. In the third phase the baby is able to respond differentially from

a distance, crying when he sees mother leave the room, crowing when
he Ainsworth believed that the onset of separation anxi-
sees her return.

ety in this phase heralds the beginnings of true attachment.

During

this period the children were also beginning to differentiate

between

strangers and familiar figures and would tend to stare at a stranger

for

prolonged periods. The fourth phase is one of active initiative — follow-

ing, approaching, burying the face in mother s lap, scrambling over her.

Children not only protest when mother leaves the room now hut typi-

cally crawl after her. They greet her actively when she returns. And they

begin leaving her, cautiously, to explore the world around them. Toward
the end of this period, at around six to eight months, uneasiness
with

strangersbecomes more conspicuous, thus initiating the phase of stranger

anxiety. Some of the babies who had accepted Ainsworth and
Kibuka

when they were younger now stared at them, would not approach even
when invited, and became tense when held. Babies who did not know
them showed more conspicuous anxiety.
In the last three months of the first year, most of the Ganda babies
became efficient initiators of attachment, bolder explorers of their envi-
ronment, and markedly more alarmed by strangers. This was a time when
strangers
clinging to mother became more pronounced in the presence of
or in strange situations. But there was enough variation among the
babies for Ainsworth to note that the development of attachment
dif-

fered from child to child and that methods of infant care

and cultural

factors both play a role in that variability.

Weaning was the most upsetting event for the babies, especially those
to being fed on demand (a practice that Ains-
who were accustomed
138 The Assessment of Parenting Style

worth nevertheless supports). The insecurity manifested around this time

was worsened by mothers who withdrew somewhat from their children to
drive home the point that breast feeding was now over. Clinging, crying,
protests over the slightest distance from mother, and other aspects of
insecurity now became more pronounced.

Parenting style seemed responsible for several broad differences among

the children. Five of them, whose mothers were absent a great deal or
who barely responded to them, did not develop any of the attachment
behaviors that were so prominent in the others. Ainsworth called them
“nonattached” or, perhaps more accurately, “not yet attached.” Nora was
left by herself in a crib much of the day and was not responded to when
she cried, a fate typical of the nonattached group. She did not cry when
her mother left the room, did not greet her mother when she returned,
did not use her mother as a secure base, indeed seemed to be precociously
independent. But Nora was not completely nonattached, for she de^
lighted in seeing her father, who came home only on weekends. When
he was there, she approached him again and again.
The causes of nonattachment seemed easy to identify. In the case of
one pair of twins, Ainsworth recalled, the mother became extraordinary
ily upset and anxious over their demands. “She would he feeding one and
the other one would he screaming, and she just didn’t seem to he able to
satisfy these babies for any length of time. She just gave up trying to
respond the way a Ganda woman would respond normally to her hahy,

which is to he very sensitive to hahy signals and to give the breast at any
sign of wanting it — at least a great many of them would do that. She had
a large hahy carriage that she had gotten trom somewhere. She put one
child in one end and one in the other end and wheeled it out to the edge
of their property so they could scream all they liked and she wouldn’t
hear them. Then when she thought it was time to teed them, she would
bring them in and tend to them, and then off they went again. So she
really was rejecting as well as neglecting.”'^

The other two nonattached babies were children ot two wives in the
same family, who had a total of five small children between them. “This
was the most affluent family in the sample,” Ainsworth would later note
“and these mothers had learned from Europeans that each hahy should
have its own room, each hahy should have its own crib.” The room was
shared by the mother, hut during the day the two babies were often left

there alone. “These two ladies rather liked polygamy — each could have
 Ainsworth in Uganda 139

completely free days. So one would go off shopping, and the other would
take care of all these kids and not really give a lot of attention to the
baby in the isolated room or heed its cries.” When Ainsworth and
Kibuka came to visit, the babies would be brought out and set on the
floor beside the mother, hut they showed none of the attachment behav'
ior so striking in the rest of the sample.
All of the other children were clearly attached, hut, again, there were
differences, and it was apparent to Ainsworth, especially when she
assessed how the babies responded to separation
— “Did the baby cry
when his mother left the room? Did he follow her? Did he cling to her to
try to prevent her going, or did he greet her return by clinging
?”^^
— that

some were more secure in their attachment than others. She noticed
that certain babies seemed to cling and cry excessively, while others

hardly responded at all when their mothers were about to leave them.
“Were these babies less attached,” she asked, “than the children who
clung to their mothers and would not let them go? Or were they perhaps
simply more secure in their relationships with their mothers?”^^ This
would turn out to be a critical question, which she would not be able to

answer until a decade later.

Ultimately, Ainsworth labeled seven children insecurely attached.

They could not tolerate any distance from mother, needed almost con^
tinuous contact with her, and remained fussy and often continued to cry
even when being held by her. Muhamidi was one such child.

Muhamidi came from a large, troubled, and disorganized home. The

parents did not get along, the father did not help with the children, and
he expected his wife to work in the fields with the coffee harvest despite
her many other responsibilities. An older daughter was perpetually ill

and occupied much of the mother’s time. The mother was anxious.
Muhamidi slept with his mother and was hreast'fed on demand, two
factors that one might think would virtually guarantee secure attach'

ment. He was healthy and well developed, and his mother took him
everywhere she went, often carrying him on her back, even when she
worked in the garden. Ainsworth gave her the highest possible rating in

total amount of maternal care given to the child. But, despite having

looked secure at thirty-two weeks, at eight months Muhamidi showed

marked signs of insecurity. He cried a great deal. He demanded constant
proximity to his mother, and he wept as soon as they were parted. He
demonstrated little interest in anyone else. When he was tested at nine
months with the Gesell Developmental Schedules (motor skills, laii'
140 The Assessment of Parenting Style

guage skills, social behavior, general adaptation), he was found to be a

hoy of high vitality with a Developmental Quotient 30 points above
average. When he was tested again at twenty-two months, he showed
less vitality and his Developmental Quotient had fallen from 130 to 109.
At months he seemed sad and dull and his DQ was 96.
thirty^four

Muhamidi’s brother, tested by the same doctor, experienced an even

more dramatic decline from 142 to 76 over a similar period. One can
only wonder, when reading such figures, whether insecure attachment
had undermined the initially robust development of these children.

Questions like this would be asked with increasing frequency as

Ainsworth refined her work.

Ainsworth, meanwhile, struggled to understand why certain children,
like Muhamidi, should be insecure. Anecdotal evidence was suggestive.
She noted that some of the mothers of the insecure children “were them-
selves highly anxious, having been separated from or deserted by their

husbands and finding it difficult to establish a satisfactory life apart from

them.”^^ One mother had to leave her child with a neighbor all day so

she could work in the garden to raise food, a desperate necessity since her
husband’s desertion. Another mother was plainly rejecting of the child, a
factor that Bowlby had called attention to years earlier. As for Muham-
idi’s mother, she was distressed that she could not give him the joyful
attention she thought she should; she was invariably preoccupied by
other things. Although she kept him with her all the time and gave him
the breast when he wanted it, there was at times a perfunctory quality to

her caregiving, as this observation by Ainsworth suggests:

We found Muhamidi sitting outside the house crying bitterly.

His sister was lying on a pile of cloths near him, appearing

very ill and paying no attention to the baby. The mother
appeared from behind the house where she had been working.
She ushered us into the house, giving first attention to the

polite greetings. She then went out to get him and put him to

the breast immediately, without stopping to wipe his nose,

which was streaming, or to wipe his face, which was muddy.^^

Muhamidi’s mother ultimately struck Ainsworth as disturbed herself

— “a masochistic person who always did things the hard way and who
compounded the difficulties that a malign fate had visited upon her.’’^^
 Ainsworth in Uganda HI

Despite such observations, Ainsworth could not confidently enumet'

ate the specific maternal behaviors that made the secure infants secure
and the insecure infants insecure. Sharp boundaries that might different
tiate the two were hard to come by. Almost all the mothers were warm
and affectionate, the only exceptions being two mothers of nonattached
children. One factor that did prove a relatively dependable demarcation
was the mother’s enjoyment of breast feeding — twelve of the fourteen
mothers who said they enjoyed breast feeding had securely attached
children. But this factor is more in the nature of a predictor than a
description of the quality of care that would lead to secure attachment.
The total amount of care a child received — from mother, father, sib'

lings, grandparents, nursemaids, father’s second wife — proved not to be

a significant factor. This finding was interesting from the point of view
of the debate over whether a child needed a single main caregiver or
would do just as well if not better with lots of caregivers. Margaret Mead
had argued for the superiority of multiple caregivers. But Ainsworth’s
data suggested that many providers were fine only as long as the care was
not so diluted that no one person would become a significant attach'
ment figure.

The amount of care provided by the mother herself, on the other

hand, did clearly distinguish among the three groups of children. The
mothers who gave the most care tended to have securely attached chib
dren, the mothers who gave the least nonattached, with the insecurely
attached falling in between. Was this, then, the variable she’d been seek'
ing? Ainsworth did not believe so. Who after all had received more care
from his mother than Muhamidi? Although quantity often suggests qiiah
ity, the two are not the same. Something about the care, the quality ot
the care, must have mattered. But how could that be factored out? She
could not yet answer this question.
Infancy in Uganda, the book Ainsworth eventually wrote, was a
unique document, rare in its meticulous attention to detail, original in its

mixture of longitudinal and crosS'Sectional techniques, and the first to

demonstrate the development of attachment, making sense of infant

behaviors and developmental sequences that were poorly understood
before. It also planted the concept of the secure base, which would grad'
ually become more important in the way we think about the parent'child
bond and about human ties in general.

The secure'hase concept has proved particularly useful in recent years

142 The Assessment of Parenting Style

in understanding the maddening selhdestructive behaviors of certain

seeking.
toddlers, behaviors formerly identified as “negative attention

The tendency to get hurt repeatedly while exploring, to wander away

from the mother in public places, or to simply walk out the front door

and disappear down the street — behavior some of Bowlby’s

that recalls

juvenile thieves —
are now seen by psychologists Alicia Lieberman and

Jeree Pawl as indications that the child does not feel his
mother is prO'
viding a secure base for him. By behaving in these ways the toddler
is

in effect, “How far do I need to go before my mother will

bring
asking,
me hack? How much danger is too much so that my mother will protect

me? How much fear do 1 need to experience before Pm helped to feel

safe?”^^ Such issues, in different form, would prove to have meaning

throughout childhood, adulthood, and old age as well.
Infancy in Uganda was not published until 1967, having been
held up

hy personal upheavals and by Ainsworths fastidious effort to make sense

of the data she’d collected. But by that time she was deeply into some^
thing that would change the course of developmental psychology.^^
 The Strange Situation

In 1956, for a third time, Ainsworth changed continents to follow her

husband. This time to Baltimore, where, within a few weeks, a teaching
and clinical job was patched together for her at Johns Hopkins Uni'
versity. It would be seven years before she managed to start her next

longitudinal study, during which time she would divorce and enter psy-

choanalysis. The connection with Bowlby had grown thin, hut when he
visited her in 1960, just as her marriage was dissolving, he came face^tO'
face with the findings that she would eventually publish as Infancy in

Uganda and understood for the first time how pertinent they were to his
work. Indeed, this was the only major study outside his own unit offering
empirical support for his theory. In terms of their relationship, says
Ainsworth, “that made all the difference.”^ From being Bowlby’s most
capable adherent, she had become an equal colleague. In a few years she
would be an acknowledged partner.
Ainsworth arrived at Hopkins when she was forty^three years old,

without offers or prospects. She started out teaching an evening course

in an area where she had great expertise and even some renown — per'

sonality assessment and intelligence testing. Supplemented with a couple

144 The Assessment of Parenting Style

of days a week at Sheppard Pratt, a private mental hospital near Bah

timore, she gradually had enough work to keep her going, but none of it

related to her true interest,which was now infant development. She

looked back on Uganda with pleasure and longing. It had been a happy
time for her. She loved doing research, and she loved contact with the
babies, which her own marriage had failed to produce. As she later said,
“I could hardly wait until 1 had the opportunity to replicate the thing.”^
But it was not until another five years had passed, in 1961, that she
was able to convince her department chairman at Hopkins that her true
love was developmental psychology and that she wanted to return to it.
In 1963 she received a modest grant to perform a pilot study for her next
project, a longitudinal observation of mothers and infants in Baltimore.
She managed with that meager funding to do the full study itself.
Although some eight years had passed since her last longitudinal
research of infants and mothers in the natural environment, Ainsworth
was still working in largely virgin territory. Psychologist Katherine Wolf
had observed babies in the home in Boston and was widely believed to
be the person who knew the most about infant development, but she had
died before being able to get her data together and the notes she left

behind were indecipherable to others. Psychoanalyst Louis Sander had

also worked on a longitudinal study, much like Ainsworth s, which she

herself considered superb. He had published several clinical papers refer^

ring to the work, in which he gave his impressions of certain develop^

mental stages in infancy based on what he’d observed. But the study ran
for a long time and funding was cut off before he was ever able to analyze

his data. Other researchers, like Sylvia Brody, had observed infant-

mother interaction in a lab or a play center. In Brody’s case the lab was
fitted out to look just like a home. But to Ainsworth a home in a lab was
not the same as a real home.^
“Just take feeding. In the home environment, 1 could see how a

mother responds to infant signals when she had a lot of other demands
on her time, with the telephone and housekeeping and other kids if
there were any. saw one mother who was working very hard to put her
1

six-week-old hahy on three meals a day — and she was breast feeding at

that! She would say, ‘1 don’t know why the baby’s crying. He was fed at
seven o’clock this morning,’ it now being after twelve. She would pick it

up and play with it very nicely for a while and then put it down, and it

would cry again. She would dangle a rattle, she would do this, do that,

she even gave it a hath one day to fill up the time till one o’clock, with
 The Strange Situation 145

the baby off and on screaming. You would never observe that type of
thing in the lab.”'^

The mothers’ different responses to their infant’s crying was also more
obvious in the home environment. Some ignored 96 percent of their
baby’s cries, while others ignored only 4 percent; some mothers averaged
two minutes in responding, while others averaged nine minutes. Some-
times Ainsworth sat and counted the minutes, waiting for what seemed
an eternity while a baby howled in anguish. Some mothers steeled them'
selves to wait it out as long as they could, often because they believed
that not responding would train their baby not to cry. Some were at

times too engaged in other activities to be able to attend to the baby.

Some, incredibly, simply did not register that the baby was crying, per-
haps because they were too anxious or depressed themselves. (This
proved particularly hard for an observer to sit through.) As Bowlby later

remarked, “Had the observers not been present to see and hear what
was going on but had relied insteadon what the mothers had told
them, the pictures they would have got would in many cases have been
.’’^
entirely false. . .

Ainsworth put together a team of four observers and a total of twenty-

six families with babies on the way. The researchers would make eigh-
teen home visits of four hours each over the course of the infant’s first

year. As in Uganda, Ainsworth and her colleagues acted as friends, not

furniture — talking, helping, holding the babies, becoming part of the
family — in order to encourage the mothers to act more naturally. “To
have somebody there for an extended period of time just watching and
taking notes could be very tension producing. Besides, I wanted to see

whether the baby would smile at us, whether he would cuddle when we
picked him up, and how the baby would behave with us in comparison
with the mother.”^
In the Baltimore study, Ainsworth Uganda
hoped to replicate the

research and make it much more systematic. But she now had some spe-
cific questions she was eager to answer. First among them concerned the

patterns of attachment behavior she’d observed among the Ganda. Were

these patterns universal? Would American babies behave the same way?
Ethology stipulated that they should, that all human babies are horn
with a certain species-specific repertoire of instinctual behavior patterns.
Ainsworth was clearly thinking along these lines: “The baby gives cer-

tain signals that he doesn’t intend as signals; he behaves in certain ways

under certain circumstances to bring the adult caregiver to attend to him
146 The Assessment of Parenting Style

and it all goes from there” — the meanings, the quality of relatedness, the

feelings of love and security. “It’s not that he is hungry and gives these
signals for a purpose; he is upset about something. When he s alone, he is

apt to give pretty much the same kind of signals as he does when he is

hungry or has a pin sticking in him or is wet or whatever.* And the thing

that was increasingly exciting to me after I had started the study of the

American babies was to find that the behaviors that I had come to call

attachment behaviors in the Ganda infants were essentially the same as

for the American infants.”^

Indeed, of the sixteen attachment behaviors she’d observed in the

two studies, she found only two differences. The Ganda children clapped
when greeting attachment figures, something the American children
never did, and the American children hugged and kissed, something the
Ganda children never did. These differences appeared to he culturally
derived, as Ganda adults often clapped with pleasure when returning after

an absence, while American adults frequently demanded that their babies

give them a hug or a kiss, which Ainsworth never observed in Uganda.
If Ainsworth had stopped there, she would have produced another

valuable and pioneering study, both replicating the Ganda research and

demonstrating its cross-cultural validity. But she had a problem in mak-

ing a certain critical comparison between the Ganda babies and middle-
class American babies. “1 all along had this idea about a secure base. It

was so conspicuous with the Ganda babies. If the mother was there, the
kid would roam all around the room and explore things, looking back at
mother and maybe giving her a smile, but focusing most of his attention

in the environment. And just as soon as the mother got up to leave the

room, the chances were the baby would shriek and absolutely stop any
kind of exploratory behavior.
“Now, the Ganda babies are much more used to having their mother
with them all the time. Whereas the Baltimore babies were used to hav-
ing their mt:>thers come and go, come and go, and they were much less

likely to cry when their mother left the room. So when they were happily
exploring it wasn’t clear if it was because the mother was there or not.”®
These questions brought to mind a paper Ainsworth had read in 1943

*Thar the hahy’s cry is undifferentiated at first is a matter of debate. According to Barry Lester, a
developmental psychologist at Brown University, the hunger cry of newborns is acoustically differ-
ent from their pain cry and their distress cry. He adds that “by four to six weeks when they have
more control over their vocal cords, hahies develop a cry to communicate frustration and the ‘fake
cry’ that is a hid for attention.” (Goodman, 91-92, p. 27.)
 The Strange Situation 147

called “Young Children in an Insecure Situation” by Jean Arsenian, who

had put babies into a playroom, some with their mothers or mother fig'

ures and others by themselves. “She didn’t talk about exploratory behav'
ior, but she made it quite clear that the ones that were brought in with
their mothers could take a constructive interest in the environment,
while the ones that were brought in without a mother figure spent most
of their time crying. 1 always remembered that, and when I started work-

ing on the infant stuff, 1 thought sooner or later 1 wanted to try this out.
“So 1 thought, all right, if you don’t see the secure base phenomenon
very clearly at home, that doesn’t necessarily mean it doesn’t exist. It

could very well be different in a strange environment, such as Arsenian

used — a playroom in a lab. If 1 could bring the children into the univer-
sity with their mothers, maybe I could see how they used the mother to
explore.”^
Around this time Harry Harlow had reported yet another of his exper-
iments with rhesus monkeys, which further confirmed Ainsworth’s
thinking. He had taken the monkeys raised with the cloth “mothers” and
put them in an unfamiliar environment. If placed there with their cloth
mother, they would rush to her and cling to her in terror, then gradually
climb down, look around, and move off to explore. If alone, their crying

and terror would not abate.

Now, it turned out that the Baltimore babies not only demonstrated
less secure base activity than the Ganda, they showed less stranger anxi-
ety, too, and they were much less likely to cry when their mother left

the room. Ainsworth hoped that an unfamiliar setting might raise the
threshold of threat on all three accounts and thereby ascertain whether
these infant behaviors were indeed universal aspects of attachment.
“I thought. Well let’s work it all out: We’ll have the mother and baby
together in a strange environment with a lot of toys to invite explo-
ration. Then we’ll introduce a stranger when the mother’s still there, and
see how the baby responds. Then we’ll have a separation situation where
the mother leaves the baby with the stranger. How does the baby
respond to the departure? And when the mother returns, how does the
baby respond to the reunion? But since the stranger was in the room dur-

ing the first departure, maybe we’d better have an episode in which the
mother leaves the baby entirely alone. Then we could see whether the
return of the stranger would lessen whatever distress has occurred.
Finally, we’ll have another reunion with the mother. We devised this
thing in half an hour.”^^
148 The Assessment of Parenting Style

Thus the Strange Situation was born, a laboratory assessment that

would come to be more widely used than any other in the histor^^ of
developmental psychology. A room was set up at the psychology depart'
ment of Johns Hopkins University with a one-way mirror for observa^

tion. It had three chairs, and a bunch of toys on the floor.

Ainsworth knew, of course, that the babies differed from one another:
She’d seen variations in security among the Ganda and she’d seen it

among these Baltimore infants, too. So she expected that they would

behave differently in the inherently stressful strange situation into which

she was now going to place them.
“The mothers who had been particularly good about responding to
signals had babies who responded just exactly the way 1 had predicted
babies would respond in this kind of situation. Indeed, she already had

a script in mind: After the brief introduction, which she labeled epi'

sode one, came episode two — the baby alone in the playroom with its

mother. Ainsworth expected that most of the babies would now fairly
readily leave the mother to explore, that they would go to the toys, and

that they would play with them. She expected that in the third episode,

when the stranger entered the room, the babies would react to the
stranger, some would go to the mother, and exploratory behavior would
diminish. She anticipated that most of these babies would become dis'

tressed and probably cry when the mother left the room in the fourth
“1 thought they might
episode, even though the stranger was still there.
go to the door and try to follow the mother and search for her, or 1

thought they might go to the chair where mother left her pocketbook

because that’s what they had associated her with.” In the fifth episode,

when the mother returned, Ainsworth expected that the baby would
approach the mother and want contact with her and then settle down.
The mother was instructed to put the baby hack on the floor if she’d
picked up during the reunion, and Ainsworth assumed the baby would
it

become interested in the toys again at that point. During this episode the
stranger slipped out of the room. In the sixth episode the mother left

again, so that the hahy was entirely alone for the first time, and
Ainsworth expected that this would be even more stressful than the first

separation. When the stranger came hack and tried to comfort the baby

in episode seven, Ainsworth did not expect the hahy to he comforted. In

the final episode, when the mother returned, she anticipated that the
hahy would demonstrate even greater eagerness for contact than it had
shown before.
 The Strange Situation 149

The very first child put through the Strange Situation behaved
exactly the way Ainsworth had anticipated. “It was the only child I have
ever seen who absolutely matched what I expected.”^ ^

Despite the surprising variability, Ainsworth’s expectations were strongly

validated. Although it was a strange environment, almost all the babies
began to explore, usually keeping visual tabs on their mother at the same
time. When the stranger arrived, exploration diminished markedly, and
most babies spent more time looking at the stranger than the toys. Their
wariness was evident mostly in gaze aversion but also in some cases in
physical avoidance and crying. Still, the pleasant stranger also managed
to get at least one smile from over half the babies.
When the mother walked out of the room the first time, half the
babies cried at some point. A large minority of the babies were lured
back into play by the stranger, but few of those who were truly distressed

could be fully comforted by her. When mother returned, the vast major-
ity of the babies greeted her by smiling, vocalizing, or crying, or more
often with some combination of these reactions. About half of them
wanted physical contact, the majority of them achieving it within fifteen
seconds.
When the mother slipped out a second time, now leaving the baby
alone in the room, the distress was usually very intense, with most of the
babies crying, many of them so pitifully the episode had to be curtailed.
When the stranger returned, she did not have much success in comfort-
ing the babies who are distressed, although many of them did accept her
soothing attentions, allowing themselves to be picked up, and, although
still upset, seeming to derive some solace from contact with her. The
babies’ responses to her efforts to get them to play were mixed.
When the mother returned the second time, the great majority of
babies greeted her in some way, often with intensified crying. More than
twice as many babies achieved contact with mother within fifteen sec-
onds of her return this time than in the previous reunion, and almost all

of them were picked up at some point, as indeed the instructions to the

mother had recommended. Interestingly, almost half the babies also

avoided the mother in some way during this episode, most prominently
by turning away from her.
Since she and her colleagues had observed all these children in the
home, Ainsworth knew that some of them were notably less secure than
the others. How, she wondered, would their insecurity reveal itself? Two
150 The Assessment of Parenting Style

main variations emerged. Much as Ainsworth expected, one group of

insecure babies tended to become extremely distressed in the separation

episodes. Then, when they were reunited with their mother, they seemed

to be going two ways at once. They eagerly sought contact with her, but

also showed unmistakable signs of anger and resistance, often kicking her

or arching away from her embrace, squirming and crying all the while, so
that they were unable to derive any comfort from the contact.
A little girl wedl call Suzie was such a child. Suzie is inconsolable
the second time her mother leaves the room, and even the
strangers

return and soothing ministrations cannot prevent an almost

panicked

level of crying. Now Suzie hears mother’s voice outside the

room and
stumbles toward the door.
Mother picks Suzie up at once, wipes her tears, and offers comforting
words. Suzie does not mold to mom or fully embrace her. She remains
fussy, keeps crying, squirms. Mom puts her down, picks her up again,
the while Suzie still crying.
wipes her hands with a paper towel; all is

“What’s the matter?’’ mother says. “What’s the matter?’’ Still crying, the

baby is struggling wildly and kicking at the same time. Again mom puts

her down, then picks her up. “Suzie!’’ she says, her tone both imploring

and scolding.
This impossible struggle between mother and baby continues for the
full three minutes after mother’s return. At certain
moments Suzie seems
on the verge of being comforted, and for an instant mother and baby
actually seem attuned, as Suzie allows herself to be cared for while she

cries. But suddenly her arms drop off her mother’s shoulder and her body
slumps. She has broken the connection, and mother is again exasperated.

Four times the mother exclaims, “Suzie?’’ She tries to bounce her on her
knee, to interest her in the world outside the window. But Suzie s cranky

and irritable crying persists.

Ainsworth called such insecure babies “ambivalent’’ (or “resistant’’).

They were extremely distressed by the separations and eagerly wanted
their mothers hack, and yet resisted them at the same time. Their reac"

tions were in keeping with Ainsworth’s anticipations, based on the inse^

cure behaviors she’d seen in the home. “1 expected that some children

were going to he more insecure than others, that they would cry harder,
that they would cry more promptly, that they would cry longer, be harder
to comfort, and that they would be angry. The thing that blew my mind
was the avoidant response.
 The Strange Situation 151

* * *

Led by mom and squawking, a little fellow we’ll call Donnie enters the
room. He is an adorable, squiggly^haired toddler, who appears vital and
engaging. His mother, as she was instructed to do, shows him the toys.
“Look! Look!” she says, and then, “Go ahead. Go! Go!” The baby takes
her cue and explores the toys, while mom, as instructed, retires to a

nearby chair and pretends to read a magazine. He is very still as he exam^

ines the toys, looking back over his shoulder repeatedly to check on his
mother. This is typical attachment behavior. He pauses, he looks around
at the room, he vocalizes a little, and gradually settles in.

The stranger, a young female graduate student, enters and says, “Hi!”
Donnie cranes his neck and watches her sit in the other chair, then goes
back to the toys. He repeatedly looks up to keep tabs on mom. He con^
tinues to play as mother and stranger talk quietly, as called for in the prO'
tocol. Five minutes into the assessment, he remains in the same position,
seated on the floor with legs outstretched before him.

The stranger sits down to play with him. He is easily engaged. He

smiles, vocalizes, seems delighted with the mutual play. After a minute,
mom rises and walks quietly toward the door. Donnie sees, turns, starts

stretching into a crawl as if to follow, but it’s too late. She’s gone. He
turns back to the stranger inquisitively, then bursts into tears. The young
woman tries to comfort him. “I know, that’s okay,” she says in a soft,

rock^a'bye voice. She reaches out and lifts him into her arms. He’s still

crying loudly. He does not protest the stranger’s embrace, but he contin^
ues to cry, and she continues to try to sooth him, physically and with
words. Gradually, she tries to reengage him in play.
The crying abates as little Donnie gets interested in the toys. He’s on
her knee now. She has one arm around him, the other free for play. He
accepts a ball from her and cooperates by trying to drop it into a hole in
a container. Eighty seconds after mom’s exit, he is fully engaged in play,

making only occasional little gasps, which seem to be remnants of his cry
and a reflection of some ongoing distress. The stranger separates herself
physically as he continues to play. He looks up at her and then starts
playing on his own. She’s behind him now and her supportive talk has

tapered off. He’s fully involved in the toys.

Donnie looks briefly at the door when he hears his mother’s voice. A
moment later she enters. Their reunion is not especially passionate, but
he is soon in her arms, and shortly afterward she returns him to the toys.
152 The Assessment of Parenting Style

Nothing seems to he amiss in any of this. But let’s pause here and play
easily be
the reunion back again. Because certain details, which might
overlooked, are telling.
When mother first enters, Donnie seems about to crawl toward her,

and he lets out a little cry. He stretches out an arm in her direction. She

stays by the door and says, “Come on, come on,” urging him toward her.

He crawls to her feet and, as he arrives, does something that doesn’t

seem

to fit. In fact, an untrained eye will be tempted to overlook it and forget

it happened. Arriving at his much missed mom, Donnie does not look up

at her and demonstrate a desire to be lifted into her arms. Rather, he

looks to his right, where one of the toy eggs he was playing
with has

rolled. reaches down and puts her two hands in front of him palms
Mom
up, and Donnie then willingly puts a hand on each of hers to accept a
lift. But then again something odd, something
like a word whispered out
her
of place in an otherwise coherent sentence: As Donnie rises in
embrace, his eyes do not meet hers and he does not smile at her. Rather,
he looks to the side, away from her, and then back in the direction of the
stranger. Into these few seconds of reunion behavior Ainsworth will read
a great deal.
Mother talks to baby sottly and walks him back to the playthings,
putting him down as the stranger quietly leaves. Still holding on to

mother’s hand, as she kneels alongside of him, Donnie gives a small yelp
and seems interested in the toys again. A few minutes later, as Donnie is

happily reengaged in play, mother quietiy rises and leaves. Donnie

shrieks and immediately follows her to the door crying. His way barred,
he breaks into a loud, hopeless cry. Twenty seconds later the sympathetic
stranger returns and immediately reaches down to pick him up. He seems
slightly relieved hut keeps crying loudly. She takes him to the play
area

and winds up a musical toy. He still looks distressed but stops crying. The
musical toy engages him. He glances at the door. The stranger and

Donnie stay in this position for a minute and a half. She has her arm
around him as he sits on her knee and she shows him the toys. He
remains interested and accepts the stranger, but his chest is heaving and
he is obviously still upset. For a moment he gives her a smile. He takes a
hall from her and tosses it down.
The stranger hacks away from him while trying to keep him interested

in the toys. He successfully drops a hall through a hole: “Good hoy!” she
says and claps. He’s absorbed in play again, still gasping a little, and look'

ing up frequently at the stranger.

 The Strange Situation 153

A minute later he hears mom and turns to see her reentry. He vocah
izes excitedly. He seems about to stand and go to her but doesn’t. This is

another telling moment. Donnie turns back to the toys. He looks again
toward mom. The stranger leaves quietly while Donnie resumes explo-
ration. He glances toward mom a third time. He’s playing quietly now
and seems absorbed, just as he was when the process started twenty min-
utes ago. Another whole minute passes before he looks again at his
mother. He seems totally calm as the Strange Situation procedure comes
to an end.
Donnie is a typical avoidant baby.
What was striking about babies like Donnie was how good they
looked — in Donnie’s case, adorable, accessible, resilient — if you ignored
those peculiar moments when their behavior seemed strangely off, when
they sought so little solace despite their distress. Such a baby seemed, as

Ainsworth said, “a psychologist’s dream baby” — competent, indepen-

dent, unperturbed. Some of the one-year-olds whom Ainsworth rated as

most secure were so distressed when left alone in the room that they
could not even approach their mother when she returned but had to
resort to sending her signals, like reaching pathetically with an out-
stretched hand, as if they were tiny infants again and unable to crawl or
walk. But some of the babies in the avoidant group went through the
entire twenty minutes with barely a ruffle. Even those who were evi-

dently upset by the separation and searched for the mother when she was
gone did not cling when picked up by her, and the few who did released
her after a moment. Except for the fact that they occasionally hit their
mothers at unexpected times — swatting her while smiling or giving a
brief hostile kick in the midst of play, without any indication of mood
change — these babies were amazingly blase.

The blase response of the avoidant babies was such a shock to

Ainsworth because of what she’d seen in the home. At home the secure
babies generally cried infrequently, weathered small separations without
great distress, and greeted their mothers positively upon reunion. They
were inclined to seek close bodily contact, to enjoy it once they got it,

and to happily give it up when released. By and large they did not display

a lot of anger. Ambivalent babies like Suzie, on the other hand, tended
to cry a great deal at home, to become markedly distressed during separa-

tions, to act disturbed in various ways when being held, and to be more
angry. Astonishingly, the avoidant babies were just the same; indeed the

Donnies were indistinguishable from the Suzies in the home environ-

154 The Assessment of Parenting Style

ment. These children, who seemed indifferent to their mother’s comings

or goings in the lab, even to the point of ignoring her on reunion who
looked so extraordinarily independent — had appeared quite insecure in

the home. They cried and showed as much separation distress and anger

as the other insecure babies. Their cool indifference in the Strange Sit'
until
nation was reminiscent of adults who seem like pillars of strength
one realizes that they are cut off from their own feelings.

Ainsworth was familiar, of course, with all the attachment^related lit'

erature. She knew about the early history of the young thieves those

who had been sent to foster care after forming attachments to their
mother and then icily avoided her when she visited, and those who came
home after lengthy hospital stays and showed no interest in their mother
or in being cared for at all, like Derek B. who called his mother nurse.
She was aware of little Reggie in Burlingham and Freud’s wartime nurS'
ery whose nurse had left him after having taken care of him for two and a
half years, and of Reggie’s snubbing her when she visited. (“My very own
Mary'Ann! But 1 don’t like her.’’) She was aware of the children in

Robertson’s studies, like Laura, who after a short hospitalization refused

to take her mother’s hand upon leaving, and who after a second short

separation some months later refused to even recognize her mother for
two days (“But I want my mummy!’’). And, of course, she was aware of
the children Robertson had observed in the long'term tuberculosis wards
who, after going through a stage of protest and then despair, had prO'
ceeded to an eerie detachment in which they no longer seemed to know
their mother when she visited or to show concern when she left.

Ainsworth now had a flash association: “The behavior of these

avoidant one'yeat'olds, 1 realized, was similar to the older child who has

had a long depriving separation and comes home and ignores mother.
Here were these kids who had never had a serious separation behaving
just that way.’’^^ The avoidant response suggested not only that both the

infant and the older child had experienced a similar sense of rejection,
hut that they were using the same defense to cope with it when placed in
a situation of heightened stress —
an emotional cutoff that disguised their
hurt and anger, even from themselves. Ambivalent children, in contrast,
although also angry, had not crossed over into this protective state of
indifference. They still hoped for solace and connection; hut their anger

spoiled the possibility of getting it. Ainsworth’s ability to isolate and

describe these reactions and their causes implied that she had hit upon
the thing that Bowlhy had only dreamed of — a procedure to assess the
 The Strange Situation 155

effects, not of drastic separations and loss, but of what he had called in
his WHO report “the partial forms of maternal deprivation.” She had
opened a research window onto the quotidian details of parenting.
For the Baltimore study Ainsworth had developed four scales to rate a
mother’s way of being with her baby: How often was the mother sensitive
to her infant’s signals? How much acceptance of the baby did she demon-
strate as opposed to rejection? Did she cooperate with the baby’s desires
and rhythms or did she tend to interfere, imposing her own schedule and
her own pace when feeding, handling, or playing? And finally, how avail-

able to the baby was she, and how often did she ignore it? With this

degree of specificity, she hoped to get beyond the vague concept of

“maternal care,” which she had come to regard as useless. As it turned
out, a comparison of the mothers of the secure babies with the mothers
of the insecure babies on these scales helped explain the puzzle of anx-

ious attachment,* which had eluded her in Uganda.

Mothers of securely attached children were significantly more respon-
them up when they cried,
sive to their infants’ signals, quicker to pick

inclined to hold them longer and with more apparent pleasure. They
were rated much higher in sensitivity, acceptance, cooperation, and
emotional accessibility. In these four scales of maternal behavior seemed
to reside the key to secure attachment, the key that she’d been seeking
since her days with Bill Blatz in Toronto. The mothers of the two inse-
cure groups rated equally low on all four measurements, the main differ-

ence being that, while the mothers of the ambivalent children were
often maddeningly unpredictable, the mothers of avoidant children were
substantially more rejecting. It was apparently the experience of rejec-

tion and the development of strategies to cope with it that gave the

avoidant children their appearance of precocious independence when

they had to deal with a stressful situation in the lab.

The behavior of the anxious mothers ranged from mean-spirited to

merely cool, from chaotic to pleasantly incompetent. Many of these
mothers were nice people and well-meaning parents who took pride in
their babies and had various means of expressing their love. Some were

good playmates or teachers; some were delighted by the positive qualities

*Ainsworth initially used the term “insecure attachment,” hut Bowlhy, trying to keep attachment
theory consonant with analytic concepts, preferred “anxious attachment,” which also made sense
given the anxiety that the child associates with the attachment. Ainsworth went along with that,
but some of her students didn’t. The two terms are now used interchangeably.
156 The Assessment of Parenting Style

they saw in their children. But what they all had in common was diffi'

culty responding to the baby’s attachment needs in a loving, attuned,

and consistent way. Inevitably this problem was compounded as the
babies became more demanding and distressed, the mothers more irri-
tated and overwhelmed. The power struggles that resulted inevitably

brought out a more hostile and rejecting side of the mother’s personality.
As Ainsworth and others spent more time analyzing the data over the
coming years, a more refined picture of the three groups emerged. The
mothers of the avoidant children, it was found, showed far less emotional

expression, and, as Ainsworth’s student Mary Main later suggested, they

seemed to be rigidly containing their anger and irritation. They behaved

less affectionately when they were holding their babies, and they were
more inclined to reinforce their commands with gruff physical interven-

(Almost half of them handled their babies roughly on occasion,

tions.

compared to only 8 percent of the mothers of secure children.) Some of

the avoidant mothers. Main would write, “mocked their infants or spoke

sarcastically to or about them; some stared them down. One expressed

irritation when the infant spilled an imaginary cup of tea.’’^^ The moth-
ers of secure and ambivalent babies did not behave this way.

Mothers of avoidant infants sometimes spoke of their dislike of physi-

cal contact and could be quite rejecting on this score (“Don’t touch

me!’’^®). Indeed, as Main reviewed the videotapes of avoidant mothers

and babies, she sensed that many of the mothers had an aversion to phys-
ical warmth, so that that their infants eventually do not respond to most
efforts to hold them. They don’t cuddle or cling, and when held, they

tend to go limp like a sack of potatoes. Late in the first year, their efforts

to make physical contact with their mother are often limited to reaching

for her arm or her foot. (None of this, it was found, was at all related to

infant cuddliness at birth.)

Further analysis of Ainsworth’s data revealed that the mothers who
responded quickly and warmly to their babies’ cries during the early

months of life not only tended to have securely attached habies at the
end of the first year but babies who cried less as well. Instead of crying,

these one-year-olds tended to use gestures, facial expressions, and vocal-

izations to get their mother’s attention. Similarly, the children of moth-
ers who had been most affectionate in holding their babies and who had
given them a lot of physical attention tended to seek less physical con-

tact at twelve months than the other habies. As Ainsworth would later
write, those infants, because they had been so consistently responded to.
 The Strange Sittiation 157

seemed to have developed confidence in their ability to control what

happened to themd^ Her findings seemed to disprove conclusively the

advice that many psychologists and pediatricians had been giving to

American mothers for decades — that they should not reinforce their
babies’ crying by responding to it. Such admonitions now seemed apprO'
priate only for parents of babies who had already been spoiled or for lim^
ited purposes like sleep training when the baby was already several
months old.

Feeding, too, revealed some sharp differences. Mothers of securely

attached babies tended to feed them promptly on demand or to hold
them off with gentle coaxing. They let the baby have some of the initia^

tive when feeding on solids and were flexible in mixing solids and fluids.

Less attuned mothers did not adjust well to the baby’s pace, they overen-
larged the bottle’s nipple hole causing the baby to struggle and gag, and
pushed solids on the baby in such a way that they ended up with a battle
on their hands.

In face^tO'face interactions mothers of securely attached children

tended to be more skilled than the others in adjusting their own pace and
behavior in accordance with the baby’s cues. Not surprisingly, the moth-
ers who smiled and spoke to their babies got more joyful responses, replete
with bouncing and vocalizations, than those who merely looked at them.
Ainsworth had now clearly established that sheer quantity of maternal

care was not the issue. On the whole, for instance, mothers of secure
babies did not hold their babies appreciably more than other mothers.
But there was a difference in the way they held them: They were much
more affectionate, tender, and careful, rarely inept, rarely distressing

their babies with an unpleasant experience of physical contact. And they

held their babies when the babies wanted to be held.^^
As Ainsworth analyzed the twenty-three babies who went through the
first Strange Situation (three of the original twenty^six babies were not
assessed due to illness or age), she was not satisfied with the three main
categories of secure, ambivalent, and avoidant. The babies’ behavior was
too varied. She divided each insecure category into two subgroups and
the secure babies into four subgroups.* Mary Main, who was there at the

*The anxious subgroups basically show different degrees or styles of avoidance or ambivalence.
Some avoidant children completely snubbed their mothers on reunion, some showed some initial
interest in contact. Some ambivalent babies were more angry, some more passive in their resistance.
The secure subgroups allow for cases where a child is basically secure but shows some signs of avoid-
ance or ambivalence.
158 The Assessment of Parenting Style

time, latercompared Ainsworth’s rigorousness to that of Nobel Prize-

winning geneticist Barbara McClintock, who was determined to explain
the pattern ot kernels on every ear of corn. Mary was like that, too.
Main “She had to explain every bahy.”^^ To Ainsworth’s own
said.

amazement, the eight groupings she created would hold up for twenty
years and thousands of children.
In the history of psychology’s move toward an understanding of the
impact of relationships, no one before had come up with a method of
There were a myriad of procedures for assessing the
assessing relatedness.
individual and new ways of diagnosing, describing, and categorizing him
were repeatedly being devised —
but no one had found a way to assess
how style of parenting contributed to individual differences. “1 did not

intend this as a way of assessing attachment,” Ainsworth says, hut it cer-

tainly wound up as that. We began to realize that it fit in with our

impressions after seventy-two hours of observation in an amazing way.

But instead of seventy-two hours of observation we could do a Strange
Situation in twenty minutes. What’s more, researchers could use her
empirical base to launch new studies, investigating aspects of relatedness,

how they are internally organized, how they are transmitted, how they

affect future behavior. Through this ingenious project, capping years of

research, Ainsworth had begun a revolution.

Ainsworth’s discovery of attachment patterns would prove a bonanza

to Bowlby. From the days of his supervision by Melanie Klein, when she

had refused to consider the family relationships that contributed to the

disturbance of the children he was seeing in treatment, Bowlby had
wanted to show the world that early parenting was crucial in emotional

development. He studied separations, not because he considered them

more important than other damaging experiences of early childhood but
because they were easy to research. No one could say they hadn’t actually
occurred. And once he got into the research, much like Freud’s early

grappling with hysteria, he found opened many theoretical doors for

it

him; so he pursued it as far as he could. His name became synonymous

with early separation. But the big game in Bowlby ’s mind had always
been the parent’s attitude toward the child, the parent’s everyday behav-
ior, be it loving, rejecting, or some mixture of the two. Ainsworth’s idea,

of course, had been to do longitudinal research, like James Robertson.

And indeed she was a pioneer — it may be fair to say the primary pioneer
— in the observation of children in the natural environment. But
Ainsworth had been interested first in emotional security and its sources.
 —
The Strange Situation 159

And suddenly, almost as if by accident, the convergence of her interests

had produced the very result that brought Bowlby’s work full circle. She
had gotten him, and attachment theory, into the home. For the first

time, something could be said with scientific accuracy about the emo'
tional impact that parents’ everyday behavior has on their young.

The impact on Bowlby’s status would be immeasurable. Until now his

stature rested mainly on his work in maternal deprivation; and as a result

of that work he was better known in Britain than most other psychoana^
lysts and fairly renowned abroad as well, though not in the United
States. But his standing as a developmental theorist was modest. When
his first volume of Attachment and Loss appeared in 1969, there were
some who believed he was successfully building a sounder metapsychol-

ogy — or overarching theory — which would one day replace other ana-
lytic models. Some of those who had worked on his Tavistock team
Robertson, Schaffer, Heinicke, Ainsworth herself — had done valuable
research that lent support to his views. This empirical strength was more
than almost any of the other analytic theorists could boast. But nothing
resounding had happened to set him apart until Ainsworth’s surprising
Strange Situation, which applied the idea of separation and reunion in a
wholly new way.
“The fact that the Strange Situation was not in the home envi-

ronment, that it was in the lab, really helped,” Ainsworth later said
with a laugh. “I only did it as an adjunct to my naturalistic research, hut

it was the thing that everyone could accept somehow. It was so demon-
strable.”^^

For the next twenty years, Ainsworth would be occupied with the tall-

out from this work. Because she and her co-workers had made such
painstaking descriptions of each mother-infant pair, the statistical analy-

ses would take years to work through. Meanwhile, she would be training
others to use the technique, supervising and co-piloting new research,

writing articles, teaching, giving workshops, and serving as the leader of

a growing attachment community.
Of the Baltimore study, she later said, it “turned out to be everything
that 1 hoped it would be, and it has drawn together all the threads of my
professional career. Each piece of data analysis we did with very few

exceptions had some sort of hang to it. It was always such a pleasure to

find things working out, and we had an awful lot of things work out.”^"^

The constantly appearing evidence, meanwhile, was more raw material

for Bowlby’s grand synthesizing machine. It was fed into his three vol-
160 The Assessment of Parenting Style

umes of Attachment and Loss, which gradually made their way into public

cation between 1969 and 1982.

Ainsworth’s work also held the potential of bringing attachment the^
ory closer to its psychoanalytic roots. For what Ainsworth called “secure
attachment’’ could also be understood as having something in common
with healthy narcissistic development, a growing concern among ana^
lysts at that time, especially the followers of Heinz Kohut. They saw the
infant as needing to be at the center of a loving caretaker’s universe and
that those who failed to get such consistent and dependable adoration
would be at risk for developing a narcissistic personality disorder later
(which typically includes a shaky sense of selTworth and clinging to an
infantile need to be applauded and ministered to). One could even imag'
ine the ambivalently and avoidantly attached children as representing in

some way different variants of narcissistic disturbance.

Ainsworth’s findings had the potential of promoting a rapprochement

between Bowlby and Melanie Klein, too. The anxious children could,
without great effort, be seen as struggling with some of the issues — para-

noia, persecution, rage and its destructive effects — that Klein had expli-

cated. And Ainsworth’s secure base, which became internalized in the

child as secure attachment, is not unlike Klein’s view that the baby inter-
nalizes the good breast, or, ultimately, a warming sense of mother love,

and that this activates the baby’s pre-existing love-attuned circuity that
cause it to see itself as good and the world as a good place.*
Ainsworth, of course, framed the issues differently from either Klein or
Kohut, speaking not about narcissism or internalization, but whether the
developing child had faith in his mother’s love, in his ability to get care
when he needed it, and felt free to explore. But this evolving attachment
view, although it added new dimensions, was hardly incompatible with
those of her analytic contemporaries. In fact, much would be gained from
free and regular interpollination, not to mention a full re-integration of
attachment theory into the analytic fold. But that would not begin to
occur for some twenty years, and then only slowly.

*Ir may He Bowlby ’s problem with klein’s point of view as it was originally formulated
recalled that
was that the baby’s capacity to experience a good breast, and therefore a sense of himself and the
world as being good, was not particularly dependent on the quality of mother love. The baby already
had the love and the hate within him and he applied it to the suckling experience according to his
own feelings of satisfaction or dissatisfaction, which might have little to do with the mother. But by
the time Ainsworth was enunciating her ideas, Kleinians were more prone to take parental qualities
into account.
 The Strange Situation 161

Indeed, it would be a while before almost anyone would realize the

importance of what Ainsworth had done. The findings of the Baltimore

study, conducted from 1963 through 1967, did not begin to appear in

published form until 1969, and her book Patterns of Attachment, written
with several collaborators, was not published until 1978. The Strange
Situation procedure, meanwhile, could not easily be learned from a man^
ual, and developmentalists had to go through training to master it. It was
sometimes difficult to recognize the avoidant response, partly because
the avoidant children often look so good — independent, engageable,

emotionally robust — that an observer might tend to rationalize or over^

look the peculiarities of their reunion behavior; while some of the

ambivalent children could also be quite hard at first to distinguish from
the secure. It was hard for the untutored eye to pick up on the subtle cues
that led a child to be placed in one subcategory or another and easy to
overlook Ainsworth’s strict precautions about not overstressing the child
if a separation proved too much for him. The studies that Ainsworth’s

students conducted, which supported and extended her work, also took
time; they did not start seeing print until the late 1970s. Beyond all that,

scientists are cautious, new ideas are slow to catch on, and the attach-
ment ideas turned out to be especially problematic for some, offending

the reigning theorists and threatening others by calling specific parent-

ing styles into question. Even Bowlby at first took Ainsworth’s work in
stride. As he himself would say, “I hadn’t yet seen the payoff.”
 Second Front:
Ainsworth’s American Revolution

“1 got interested in the field because I went to her lectures,” says Inge

Bretherton, who was a thirty'three'year'old undergraduate, returning to

school after raising children, when she first heard Ainsworth speak in
1969. “I thought. Oh, here is somebody who’s studying real children in

real environments. Almost nobody else was doing that at the time. Back
then, everybody was a behaviorist. You couldn’t talk about the inner life,

so to speak, or the internal world. Not in developmental psychology. I

had gone to lectures in Cambridge where every time the person talked
about consciousness, he made quotation marks in the air. That was the
sort of climate in which this developed.”^

Bretherton, now a leading attachment scholar and theorist, was just

one of many bright students Ainsworth began attracting at that time,

people who would carry attachment work with them to other universities

throughout the seventies and eighties.

Everett Waters was an undergraduate in chemistry at Johns Hopkins
when he met Ainsworth in 1971. Looking for a summer job, he told her
that he would be willing to start out as a volunteer. “She said, ‘Whoa-ho,
a volunteer!’ And she puts her arm around me and walks me down the
 Second Front 163

hall to Mary Main, who’s doing her dissertation.”^ Waters soon aban^
doned chemistry and entered the psychology doctoral program at the

University of Minnesota in 1972. There he met a young assistant profes-

sor named Alan Sroufe, a soft-spoken and deliberate man with a com-
manding presence who over the next two decades would become one of
the leading figures in developmental psychology.
Sroufe, who had only a passing awareness of attachment theory, had
recently read a major paper by Ainsworth on attachment and dependency
and was drawn to her ideas. He was thus particularly receptive to Waters
enthusiastic recounting of Ainsworth’s work. The Strange Situation

struck him as a potentially powerful tool, and he immediately got Waters

all the lab space and money he needed to replicate the procedure in order

to study whether an infant’s attachment pattern would prove stable over

time —an important question to anyone who, like Sroufe, had been
trained in behavioral measures. Sroufe was equally impressed with the
methodology Ainsworth had used in the longitudinal end of the study, in

particular the various means she adopted to describe the maternal care

she was observing in the home. He recognized that there was something

new here, something that might alter the trajectory of his career.
Sroufe soon became the first significant “independent” to join the

attachment group. He was neither a disciple nor student of either of the

principals and was thus able to enrich attachment thinking with a

wholly different orientation of his own. His placid face, somewhat remi-
niscent of the singer John Denver’s, and his measured manner belie
Sroufe ’s passion for ideas and for disseminating them. He had been a

preacher for a small Protestant group as a youth, and some of his ideolog-
ical opponents would find him overly dogmatic in promulgating the new

attachment theology. But if this quality occasionally seemed a flaw, it was

also an aspect of an extraordinary gift. For Sroufe has a talent that is rare

in academia: He has the capacity to speak and write powerfully and

accessibly, one idea building fluidly upon the other, illustrated by telling

examples, and buttressed by a ready access to mounds of data. His 1977

article, “Attachment as an Organizational Construct,” written with Ev-

erett Waters,would awaken many developmental psychologists to what

attachment thinking was all about. Sroufe would also help seed psychol-
ogy departments across the country with attachment scholars, becoming
a mentor new generation of developmental psychologists. One of
to a

them, Dante Cicchetti, now a leading developmentalist in his own right,

would later recall the “electric atmosphere” of his Minnesota years.
164 The Assessment of Parenting Style

As Sroufe examined Ainsworth’s material, he became convinced that

she had achieved what no one had achieved before: the ability to study
the complexity of relating in a way that could be quantified and used by
other researchers. His enthusiasm for what she’d accomplished derived
partly from his own history of difficulties with the research methods that
preceded hers. His verbal mastery is evident in the way he discusses the
originality of Ainsworth’s work and the revolutionary potential he saw
in it: “In the past, developmental psychology thought there were two
ways of doing things —you either counted discrete behaviors, or you

did global ratings. The problem with discrete behaviors is, while they’re
very reliable —you can code them very accurately and you can get every'

one to agree, yes, he did seven of them, and he did four of those — it takes

a tremendous amount of observation to get anything that’s worth'

while, and it’s hard to know what they mean. They’re very situationally
influenced, very unstable over time, and of quite limited validity. To
know that one mother picks up her kid more than another mother
does, or that one child talks to other children more than a second child
may tell you something, but
does, that it probably doesn’t. But if you
know somehow that one child is able to emotionally engage another

child and the second child isn’t, that tells you a lot. Now how the heck
do we measure that? It turns out counting frequency isn’t an especially
good way.’’

The problems that arose when trying to squeeze meaning from

counted behaviors had caused some researchers to favor what was called
“global ratings,’’ which allowed an observer to make an overall judgment:

how sensitive is this mother? how sociable is that kid? “But,’’ Sroufe says,
“global ratings always had the reputation of being subjective and unreli'

able. People couldn’t agree. Well, Ainsworth’s methodology is actually

neither of those.
“She has one scale called Cooperation and Interference. On one end,
the cooperative end, these parents fit what they do to the child. They do
things in a timely manner, they do things when the child is open to

them, they don’t do things at cross purposes to the child. On the other
end, interfering, the parent is coming in doing things when the child
isn’t ready. Think of a feeding situation. You’re feeding the child, he
turns his head, and you cram the food in his ear. This would not be a
cooperative behavior.
“So now you’ve got both things. You have ratings that are reliable and
are based on behavior. But they’re broader and more integrative and
 Second Front 165

more attending to the meaning of behavior. Ainsworth showed that

mothers of babies who are later avoidant hold their babies as much as
mothers of babies who later are secure. So if you just measure frequency
of holding you get no difference. But there’s one circumstance in which
mothers of babies who are later avoidant do not hold them, and that’s

when the baby signals that it wants to be held. So you could have
counted a lot of holding and you would have gotten nothing.”
Sroufe credits Ainsworth with enabling developmental psychology to
recognize and measure the qualitative aspects of behavior, not just the
mother’s but the child’s as well, so that in the Strange Situation one does
not merely observe what the child does but how his behavior is orga^

nized — that is, what central themes and dynamics underlie it. “It isn’t

how much the baby cries,” Sroufe says, “it isn’t how much the baby looks
at the mother, it isn’t how much contact the baby seeks; it’s when it does
those things and with what other behaviors and to what end.”
Sroufe now adopted Ainsworth’s methodology to his own work: assess^
ing such qualities as aggression and dependency in older children — in par-

ticular, children who had been through Strange Situations when they
were a year old — in the hopes of seeing whether early quality of attach-
ment would predict later psychological tendencies. “If you want to define

aggression as one child comes up to another child and juxtaposes his

hand, which is in a closed position, to the other child’s body, that’s all

well and good, but if somebody goes, ‘Hey, babe!”’ —Sroufe throws a
friendly punch at an imaginary shoulder
— “that isn’t aggression. And it

can also be aggression if you just go up and stand next to the kid and
never hit him. You can keep redefining your behaviors to rule one thing
or another out, but it’s so much easier to let a coder attend to the meaning
of the thing and take into account the various cues that are available.”
In Ainsworth’s methodology, however, Sroufe saw a shift not just from
counting behaviors to assessing a deeper quality ot the child’s psycholog-

ical organization but also from studying the child in isolation to studying
the bond between mother and child. “This movement of psychology

from studies of individuals as units to thinking of relationships as units,

that’s a profound change in the field. And here’s Ainsworth who comes
along and has a way of conceptualizing and assessing a relationship. The
thing that’s profound about her assessments, both the home observations

and the lab assessments, is that she is assessing the quality of the rela-

tionship between the infant and the caregiver. Not the baby, not the

mother, the relationship.”^

166 The Assessment of Parenting Style

Waters’ study was the first of many to come out of Minnesota. He

assessed a group of babies at twelve months and then again at eighteen

and found that 48 of 50 fell into the same Ainsworth attachment cate^

gory on each occasion."^ This was a significant finding that flew in the
face of the predictions of leading developmental theorists, who assumed
that the attachment classifications were transient and situational rather
than incorporated into the child’s psychology. Most interesting from

Sroufe’s point of view was that the classifications remained the same
despite the fact that eighteen-month-olds behave very differently from
twelve^month^olds and thus manifest their security or anxiety in entirely
different ways.

Before long Minnesota was buzzing with attachment research. That

the Institute of Child Development, prestigious and centrist, had gotten
into attachment, was enormously helpful to Ainsworth, who needed the

support. For her findings were not universally accorded this friendly
reception. Behaviorist thinkers found her conclusions flawed and unac'
ceptable and lost no time pointing that out.

In 1969 the world of psychoanalysis in which Bowlby operated and the

world of developmental psychology that had become Ainsworth’s home
were almost completely foreign to each other. Although psychoanalysis
had a theory of human development, it was mainly a clinical discipline

dealing with patients. Its concepts of psychological progress from infancy

through childhood and adolescence —whether Sigmund Freud’s, Anna
Freud’s, Melanie Klein’s, or Erik Erikson’s —were based largely on infer'

ences about what the developing child must have experienced at various
stages. These inferences were first built on clinical experience with adult
patients — their fantasies, their memories, their feelings when in re'

gressed states. They were later reinforced by clinical experience with

children. But they were almost never based on the direct observation of

infants or of children outside the consulting room.

Developmental psychology, on the other hand, was strictly an acadc'
mic discipline, centered in the psychology departments of universities,
and interested in the actual experiences of normal children. Develop'
mental ists had wanted to break away, from the testing and measuring of
groups of children, which had been the purview of child psychology, in
favor of observing children in the natural environment and discerning
the laws of maturation, stages of their growth, and how early experience

might affect later behavior. Their special tool was the longitudinal study.
 Second Front 167

which was providing numerous insights into the processes of normal

development. Ainsworth herself had come out of a developmentally ori-

ented program in Toronto.

By the late sixties, however, when she was beginning to make her
views known through American publications, the dominant ideology of
the psychology departments where developmentalists did their teaching
and research was behaviorism, and, in particular, learning theory. Unlike
psychoanalysts, behaviorists were not concerned with what the infant
felt or any other aspect of its internal experience. They focused mainly
on behavior and learning — aspects of the child’s being that were more
readily accessible to direct observation; and counting behaviors was the
standard means of research.
Behaviorism had grown out of the turn-of-the-century work of Edward
Thorndike and Ivan Pavlov, and later James B. Watson, who first coined
the term and who believed that psychology should be a hard-nosed,
empirical science. Watson rejected the value of introspection — a tech-

nique for exploring mental functioning that had been acceptable in psy-
chology up to his time —denounced the concept of inherited personality
traits, and insisted that all behavior, animal and human, was the result of

environmental stimuli. Watson also belittled the concept of conscious-

ness as unscientific and saw mental processes as bodily movements.
Thoughts in his view were merely subvocal speech.
Watson had been impressed with Pavlov’s experiments in Russia, in
which a dog was made to salivate at the sound of a bell because the ring-
ing of the bell occurred repeatedly just before the dog was fed. This con-
ditioned response became the core of classical behaviorism. Watson
promoted much important experimentation on conditioned responses,
perhaps most notoriously, conditioning an eleven-month-old child.
Little Albert, to become terrified of cuddly animals. This was achieved
by sounding a loud clang over his head each time he reached out to
touch a small white rat. Classical behavior therapy assumed that if peo-
ple could be maladaptively conditioned, they could be deconditioned

and thus made well. This approach sometimes proved effective, espe-
cially with phobias like the one Watson induced in Little Albert (al-

though, unhappily, Albert’s mother took him away before he could

benefit from the reverse treatment).
One of behaviorism’s most brilliant modern adherents, B. L. Skinner,

believed that almost all behavior, from language acquisition to attach-

ment, is shaped by rewards. It has little to do with inborn tendencies,

 —
168 The Assessment of Parenting Style

way such things might mix with experience to

capacities, or needs or the

produce unconscious fantasies and desires. If a hen happens to peck a

disk or if a rat happens to depress a lever and each then gets a pellet of

food, and if this happens consistently, the animal will eventually come to

associate the behavior with getting fed. Much the same happens with
more complex and subtle way. Experiments
people, Skinner said, but in a
with animal conditioning, many of them done in the famous Skinner
box, have enabled researchers to shape animals’ behavior in astonishing
ways, such as teaching chickens to play Ping-Pong. Such feats demon'
strated not only how potent a force rewards are but also how far afield

they can take an organism from its instinctual behaviors, accounting, in

human beings, not only for complex athletic and scientific endeavors but
for neurotic distortions as well. What Skinner called “operant condition'
ing’’ became an important tool of behavior therapists. In child treatment,

for instance, the rewards could be systematically applied by the therapist

and taught to parents or teachers in order to reshape the child’s behavior.

Beyond that, Skinner’s behaviorism was also a philosophy for the ratiO'

nal management of society, and throughout his career he offered sugges'

tions, sadly unheeded for the most part, for ways in which public policy
could make use of rewards to improve social functioning without need'
less coercion.
For Skinner and other learning theorists the forces of conditioning
explained it all. Speculation about what went on inside the individual
his emotions, his insights, his imaginings —was needless, useless, and det'

rimental to science. They were similarly unimpressed by the theories

about biological maturation, which suggested that certain biologically

timed abilities clicked in at certain ages. Development, they said, was
not a succession of stages, as Arnold Gesell had argued, but a smooth
progression based on the gradual impact of the environment. To behav'
iorists, the child was almost completely malleable, was inclined to seek
pleasure and avoid pain from the first moments of infancy, and was

exquisitely attuned to the subtle conditioning forces that most of us

overlook.
Bowlby and Ainsworth did not deny the importance of behavioral
i-heory —
Bowlhy had made it a key'element of his system. But they saw
learning as just one part in the complex web of human nature and relat'

edness. Like Noam Chomsky, who was battling Skinner over the issue of

speech acquisition, arguing that speech was not randomly learned but
something that the child was biologically prepared for and actively pat'
 —
Second Front 169

ticipated in through innate abilities, Bowlby and Ainsworth made simi-

lar claims for attachment. Attachment developed because of instinctual

needs and had little to do with the rewards and punishments that medi'
ated many forms of learning. Indeed, they argued that attachment will
develop in the face of very little reward and a great deal of punishment,
as is the case with abused children who show very strong attachments to
their parents.

Learning theory, however, saw nothing inherent in being human that

caused people to seek intimate attachments. Attachment was simply a

remnant of dependency. In the classical behavioral view, the child asso-

ciated his mother with taking care of him and eventually came to feel

that he needed her as much as food or water. In the more recent Skinner
model, dependency came about because certain dependent behaviors
crying, clinging, seeking attention, seeking praise —were constantly
rewarded by the mother. At normal levels such reinforcement accounted
for all normal human ties; excessive reinforcement, for all the negative
aspects of dependency, from clinginess to spoiledness to lack of self-

reliance. But in both views one theme remained constant: Relationships

were a secondary phenomenon and not particularly important in them-
selves.^

In the battle that developed between Ainsworth and her behavioral

colleagues, the infant’s cry became a central point of contention. In learn-

ing theory, crying is a random activity that means nothing until it is rein-

forced by a mother’s responsiveness. Like the rat that learns to depress a

lever to get a food pellet and is soon pressing the lever like crazy, an infant
has no initial reason to cry and would not continue doing so unless the
crying were reinforced. Because he learns that crying gets him what he
wants, he cries all the more. A random, accidental behavior thus becomes
a furious habit. Hence the injunctions since Watson’s day against spoiling
children with too much responsiveness: If you picked up your baby every'

time it cried, you’d soon have a ferocious crybaby on your hands. This
warning. Dr. Spock notwithstanding, remained a teature of American
bahy-rearing ideology (and, indeed, lingers on today). Bowlby ’s idea, as it

arrived on the developmental psychology scene in the Uriited States

through the articles of Ainsworth and her students, that the infai'it is

prewired for attachment and has certain inborn attachmei'it behaviors

that must be responded to by the caregiver if the child is to develop emo-
tional security — in other words, that the hahy cries for a purpose and
requires a response — struck behavioral theorists as pitifully misguided.
170 The Assessment of Parenting Style

Ainsworth’s Baltimore research was similarly incomprehensible to

them. Her statistics suggesting, for instance, that babies cried less at

twelve months if their cries had been responded to conscientiously when

they were younger flew in the face of learning theory logic. If you rein^
forced a baby’s crying, proximity seeking, and the like by responding to it
positively,you should get more of such behavior, not less. But A.insworth
insisted that warm, sensitive care does not create dependency; it enables

autonomy. “It’s a good thing to give a baby and a young child physical
contact,” she would later say, “especially when they want it and seek it. It

doesn’t spoil them. It doesn’t make them clingy. It doesn’t make them
addicted to being held.”^ This was stark disagreement —and the fact that

it had such immediate consequences for reahworld issues of child rearing

heated the debate up all the more.

Ainsworth did not deny that excessive dependency existed. But she
held, as Bowlby had, that dependency and attachment represented dih
ferent realms of relatedness, attachment being closely synonymous with
love; dependency more often representative of neurotic anxiety. “Does a
phobic wife who clings to her husband and consistently seeks his proximo
ity,” she asked, “love him more than a woman who is less neurotic and

more competent loves her husband?” The answer, she said, was not nec'
essarily. Indeed the phobic wife might actually love less than the health'
ier wife. Similarly, she asked: “Is who especially clings to his
the child

mother more attached to her than a child who clings less or is he —

merely more insecure?”^ Again, as her own studies had shown, clinginess

did not indicate greater attachment or love but greater anxiety.

Ainsworth repeated what Bowlby had been at pains to insist: that
attachment itself should not he confused with attachment behavior. The
anxiously attached child shows heightened attachment behaviors — cry'

ing, seeking, clinging —hut he is not more strongly attached. He doesn’t

love his mother more than the securely attached child. Rather, his love

(which is not weaker either) is riddled with uncertainty and anger.

Indeed, the anxiously attached group that Ainsworth labeled ambivalent
seemed on their way to becoming dependent characters, in the negative,

clinical sense, like the clingy, phobic spouse.

With the results of her Baltimore study in hand, Ainsworth could
state with assurance what her early mentor. Bill Blatz, could only propose

as theory: that the responsive mother provides her hahy with a secure

base. The infant needs to know that his primary caregiver is steady,

dependable, there for him. Fortified with the knowledge ot his mother’s
 ^ —
Second Front 171

availability, the child is able to go forth and explore the world. Lacking
it, he is insecure, and his exploratory behavior is stunted.
Although numerous methods had been devised to measure conceptual
and cognitive development — many ofthem introduced by Swiss psy^
chologist Jean Piaget — until now there had been almost no measurement
or assessment procedures for an infant’s social and emotional develop^
ment, certainly not at this level of complexity. Although it was widely
assumed that reaLlife events shaped personality, no one had been able to
come forth with evidence about exactly which events or experiences
mattered. To a few visionaries like Sroufe it seemed that Ainsworth, in a

stroke, had changed all that. But in the hehaviorist'dominated atmoS'

phere of developmental psychology in the late 1960s, her assertions
about the baby’s knowledge of the mother’s availability or about the
security to explore the world that such knowledge confers bordered on
the mystical and fell on deaf ears.
To many mothers Ainsworth’s prescriptions might seem as natural as

maternity itself. (Of course you pick up your baby when it cries!) But as

pleasing as it may he for some to discover that psychology is catching up

to what has always seemed intuitively correct to them — that little chib
dren do indeed need nurturance and consistency, that the way you are
with your baby will profoundly affect its personality development, that
what happens to it when it’s little will influence what it becomes later

it is equally displeasing to come face-to-face with a body of evidence that

suggests that you yourself didn’t or aren’t or won’t be doing it right. Her
statements were thus not only threatening theoretically hut personally as
well, for there was no shortage of developmental psychologists who were
also parents.

In any case, prominent developmentalists, most of them trained in

learning theory, attacked her on every front. They insisted that the pat-

terns of attachment that Ainsworth thought she had demonstrated

would not prove stable and, even if they did (as Everett Waters’ study
seemed to demonstrate), that they should not he thought ot as represent-

ing the child’s quality of attachment hut merely the way the child
behaved when given certain cues from the mother. Take the mother
away, and the child would look entirely different. Any other conclusion
could only mean that Ainsworth had misread her data. They turther

contended that any assessment based on twenty minutes ot behavior was

inherently flawed because the data base was too small.
During those years Ainsworth sought funds to replicate her work, but
172 The Assessment of Parenting Style

the funding agencies, while respectful of her research capabilities, replied

that there was no point replicating something of so little value. It was a

depressing development, and it fueled her own selhdoubts.

But the Strange Situation’s value did not escape a handful of infant
researchers, who saw that they had been given an extraordinary tool, a

Rosetta Stone of sorts, with which they could decipher the residue of the
infant’s experience with its parents. For the first time they had three cat'

egories of children who demonstrated distinct behavior patterns in a lab'

oratory procedure:

• The securely attached, who sought their mother when dis'

tressed, who seemed confident of her availability, who were
upset when she left them, who eagerly greeted her upon her
return, and who warmly accepted and were readily comforted

by her soothing embrace.

• The who seemed to depend less on their
avoidantly attached,

mother as a secure base; who sometimes attacked her with a

random act of aggression; who were far more clingy and
demanding than the secure children in the home environ'
ment; and who, despite in some cases being just as openly
upset by the mother’s departure in the Strange Situation,
showed no interest in her when she returned.
• And the ambivalently attached, who tended to he the most
overtly anxious; who, like the avoidant children, were also
clingy and demanding at home; who, like the secure, were
upset when abandoned by the mother in the Strange Sit'

uation; but who, despite wanting her desperately when she

returned, arched away angrily or went limp in her embrace, so

that they could not he soothed.

Once researchers had these three categories and knew what sort of care'

taking the three types of children had experienced — sensitively attuned

and reliable in the case of the secure children; rejecting and somewhat
harsh in the case of avoidant children; inconsistent or chaotic in the case
of the ambivalent children — all sorts of questions previously confined to

theoretical speculation were suddenly accessible to empirical study.

Because now children whose quality of attachment had been assessed at
twelve months could be followed and tested for other qualities as well in
order to compare how anxiously attached and securely attached children
 Second Front 173

developed. In the coming years psychologists would — sporadically at first,

then with greater frequency, and finally in a flood of empirical excite^

ment that would make Ainsworth one of the most cited authors in the
history of developmental psychology — use the Strange Situation to corre-
late attachment style (and hence the mother’s parenting style) with the
child’s character development, schoolwork, problem-solving ability, self-

reliance, self-esteem, peer relations, general sociability, and just about

every other important issue in his young life, with some stunning results.

None would he more prolific at this than Sroufe.

I

I
 PART III

The Fate oe Early Attachments

(

i
t

li

>

'•

m
 13
The Minnesota Studies:
Parenting Style
AND Personality Development

Armed with Ainsworth’s published data, Bowlhy was able to advance his
attack on the unfeeling style of child care he deplored and in particular

the assumption that a stoic, somewhat depriving upbringing built auton^

omy, self-reliance, and strong character. Bowlhy could now argue with
greater force that just the opposite was true — that self-reliance both in
childhood and later life rested not on benign neglect or stern disciplinary
regimes but “on secure attachment to a trusted figure,’’^ which was, in

turn, built on sensitive responsiveness to a baby’s needs. He predicted

that the babies from Ainsworth’s most securely attached subgroup, the
ones who had received the most consistently positive care, “would be
most likely in due course to develop a stable self-reliance combined with
trust in others.’’^ Bowlhy was obliged to acknowledge that his argument
could only be considered speculative, that the empirical data were “still

woefully insufficient.’’ But it was just this empirical data that Alan
Sroufe was equipped to provide.
As a result of Waters’ study, Sroufe now had a group of toddlers whose
attachment histories were known. His goal was to construct a follow-up
that would determine what benefits or disadvantages their quality ot
178 The Fate of Early Attachments

attachiTienthad conferred. The problem was, W^hat to look for? Certain

behaviors that were prominent indicators of welhbeing or disturbance at
one age (like failure to coo as a baby) could hardly be expected to persist
as strong indicators later (failure to coo as an adult). Indeed,
the im^

mense changes that children go through as they develop had led many in

the field to dismiss the very idea that some form of stable personality
existed from one developmental stage to the next.^ They held that there

was little if any connection between what happened in the early years

and what happened later. Sroufe believed that there were connections

and that an underlying pattern did persist. It could not be found in par^

ticLilar behaviors but rather in the quality of the child s adaptation at

each stage of life. Indeed, he and Waters had already estab'

felt that he

lished that when they determined that quality of attachment was stable
between twelve and eighteen months.
“If you think about it,” Sroufe related some years later, “the differences

between twelve^month-old babies and eighteen^month-old babies are

dramatic. In fact, eighteen'inonth^old babies are more like you and 1
than they are like twelve^month^old babies. At age twelve months
infants that have a secure attachment history are, generally speaking,

very distressed by separation, especially if they’re left alone. If a mother

leaves a twelve-month^old haby, even for two minutes, when the mother
comes back, that baby wants to be held and wants to be held for some
time. He crawls over, he’s crying, he climbs up her legs, and when he gets
picked up, he hugs in and holds on. At eighteen months, infants who
have secure attachments with their caregiver, generally speaking, are not
so distressed by separation, they don’t so much need physical contact
upon reunion, and the most dramatic change is that contact need not be
long. An eighteenmionth'old can be upset, the mother comes back in,

the baby comes over and gets picked up, couple of seconds, that s good
enough, back down to play. And in fact, many times what an eighteen^

month'old needs is interaction with the mother upon reunion, not physi^
cal contact — he’ll actively greet her, maybe show her a toy, maybe smile,

maybe take her a toy. But those relationships have the same quality in

that in both cases you see activity initiated by the infant to reachieve
contact — on the one hand more physical, on the other hand more psy'

chological. Secondly, you see that reachieved contact is effective in set'

tling the infant and returning the infant to exploration and play.”"^

Sroufe believed that if he could show stability in the quality of the

baby’s relationship with the mother, he could also show stability in the

general quality of the child’s adaptation. This would include the child’s
 ^

The Minnesota Studies 179

orientation toward others, his ways of dealing with stress, his expecta^
tions, and his overall approach to the world.
Working with Sroufe, graduate students Leah Albersheim and Richard
Arend rounded up forty^eight two^year-olds whose security of attach--
ment had been assessed by Waters six months earlier. Leaning on the
work of Erik Erikson and psychologist Jeanne Block, they expected that
children of this age should display greater autonomy, flexibility, resource^

fulness, and be able to use the assistance of their mothers without being
unduly dependent on it. To test this, a series of activities was created.
The team predicted that the welLadapted two-year^olds would become
readily involved in the play tasks, be able to persist when frustrated,

resist succumbing to self-defeating behaviors, and use their mothers’ help

when They would not automatically comply with mother:
they got stuck.
Some opposition was normal at this age, especially when asked to do
something onerous, like stop playing and clean up the toys. But they
would gradually cooperate with her, or at least try to, especially in a diffi-

cult play problem where her assistance was crucial to success. Sroufe

expected that they would share their enthusiasm with their mother and
frequently with the experimenter, too.
“The main task was well suited to children at that age,” recalls
Albersheim, now a child psychologist. “And for the most part they
responded with a lot of enthusiasm. There were two sticks, each one too
short to poke a toy out of a tube. So they had to figure out how to put
them together to make one stick. Usually kids that age will play around
with the tube and put the little sticks in, and after a while they figure out
that it’s not working the way it’s supposed to, and through trial and error
they get to where the two sticks will fit together to make one long stick.”

She still remembers how they struggled with the problem. “It’s a wonder-
ful age and a wonderful time to see kids trying to do a task that’s going to
challenge them and watch their pleasure as they see themselves succeed
at what they’re trying to accomplish. That’s a really exciting thing for a
two-year-old.”^
In almost every respect, however, the securely attached children* did

*This an inaccurate but convenient shorthand. All we know about the attachment status ot these
is

children is their assessment at age twelve or eighteen months in the Strange Situation with their

mothers. I will often use “securely attached” as a shorthand description for children with secure
attachment histories with their mothers. But it should be kept in mind that a child who was assessed
as securely attached at twelve or eighteen months is not necessarily securely attached at two years or
later and may have a different quality of attachment with another significant figure, like the father. 1

will also sometimes use the shorthand “secure” or “anxious” when referring to attachment status. But
this, too, is not quite accurate, since one can be secure materially or intellectually or in physical abil-

ity without being securely attached.

180 The Fate of Early Attachments

better. Tbey were all able to flexibly manage tbeir impulses and desires,

while few of the anxiously attached children had this capacity,

many of
them falling apart under stress. The secure group also showed
substan-

more responsiveness to instruc-

tially more enthusiasm, more persistence,

And they were much more likely to be rated

tions, and less frustration.
None of the ambivalent children
high in the display of positive feelings.
smiled, laughed, or expressed delight at the same level, and almost half
pouting, whin-
the avoidant children engaged in prominent displays of
ing, and hitting. “I remember one avoidant boy,” Albersheim says, “who
just got so upset with the task he started to hit everything, including his
mother, with the stick.”^

Securely attached children also engaged in more symbolic

play —
pouring an imaginary cup of tea and offering it to mother to
drink,

putting a toy man in a toy car and having him drive around,
placing ani-

mals in a barn. Such symbolic play an important aspect of the child’s

is

mental and emotional development. As Albersheim says, it’s a measure

in part of the child’s ability to tune in to his own inner life, a precursor of

self-expression. Also, by creating an imaginary world, and temporarily

living within it, the child practices his understanding of
how the world
operates and has a chance to work out problems, both emotional
and

practical.

Securely attached children also showed more opposition initially at

cleanup time. This might not seem desirable, especially to a parent strug-
gling to get compliance, but it nevertheless indicated that the children
much what they needed to do at this age expressing
were doing pretty
their growing desire for autonomy, as well as a certain confidence in

themselves. Indeed, despite their open resistance, securely attached chil-

dren displayed less anger during cleanup than either of the anxious
groups. They were able to assert what they wanted —
even if it differed
from mother’s wishes — rather than mix compliance with aggression.

What’s more, during the tougher problem-solving tasks, when compli-

ance with mother could make the difference between success or failure,
O
the secure children were the most cooperative.

But what about the mothers? How* were they affecting their childrens
performance?
Many of the key relationship issues for two-year-olds and their moth-

ers were being worked out at this time by Margaret Mahler, the distin-
guished psychoanalyst who was observing infant-mother pairs in a
 ^

The Minnesota Studies 181

playroom. Mahler emphasized the importance of the mother’s “emo'

tional willingness to let go of the toddler — to give him a gentle push, an
encouragement toward independence.”^ In this her thinking was close to
Ainsworth’s, who believed that a sensitive parent picks up on the child’s

cues that “he enjoys the adventures of exploring . . . and he is gratified

when he masters a new skill or problem on his own.”^^
Such sensitivity becomes particularly critical in what Mahler called
the “rapprochement” phase, which overlaps much of the second year. At
this stage, Mahler believed, the child is developing a clearer sense that
the mother is a separate individual whose wishes do not always coincide
with his. As eager as he is to expand his newfound autonomy, he is also
flooded at times with both anxiety and a feeling of loss: He and mommy
are no longer one. He senses, Mahler wrote, that “the world is not his
oyster, that he must cope with it more or less ‘on his own,’ very often as a
relatively helpless, small, and separate individual.”^
During this period he wants his mother to participate in everything he
does and to share his things with him. He insists that she come with him
on his explorations. He brings her one toy after another for her apprecia^
tion, often piling them in her lap. He shadows her incessantly. He darts
away from her, sometimes under perilous circumstances, demanding, in
effect, that she follow and catch him. But at the same time, Mahler said,
he is constantly on guard against being reengulfed by her. By the latter
part of the rapprochement phase, conflicts and crises frequently develop,

in which the toddler demonstrates, in Mahler’s words, an “alternating

desire to push mother away and to cling to her.”^^

Mahler observed that, while some mothers cannot tolerate the child’s

growing separateness, others cannot stand his mounting demands and

begin to withdraw. But the less emotionally available the mother is, “the
more insistently and even desperately does the toddler attempt to woo
her,”^^ with disruptions, displays of babyishness, and fulhhlown tantmms.
As a result the mother’s life becomes more miserable and the child’s

energy is drained away from explorations and the tasks of independence.*

Assessment of the Minnesota mothers suggested that they were negO'

*Mahler’s assumptions about this age period are still debated. While not disputing the behavior
Mahler observed or its importance, analyst and developmental theorist Michael Basch questions her
interpretations. “A child at this age looks back at the mother to see if her face displays an approving,
mirroring appreciation for his achievement. The recognition that ‘the world is not his oyster’ is

something that comes, if it comes at all, at the end of the oedipal phase, when the child should face
and accept his relative smallness vis-a-vis his parents’ and other adults’ power.” (Personal communi-
cation, June 1993.)
182 The Fate of Early Attachments

same degree of sensitivity they d

tiating the toddler period with the

shown in the first year. The mothers of secure children were, for in-
‘‘supportive presence” and
stance, rated almost twice as high in both
child what to do
“quality of assistance.” They did not simply tell their
him the information he needed to complete the task himself
they gave
actions and the
and helped him to see the connection between his
Highly rated mothers gave minimal assistance along the way,
just
results.

keep the child engaged. They seemed to know how to

love at
enough to
impinging.
a greater distance now, without withdrawing or
The mothers of the anxiously attached children, by contrast, seemed
unwilling or unable to maintain an appropriate distance.
Some became
intrusive and made it impossible for the child to
have his own experi-

ence. “They couldn’t tolerate the child having any frustration,” Albers-
problem
heim says. “They would just get in there and almost solve the
for him because it was too painful for them to
watch the child struggle.
But if children don’t get to struggle a little bit
—
and be able to see either
help, and to be able
that they can accomplish it or that they need a little
to figure that out on their own — if that’s interfered with, it s a real loss

for the child.

The mothers of other insecure children hung hack, giving no assis-

tance even when it was needed, or got involved in ways that the child

couldn’t use, with the result that some passive children never got the
experience
push they needed and others became so frustrated that the
only have
was ruined for them. In either case their confidence could
been undermined by such experiences, and the Minnesota team
felt they

saw in
were observing incompetence in formation. The consistency they
quality of parenting suggested that the benefits of secure
attachment in

the first year and the handicaps of anxious attachment would in

many
cases become solidified with age.
The correlations that came out of this early study convinced Sroufe

that he was on to something. Correlations, of course, do not prove

That height correlates well with reading ability in people
causality:

under twelve does not signify a causal relationship between height

and
small
reading ability; both just happen to increase with age. Similarly,
correlations often demonstrate no more than gross tendencies of popula-

fiQj-js — slightly more Jewish men, for instance, may lose their hair than

Italians, hut that statistic means very little to any particular individual.
powerful
But the correlations from this early Minnesota study were very
so pro-
the differences between the secure and insecure children
 The Minnesota Studies 183

nounced on some scales as to show almost no overlap (comparable to: all

Jewish men lose their hair and no Italian men do), as if two different
breeds of children were being observed.
The children had a different style of relating to their mothers as well.
In another study, this time of eighteen^month-olds, Minnesota research^
ers found that the secure children were much more likely to share posi'

tive feelings with their mothers. Almost all the securely attached
children (eighteen of nineteen) smiled spontaneously at their mothers
during a play episode, while less than half of the anxiously attached chil-

dren did so. Almost half of the secure kids showed their mother a toy;
while only three of the seventeen anxious kids did so. And only secure
children both showed a toy and smiled — or showed a toy, smiled, and
cooed all at the same time.^^
In a third study, assessing children at three and a half, the secure group
appeared considerably more advanced in other relationships, too, being
almost twice as likely to suggest activities, to be sympathetic to peers’
distress, and to be sought out by other children.

Sroufe and his students were now convinced that Ainsworth’s twenty-
minute assessment had not been just twenty minutes of random behavior
but that she had tapped into a profound issue in early development.
They looked forward to seeing if the correlations would hold up at later
ages and approached each new study with growing enthusiasm.
They were not disappointed.

Until now, all the Minnesota studies used samples of children from mid-
dle-class homes. But in 1974 psychologist Byron Egeland, a colleague of
who was studying risk factors in child abuse, began
Sroufe ’s at Minnesota
putting together a new sample. He recruited 267 expectant mothers, all
of them with low income and few resources, who were receiving prenatal
care at a nearby public health clinic, to participate in a longitudinal
study. For the next two decades these mothers and their children (whit-
tled down by attrition over the years to some 179 families) would he the
subjects of innumerable studies by Egeland, Sroufe, and their graduate
students.
The women were young (average age just over twenty) and mostly
single, with few social supports. Many had dropped out of high school.
Very few had planned their pregnancies or were well prepared tor parent-

hood. Eighty percent were white, 14 percent were black, the rest His-

panic or Native American.

184 The Fate of Early Attachments

Egeland and Sroufe lavished enormous attention on this poverty sam^

pie,”Egeland with a view to abuse and its precursors, Sroufe looking for
prenatal mea^
the correlates of secure and insecure attachment. They did
sures,obtained nurses’ ratings of each mothers degree of interest in her
newborn, and assessed the children at birth with numerous measures to
question-
rule out constitutional liabilities. Psychological tests, interviews,

naires, and third-party observations soon followed at regular intervals.

One of the first fruits of the new study was the ability to predict secu-
rity of attachment at birth. The child’s constitutional
and genetic make-

up, based on neurological, motor, and cognitive tests, proved largely

although one neurological assessment was able to predict later

irrelevant,

ambivalence, a fact that would explode into a huge debate in coming

years. But mother variables were more powerful. Depressed mothers and
those who had been rated by nurses as having a low interest in their baby
before it was bom were more likely to have anxious children at one year.^^
Whenthe children in the poverty sample reached aged four and a half
to five, Sroufe launched his most ambitious study to date,^^
enrolling

forty of them in a specially created nursery school on the Minnesota

campus. groups of children were observed over a period of fifteen to

Two
twenty weeks in a variety of contexts —
playing outdoors, during free

playtime, all together during circle time (when children and counselors
got together for songs, games, and announcements), and during small
group activity. There was one teacher for every six children, allowing for
a lot of teacher-child contact and the opportunity for teachers to get a

good feel for how each managed his impulses

child expressed himself and

and feelings. In addition to the teachers and videotapes, some twenty

specially trained observers were used as well, keeping tabs on the chil-

dren from a discreet distance. “We even set up booths to observe the
kids,” Sroufe says. “We observed every speck of their behavior.”^^
Numerous qualities were singled out for assessment, and in the end the
teachers were asked to select a single phrase to describe each child. As
usual, none of the teachers or raters knew anything about the children’s

prior attachment classifications.

The world of the preschooler is quite different from that ot the tod-

dler. Children at this age tend to he actively engaged with one another
and to get hy with significantly less adult assistance. Most are able to

contain their impulses and to follow rules. Sroufe predicted that children

with secure attachment histories, because they had been sensitively

responded to, would be more likely to enjoy themselves and less inclined
 The Minnesota Studies 185

to be whiny, aggressive, or have tantrums. He predicted that, because

they had developed positive expectations of others based on their experi'

ences at home, they would also have positive ways of displaying their
needs for attention and would thus present fewer management problems.
On numerous measures these predictions proved accurate — not for

every one of the secure children, but certainly for the group as a whole.
Indeed, children with secure attachment histories scored higher in
every area, from ego resiliency, to self-esteem, to independence, to the
ability to enjoy themselves and respond positively to other children.
They were seen as having superior social skills — initiating more interac-
tions with other children, sustaining them for longer periods, and, when
approached, reacting with positive feelings. They had more friends.
Indeed, they held the majority of the top positions in popularity — an
important finding in light of the fact that degree of popularity among
one’s peers in third grade had been shown to be a strong predictor of
emotional well-being or disturbance in adulthood.
The secure preschoolers also seemed to have more empathy for peers

in distress. Ambivalent children, by contrast, seemed too preoccupied

with their own needs to have any feelings left over for others, and
avoidant children sometimes seemed to take pleasure in another child’s
misery. “When a child was injured or disappointed in the nursery
school,” Sroufe recalls, “since we were filming, we were able to go back
and study the reactions of different children. We found that a number of
ambivalent kids had difficulty maintaining a boundary between them-
selves and the distressed child. That’s customary for really young children
— two-year-olds would do this — but these kids were four or five. So, for
example, a little girl fell and hurt her lip, and one of the ambivalent kids
immediately put his hand to his own mouth and went and got up on a

teacher’s lap. It was as though it had happened to him. In the same situa-

tion an avoidant child would do something like call her a crybaby.

Whereas a secure kid would get a teacher and bring the teacher to the
child or stand by and look concerned.
Some of the ambivalent children, because they were thrown into a
tizzy by aggressive encounters, became easy targets for bullies. And their

unpredictable, sometimes disruptive behavior often made the other chil-

dren antagonistic toward them. The aggressive avoidants tended to be

widely disliked, especially by the secure children, who saw them as

mean.^"^ Only a few of the children were considered severe disciplinary

problems; they all had anxious histories.

186 The Fate of Early Attachments

None of the behavioral patterns that the Minnesota team witnessed in

these children was new to them. They were observing the normal speC'

trum of childhood qualities. But what neither they nor anyone else had
had before was empirical evidence regarding the way the children had
been raised and thus, presumably, the source of their differing social

responses.
Sroufe identified three types of avoidant children at this age the

lying bully who blames others; the shy, spacey loner who seems emotion^
ally flat; and the obviously disturbed child, with repetitive twitches and
tics who daydreams and shows little interest in his environment.^^ Sroufe

was also able to identify two ambivalent patterns the fidgety, impuh—
sive child with poor concentration who is tense and easily upset by faih

ures; and the fearful, hypersensitive, clingy child who lacks initiative and
gives up easily. None of these children were as damaged as Bowlby’s

affectionless thieves. But each of the five troubling behavioral styles

Sroufe witnessed suggests a degree of pain and haunting separateness in

these preschoolers that is sad to contemplate.

Contrary to the predictions of behaviorists, who attributed depen^

dency to pampering, both groups of anxiously attached children — who,

in Ainsworth’s study, had experienced anything but pampering
— turned
out to be highly dependent. When the children were ranked in terms ot

dependency, anxious kids held almost all the top positions. From the
beginning, Ainsworth’s critics had claimed that avoidant children were
not anxiously attached but precociously independent. That was why they
hadn’t greeted their mothers or seemed much interested in them on
reunion. This argument (which persists to this day) exasperated Sroufe.
For it was apparent in the preschool nursery that avoidantly attached
babies grew up to be four^ and five^year^olds who were just as dependent
as ambivalent children, seeking attention in negative ways and making
frequent contact with their teachers, climbing into their laps or sitting
next to them at circle time at nearly three times the rate of the secure
group. “We looked at it every which way,” Sroufe says, “teacher ratings,
observer ratings, frequency of child-initiated contact with teachers
and by every measure those avoidant kids were highly dependent.
Avoidant children were least inclined to behave dependently, however,
when they were injured or disappointed. One avoidant child, after having

banged her head, crawled off into a corner by herself. Another folded his

arms and withdrew when disappointed. Still others expressed their needs

for adult attention in bizarre ways. One approached his teacher “through
 The Minnesota Studies 187

a series of oblique angles (much like one tacks a sailboat into the wind),”

until finally, backed up right alongside her, he would wait for her to initi'

ate contact. Such behavior, much like their reunion functioning in the
Strange Situation three or four years earlier, spoke poignantly of an
expectation that they would be rejected when in distress.

Sadly, their expectations often acted as self-fulfilling prophecies.

Because avoidant children frequently tended to be sullen or oppositional,
because they sometimes preyed on other insecure children, and because
they tended to come across as arrogantly self-sufficient, they were least
likely to elicit their teachers’ concern and most likely to incur their

wrath. One little boy stole a toy, hid it in his pocket, did everything he
could to elude the teacher, and then absolutely denied knowledge of the
toy when questioned.^® Such behavior infuriated the teachers. The
phrases that teachers used to describe avoidant children expressed the
frustration they felt. Of one little girl the first “Mean to
teacher wrote:
other children, kept things which didn’t belong to her.” The second
wrote: “The most dishonest preschooler I have ever met.” The third:
“Mean, lying — everything is hers.”^^

According to Sroufe, the teachers generally did not expect the

avoidant children to comply or follow rules. “The teacher would say,

‘John, I want you to put those toys away over there; John, come on, 1

want you to put those toys away’ — she wouldn’t even leave any time
between the first request and the second because she had no expectation
whatsoever that the child would put the toys away because she asked.
The consistency of the teachers’ response was such that Sroufe would
later say, “Whenever 1 see a teacher who looks as if she wants to pick a
kid up by the shoulders and stuff him in the trash barrel, 1 know that kid

had an avoidant attachment history.”

Although the teachers’ summary statements for the ambivalent chil-

dren were equally disturbing, their behavior toward them was usually
more indulgent. They saw them as ineffective, emotionally immature,
and incapable of following rules, and so they held them to a lower stan-

dard. Little Paulie, for example, was a “wild-eyed man,” perpetually anx-
ious, with almost no frustration tolerance, and little ability to cope.

Paulie’s immaturity, his inability to engage in the ordinary give and take,

and his impulsiveness alienated the other children. But Paulie, as Sroufe

put it, “wore his heart on his sleeve. ... He was so clear about his desire

for closeness and care that his behavioral and emotional problems did
not alienate the teachers. In fact, they were continually supportive.”
i88 The Fate of Early Attachments

On the whole, the teachers reacted with a poignant consistency when

dealing with each of the three types of children (although they knew
nothing of their classifications). They tended to treat securely attached

children in warm, matter^oTfact, age^appropriate ways; to indulge,

excuse, and infantilize the dingier, more scattered ambivalent children;

and to be controlling and angry with the avoidant children, despite the
fact that they were equally needy.
The findings were such that Sroufe and his students at times felt
depressed and guilty as they examined the tapes, especially when they

considered what the future was likely to hold for some of these children.
Subsequent studies of preschool children, coming out of Minnesota
and elsewhere, confirmed and elaborated their findings. It was discov-
ered that children’s fantasy and play themes,^^ their family drawings,^"^
their concentration, their exploratory behavior,^^ and their ability to

negotiate short separations with their mother^^ all looked significantly

healthier or more fully developed if they were securely attached. Former
Sroufe student Dante Cicchetti found a similarly divergent development
among securely and anxiously attached retarded children.^^
In the Minnesota studies early secure attachment did not in most
cases prove a ticket to a problem-free life. Many of the children with

secure attachment histories seemed to have problems, some of which

may have been attributable to unstable or troubling home circumstances,

including a distressed mother, lack of adequate attention and stimula-

tion, and frequently an absent father, all of which were characteristic of

Sroufe ’s sample. But, on the whole, they showed more competence, flex-
ibility, resilience, empathy, and enough relational abilities to keep them
from appearing emphatically disturbed. For instance one of the secure
girls was considered a “sparkplug” by one teacher and a “queen bee” by
another. She had difficulty waiting her turn and she overstimulated the
other children at times with her antics. But she was seen as a competent
child and none of the descriptions suggested the anger or aggression that

might imply a seriously impaired ability to relate to others. Other chil-

dren with histories of secure attachment presented a more troubling pic-

ture — “a spacey, undersocialized kid,” “dependent, sad, frail” — not as

troubling as some of the anxious children perhaps, but disturbing enough

to make clear that security of attachment in the first year is not a guaran-
tee of future emotional well-being.^® Indeed, the security itself might not
last, especially if the environment is as volatile as in the Minnesota
poverty sample.
 The Minnesota Studies 189

The preschool study was a landmark in attachment research and in

the validation of Ainsworth’s work. From very small differences in
behavior — over the course of just twenty minutes or so in the Strange
Situation — very large differences were now apparent in the children’s
adaptation as four- and five-year^olds.
Certain caveats must be kept in mind, however. First, in the excite-
ment over the Minnesota results, it was easy to forget that questions
were going unanswered, partly because no one was doing for subsequent
years the sort of careful observational studies that Ainsworth had done
for year one. Flow is anxious attachment that develops in the second,
third, or fourth year different from anxious attachment that takes hold
in the first? What is retained, if anything, of the initial security? What
unique problems arise depending on the month or year of anxious onset?
And how does the child deal with the loss of his secure love? My own
clinical experience, like that of others, suggests that although much is

retained and much more can, ideally, be rekindled later, the loss of early
security is a major cause for emotional distress in adult life. It may be
occasioned by the arrival of a new sibling, or the disappearance of a lov-
ing grandmother from the home, or the harsh reemergence of the
mother’s narcissistic needs with the end of infancy, any or all of which
can leave the small child feeling bereft and abandoned. Loss is a big word
in the attachment lexicon, but it has not been applied in this context.
Also, is there a time for most children — at the end of the second
year? the third? — when the attachment question is largely settled, re-

gardless of what new problems may arise in the family? Could it he said,
for example, that the acquisition of attachment patterns is largely a fac-

tor of infancy and toddlerhood, when the child is most dependent and
susceptible; that unlike the lower animals studied by ethologists, there is

no rigidly set sensitive period hut that the receptivity to change rapidly
decreases after a certain age? This makes a lot of sense, but there is as yet

little in the way of research data to substantiate it.

Regarding the Minnesota studies, it should he noted, too, that some of

the children only showed subtle signs of their attachment histories, while
a few others did not behave at all the way the theory predicted. What’s

more, neither the preschool study nor the others that preceded it proved
a causal link between secure attachment and healthy emotional develop-
ment; others could argue — and indeed would — that the child’s inborn

temperament was the cause of both. And, finally, Sroufe’s research did

not prove a link between sensitive, responsive parenting and secure

190 The Fate of Early Attachments

attachment: Anxious attachment might after all arise for other reasons,

various unfortunate circumstances, like the family’s moving or suffering

financial trauma, or, as mentioned above, the unhappily timed arrival of

a new sibling or the loss of a loved grandparent, any of which could cause
emotional problems that might drive a wedge between parent and child.
But none of these limitations slowed the enthusiasm that was develop^
ing for attachment research. Nor could they prevent attachment theory,
at least as interpreted by those with a political agenda, from acquiring an
even more strident moral imperative than Bowlby had endowed it with,
with the result that mothers were given new cause to feel guilty or inade^
quate. Coming at a time when women were asserting their rights to a

career and to equality in the workplace and when many families found
they needed two incomes to get by, the new attachment research, in^
spired by Ainsworth’s revolutionary work, engendered anxiety and hos^
tility once again.
It should be remembered that the driving force behind attachment
thinking was a quest to discover what children need, and Sroufe’s
research was designed to show what happened if they got it and what
happened if they did not. In our society mothers tend to he the vehicle
through which the community nurtures its young, often in a more exclu'
sive way than earlier or elsewhere on the planet. If a child is not being
properly nurtured, blame could just as logically he placed on our social

organization, which fixes too many burdens on the mother, on family diS'

ruptions, on paternal absence, on poverty, or on the human condition

itself, which leaves us eternally vulnerable to losses and other tragic

hurts. But attachment theory, as it was evolving, could be quite logically

taken to imply that mothers were the problem. It was studying the infant'
mother relationship and showing, in study after study, the miseries that

befell the child when that relationship was disturbed. Despite cautionary

words from spokesman that mothers needed more support, or that the
its

real issue was the primary caregiver, regardless of sex, or that fathers and

other attachment figures mattered, too, attachment theory was being

seen as a vehicle for blame. Opposition to attachment theory in develop^
mental psychology would eventually coalesce around this issue.

But at the time when the Minnesota papers were first seeing print,

attachment research in the United States was still young, its momentum
was high, and it was not looking over its shoulder. The Minnesota team
had amassed some of the most powerful evidence yet about the impor^
tance of early environments. More would follow.
 14
The Mother, the Father,
AND THE Outside World:
Attachment Quality
AND Childhood Relationships

One day in 1988 an eleven-yeanold girl named Amy was talking excit'
edly with three friends. They were making plans to open a “craft store” at
their summer day camp, where they could barter some of the jewelry
they’d been making.

Amy when she

surprised her friends revealed a small bag of
necklaces she had made at home the night before. She
explained she had been thinking about it at home and
decided that the store would ‘look better if there were lots of
The other
things to display.’ girls agreed that more mer^
chandise would make the shop look more attractive. As they
quickly scooped up their belongings in a joyous response to a
camp counselor’s call to prepare for swimming, they all agreed

to take some yarn and beads home with them that night, so
that they could make even more items to display on the next
day of camp. Walking toward the lockers to get their swim^
ming gear, Amy noticed that Clarissa did not have any yarn
and offered to share some of hers. Clarissa gratefully accepted
192 The Fate of Early Attachments

the yam, put it in her locker, and the two girls rushed to catch
up with the rest of the group, now on its way to the poold

By the late 1980s the children from the poverty sample had reached
their eleventh year. This is an age when Erikson stressed the importance
of industriousness in a child’s life and when the American psychoanalyst
Harry Stack Sullivan spoke of the emergence of loyal friendships, or
“chumships.” It is a time when group membership becomes critical to a

child and when he learns selTconfidence and constructiveness in social

settings. Now forty^seven children from the sample spent several weeks
attending a summer day camp, and were closely watched by the Minne-
sota team. Among those selected for this study — by graduate students
James Elicker and Michelle Englund — was the girl they called Amy.
Amy and her friends, like the rest of the Minnesota poverty sample,
had been studied and evaluated in one way or another since they were
born. When Amy was one she had been observed with her mother in the
Strange Situation and was found to be securely attached. The mother’s
behavior in those few minutes was taken to he representative. She was
encouraging when Amy exclaimed over a new toy, reassuring when Amy
became momentarily disturbed, pleased to watch Amy’s explorations.
Mother and daughter were smooth and relaxed with each other and, as
Amy picked up a toy to mouth or shake, she sometimes looked back and
smiled at her mom.^
The researchers assumed that, as a result of their mothers’ sensitivity
and responsiveness, children like Amy develop positive social expecta^
tions, come to understand the idea of reciprocity, and develop a sense of
self'worth and efficacy. Assuming no jarring changes in this relationship,

this positive experience should make it possible for them to enjoy the

spontaneity, ease, and giddy pleasures that Amy shared with her friends
at eleven and a half years.

Relationships with peers had been a component of attachment studies

for some time now. As early as twenty to twenty^three months of age
securely attached children were rated as more sociable.^ By the age of
three it became apparent not only that anxiously attached children were
less sociable but that other toddlers didn’t respond as positively to them."^
When they were four years old, a time when social skills and sensitivities

are rapidly developing and true connections with other children begin to

form, Sroufe and his students divided up a group of preschoolers into

same'sex playmate pairs, carefully arranging for every combination of
 The Mother, the Father, and the Outside World 193

secure, avoidant, and ambivalent children, and videotaped their interac-

tions. The children with secure attachment histories consistently partici'

pated in the best functioning pairs. They were more sensitive and
attuned to their partners, developed true friendships with them, and
engaged in creative and elaborate play. Indeed, the secure children had
the most healthy partnerships across the board, even when they played
with children who had anxious histories.
The avoidant children on the whole made the worst partners. Not
only did they have difficulty forming a mutual, positive connection, but
they sometimes took advantage of their less competent, more vulnerable
playmates — by tricking them during a swap, for example, or attempting

to steal their toys.

Many of the secure children refused to be put off by the rebuffs of their
anxious partners. Former Sroufe student Van Pancake recalls a case like

this:“The boy with the secure history would initiate play, with blocks or
with cars, and the other boy would ignore him or say something reject'
ing. But the secure boy continued to initiate. And he was very creative.
He kept coming back in a way that had variety in it and that took into
account the kinds of activities and interests of his playmate; and, eventU'
ally, became very good playmates.”^
these two boys
Anxiously attached children on the whole lacked this ability. The
avoidant children tended to show little interest in closeness. On one
occasion the teachers had made a “space capsule” out of cardboard and
aluminum foil. An avoidant boy saw it across the room, asked the

teacher about it, and became interested. But as he started toward it, he
saw the movements of other kids inside and turned and went hack to his
Legos.^ The ambivalent children, on the other hand, were drawn to rela-
tionships but often were not competent in them. They usually did well
“1
with secure partners, but they ran into problems with avoidants.
remember one of these children with the ambivalent history was very
interested in engaging her partner,” Pancake says, “but the initiations

were repetitive and when they were irritating to the partner, the child

did not modify them and find some other way of connecting.”^
The most disturbing material to come out of these fout'yeat'old play'

mate pairs were repeated acts of cruelty. Lucy, a little avoidant girl, tre'

quently antagonized Nancy, her ambivalent playmate, by saying no to

every comment Nancy made. At other times she froze Nancy out when
she sought interaction. Once, when Nancy complained of stomach pain,
Lucy poked her in the aching spot. When Nancy cried out, Lucy poked
her there again.
194 The Fate of Early Attachments

Sroufe and student Michael Troy reviewed the videotapes of fourteen

of the preschool playmate pairs for evidence of victimization. In all five

cases where an avoidant child was paired with another anxiously at-

tached child victimization was noted, usually by all three judges. None of
the judges saw victimization in the remaining nine pairs.
Elliot and Oscar both had avoidant histories. Elliot continually tor-

mented Oscar, who was gullible and passive in the face of Elliot’s aggres-

sion. In one play session, Elliot whispered expressions like “bugger nose,’’

“bugger face,’’ and “Hey, poop!’’ into Oscar’s ear. He then became more
hostile, throwing Oscar’s toys or claiming they were his. When Oscar
somewhat meekly called Elliot a liar, Elliot acted shocked and said, “Oh,
you said a naughty word. I’m telling my momma, you said bad words. I’m
telling the teacher. I’m gonna tell everyone!’’^ Oscar became more flus-

tered and upset.

The securely attached children did not allow themselves to be bullied
or even pulled into a relationship that had hurtful dynamics. They either
found a way to make the relationship positive, withdrew, or met the
aggression with just enough force to discourage it. It was as if such behav-
ior was foreign to them and they would have nothing to do with it. The
anxious children who became victims, on the other hand, seemed at

times to facilitate their own exploitation. In the case of another pair of

little boys with avoidant histories, Ronnie repeatedly victimized Ralph.
At the beginning of one session, after the two had been playing sepa-
rately for a minute, Ralph approached Ronnie and said with an implor-
ing voice, “Why don’t you tease me, Ronnie? I won’t get mad.’’^ In
another pair, the victimized child made 119 overtures, of which only
nineteen received positive responses.
To Sroufe it was self-evident that such piteous behavior patterns had
their origin in the rebuffs the children had suffered when they’d taken
their needs to their mothers. These children must have experienced
emotional unavailability, rejection, or physical abuse at home such that
it only seemed natural to experience or inflict it in other relationships.

“It’s unlikely,’’ Sroufe said of Lucy, “that the child’s mother actually hit
her in the stomach when she had a stomachache, hut she has had count-
less experiences with what happens* when a person is vulnerable and
needy. Does a partner exploit that vulnerability? Is a partner rejecting
because you’re feeling vulnerable? If that’s been my experience of what a
partner does, that’s what I know to do as a partner when I’m in that situ-
ation.’’
 The Mother, the Father, and the Outside World 195

The behavior of securely attached children who reacted sensitively to

their playmates and supported their more vulnerable friends also must
have derived from their home experience. “If you’re in a relationship, the

relationship is part of you, there’s no way around it,” Sroufe said. “How
do you get an empathic child? You get an empathic child not by trying to
teach the child and admonish the child to be empathic, you get an
empathic child by being empathic with the child. The child’s under'
standing of relationships can only be from the relationships he’s experi'
encecl

As the children reached preadolescence, the superior relational abilities

of the children who had been securely attached in infancy continued.

Elicker and Englund found that secure children spent more time with
each other and less time with adults, and the interpersonal sensitivity
they displayed at four continued to develop. Seventy'Six percent of the
securely attached eleven'yeat'olds made friends, compared to only 45
percent of those who had been anxiously attached.^ ^ When groups

formed, like Amy and her friends, they were composed almost exclu'
sively of children with secure histories, sometimes with one anxiously
attached child included.^^ The emotional demands of group functioning

— status, role apportionment, conflict resolution — were apparently too

most insecurely attached kids to negotiate without arousing
difficult for

more anxiety than they could effectively handle or behaving in ways that
alienated their peers.
The secure children were also more able to hold on to their friend'

ships while participating in a larger group. They could play near others or

invite others to join them without fearing that they would lose their

friend. Ambivalent pairs could not do this. In the course of three sum'

mers, only one pair of avoidant children formed a friendship: “Their play
often took place in private areas at a distance from the others or behind
some sort of barrier. They were jealous of the advances of any other child

and did not invite others to play with them. When one or another part'
ner was absent from camp, the other appeared ‘lost,’ unable to join in
with the others.
Ten'year'olds with behavior problems — who were hostile, noncom'
pliant, hyperactive, who tended to give up and cry,who displayed ner'
vous habits, who were passive and withdrawn, or who were worried and
unhappy — were usually children who had been anxiously attached to

their mothers. But quality of attachment seemed to affect boys and girls
 —
196 The Fate of Early Attachments

differently. In the poverty sample, for example, aggression was particu'

larly marked among avoidant boys. They were more likely to bully, lie,

cheat, destroy things, brag, act cruelly, disrupt the class, swear, tease,
threaten, argue, throw temper tantrums, become defiant. (Ambivalent
boys, on the other hand, were more prone to be shy, apathetic, and with'
drawn.) The picture was quite different with avoidant girls, perhaps
because they are biologically less aggressive and because girls, for a vari'
ety of reasons, are more inclined to cope with their anxiety by behaving
in socially approved ways. They tend to internalize their feelings more
blaming themselves, becoming depressed, or feeling ashamed as opposed
to fighting off such feelings by striking out at someone or through some
other behavioral strategy.
Patricia Turner, a doctoral student at Cambridge University working
with Joan Stevenson'Hinde, wife of ethologist Robert Hinde and a lead'
ing attachment researcher herself, has performed one of the few studies
that have attempted to build on this question. Turner found that some of
the stereotyped qualities we associate with boys and girls — such as smih
ing compliance in girls and angry aggression in boys — may actually rep'
resent the ways in which insecure attachment works itself out in each
group.
The anxiously attached boys, for instance, were off the charts on every
measure of aggression, assertion, and control seeking. They appeared to
be very hungry for attention and approval (their dependency scores were
also the highest), but they were unable to seek attention openly, perhaps
because their hunger for it and their expectation of rejection were both
too terrible to allow into consciousness. They thus sought attention in
indirect ways, which were often disruptive and aggressive. Their peers
would either resist them or forcefully ignore them, which only height'
ened their frustration and need, reconfirmed their expectation that they
would he rejected, and increased their offensive behaviors. Secure boys
did not behave this way and, in fact, they were no more assertive, con'
trolling, or angrily aggressive than secure girls (although they did show
twice as much playful aggression).*

*The number of dependent behaviors among

anxiously attached boys was almost twice that of the
anxiously attached girls and approximately three times that of the securely attached children.
Unfortunately, Turner did not distinguish between children with avoidant and ambivalent histories.
It is hard to know, therefore, whether her findings conflict with the Minnesota finding that hoys

with ambivalent attachment backgrounds are very different from the aggressive avoidants. See Note
16 for figures.
 The Mother, the Father, and the Outside World 197

Secure boys and secure girls were also about equal in the degree to
which they expressed positive feelings. Insecure girls, however, by a huge
margin — about two to one — expressed vastly more pleasure and con-
many more smiles on their peers than any other group. Again the
ferred
—
management of relationship anxiety the sense that one wants to be
loved and connected but that one will not be treated well — seems to

explain this behavior. It has been found in other studies, for instance,
that low'Status, low'pecking-order children in a nursery school tended to
be big smilers when initiating contact with higher^ranking children. The
smiles of such children, far from being expressions of pleasure, are most
likely a demonstration of submission and plea not to be maltreated.

Although the data linking secure attachment with better peer relations
in childhood has been impressive, Ainsworth has cautioned against

jumping to conclusions. “There is evidence that the securely attached

child will do better with other kids, but it’s not true across the board.
There are kids that are doing much better with their same-aged peers

than you’d expect from the nature of their attachment to their parents.”
She believes that something besides the attachment system is at work
here — what she laughingly calls “the sociable system” — and that no one
has yet fully grasped the complexity of what’s going on.^^
It should be remembered, too, that the Minnesota sample consists of
them headed by a single parent.
hard'pressed families, over 60 percent of
Many of the children have presumably suffered more than the insensitive
ity, inconsistency, or rejection of attachment needs that Ainsworth oh-
served in her middle'class Baltimore homes. Such circumstances would
naturally exaggerate their problems in relating to others, suggesting more
drastic consequences to anxious attachment than is necessarily the case.

By the early nineties the children in the Minnesota study had reached
their teenage years, and Sroufe found that whatever relationship advan^
tages secure attachment does tend to confer persist through age fifteen.^®
He sees in his longitudinal study the opportunity to trace the fate of early

attachments into later years as well. If the funding for the research can
be maintained, he will be able to keep additional generations of graduate
students studying the poverty sample at successive ages right on through
adulthood. From the point of view of the National Institute for Mental
Health, which has been providing much of the funding, the study would

seem irresistible. “You couldn’t name an NIMH priority,” Sroufe says,

“that we can’t access with the data coming up: drug abuse, delinquency,
198 The Fate of Early Attachments

AIDS, teenage pregnancy.” He will also, before long, be able to assess

what kind of parents they become and whether their children’s attach'

ment status mirrors their own. If, as he suspects, attachment history plays
a role in predicting such a wide variety of outcomes, the payoff of this

research will have proven to be considerable.*

In the mid' 1970s attachment workers began showing an increased inter'

est in the child’s attachment to his father — a person who tends to get
overlooked in most developmental research, either because he is consid'
ered less important or because he is less available for observation. (In
Sroufe’s poverty sample only 12 percent of the biological fathers were
present at eighteen months and many were reluctant to participate;

assessments of their impact could only be indirect. One of the first

efforts to overcome this was made by Michael Lamb, a former student of

Ainsworth’s. Observing children in the home at seven, eight, twelve,

and thirteen months of age, he found that on most attachment measures

— proximity seeking, touching, crying to, signaling a desire to be held,
protest on separation, greeting on return — infants showed no preference
for mothers over fathers. When infants are distressed, their attachment
behaviors become more pronounced, but they display them in the same
way to both mother and father, depending on who is present. When both
are present, distressed infants consistently prefer their mothers,^ even
little boys who otherwise prefer their fathers. Thus, Lamb’s work seemed
to support Bowlby’s contention that children arrange their attachment
sympathies in a hierarchy.^ ^ (This certainly seems to be the case in my
home.) Most children, however, are flexible enough to switch the father
to the top of the hierarchy if the mother is unavailable for a period of
11
time.
The first study to assess quality of attachment to the father was re'
ported by Mary Main and Donna Weston in 1981. They found the pet'
centage of children securely attached to their fathers was about the same
as that for mothers. But there was no correlation between the two.^^ A
child could be securely attached to one, both, or neither of his parents.
While one secure attachment, especially if it was to mother, was better
than none, children who had been found to be securely attached to both

*A chart briefly summarizing the findings to date regarding secure and anxious attachment appears
in the appendix.

*Ot, probably more accurately, their primary caregiver, assuming secure attachment to both parents.
 —
The Mother, the Father, and the Outside World 199

parents tended to be the most confident and competent. When empathy

was measured, the children with two secure attachments were notably
superior. This has since been supported by other studies. On the whole,

it seems that everything that has been learned about mothet'child

attachment relates in some way to father^child attachment as well, as

many readers have no doubt already assumed. And yet there are differ'

ences.
The primary caregiver — and therefore the mother in most cases — is,

of course, especially important. Something fundamental seems to get

established in the infant’s relationship with her during the first year or

two that often considerably outweighs the contribution of any secondary

attachment figure. But the formative power of the second parent
whether he is harsh or accepting, tyrannical or easygoing, highly

involved or abdicating, living at home or long gone — is critical, too.

This fact has been established over and over again in clinical work,
where unsatisfactory relationships with or abandonment by fathers often

require years of working through. Today 27 percent of American chib

dren are born to single mothers, many of them poor teenagers.^^ We
know from Sroufe’s studies that many such children do not do well.
Although fathers are usually secondary caregivers, they are not merely

secondary mothers. They usually provide a higher level of stimulation,

are often seen by children as more exciting playmates, and their (on-

average) somewhat lesser degree of intimacy with the child and greater
resistance to coddling makes them more of a stepping-stone to the out-

side world where the child will have to relate to people who are not in

perfect sympathy and attunement with him.^^ Fathers’ growing involve-

ment with their children (especially sons) during the toddler period,
which has been documented in many cultures and even in primate stud-
ies, may facilitate the child’s ability to move outside his mother’s orbit.

Finally, of course, fathers are men and therefore able to offer something

to both sons and daughters that mothers cannot.

Fathers are role models for their sons; they are sought after, imitated,

competed with, worshipped. Boys hunger for this relationship and are
extremely attentive to everything that goes on within it — acceptance,
rejection, cooperation, degree of openness, limit setting, respect and self-

respect. Both the quality of the father’s life and the quality of his rela-

tionship with his son will deeply affect the developing boy’s sense of self

and possibilities. Which is to say that boys identify with their fathers.

Identification is not an area much explored by attachment theorists, but

200 The Fate of Early Attachments

its psychological effects are huge. It will certainly have much to do with
a hoy’s character structure, influencing it in ways that attachment theory
cannot now predict. (Of this, more later.)

A father’s impact on his daughter tends to be somewhat different but

is also profound. Whether and how he is able to express his love and
show that he values her will affect the extent to which she feels valuable
and lovable herself, especially to other male figures. The sexual attrac-
tion between father and daughter can be a source of self-confidence or
shame and guilt. If the father is emotionally estranged from his wife and
becomes overly seductive with his daughter, that may adversely affect her
development, skewing it toward seductiveness or enmeshment.
There is every reason to believe that husbands also have an impact on
the quality of mother-child attachments. Marital satisfaction, for in-
stance, as well as the father’s general support of the way the mother han-
dles their child, have both been associated with the quality of the child’s

relationship to the mother.^® And even troubled girls, if they later marry
relatively supportive husbands, are found to be adequately nurturant
mothers. Thus, how the father relates to the mother; whether he sup-
ports her caregiving efforts; whether he acquiesces in or collaborates with

her unfairness, imperiousness, or other failures; whether he demonstrates

love and respect for her, distances himself emotionally from her, gets
involved in manipulations and power struggles with her may all have a
lasting impact on the child’s psychology. Studying that impact, which
may be different for boys and girls, is an important direction for future

attachment studies. It will be promoted if attachment workers begin per-

forming for later years the sort of home studies Ainsworth did in the first

year — because many fathers only become significantly involved with the

child when it is older.

Other attachment workers, notably former Ainsworth student Robert

Marvin, at the University of Virginia, and John Byng-Hall, a colleague
of Bowlby’s at the Tavistock, have attempted to include more complex
family dynamics. Their work suggests that family secrets, covert family
alliances, the way conflicts are managed, the solidarity of the parents in

maintaining their roles as authorities in the family — all the issues to

which family theorists have drawn attention — affects the formation,

maintenance, and quality of attachment bonds. This, of course, puts

parental sensitivity and attunement — the principal factors measured by
Ainsworth — in a different perspective. Needless to say, all of this begs
for further study.
 The Mother, the Father, and the Outside World 201

* * *

By the late 1970s, Minnesota papers had begun appearing with regularity

in the child development journals. That the quality of early attachment

correlated so powerfully with so many important aspects of the child’s

emerging personality brought attachment theory a level of interest and

recognition it had not known before. For many in the field, Sroufe’s data,

as well as his dramatic way of organizing and presenting it, permanently

altered the contours of the naturemurture debate. Evidence of this sort,
about the impact of early experience, emerging from solid empirical
research, had never before been available.

Students getting their Ph.D.’s in psychology while working with

Ainsworth, Sroufe, and Main in the seventies, as well as others who only
read their work, became enchanted by the potential of this new area of
investigation to answer age-old questions about what children need.
Attachment research offered them something that was rare in academic

psychology — combining the pleasure of testable hypotheses with the

prospect of changing the world. And at the center of it was this amazing
device — the Strange Situation — which could be performed in twenty
minutes and which could presumably reveal something fundamental
about a baby’s unfolding psychology and the way it had been raised. By
the early eighties, universities all over the United States were populated
with attachment scholars.
 Structures of the Mind:
Building a Model
OF Human Connection

Why do we persist as we are? Why is change so difficult? Why should we

assume that many of Sroufe’s anxiously attached children will not find in
other relationships what their parents failed to give them? If the
avoidant child has dreams of one day finding his fairy princess or her
knight in shining armor, why shouldn’t those dreams come true? Why
should a child who’s been neglected, rejected, or abused torm relation-
ships that seem to reenact those conditions? Why should the rich get
richer and the poor poorer, even in emotional life? Why does the past so

often predict the future?

Psychoanalysts had long assumed that the answer to such questions lay
in the unconscious processes that begin early in life. It’s not simply a
matter of learning — in the sense of being conditioned to repeat certain
behaviors — or imitating elsewhere what one has seen at home. One
forms images of the self and others and of how they fit together, which
have a powerful hold on the personality and serve as a blueprint tor
future relationships. The analytic literature was crowded with terms and
metaphors that tried to grasp this inner process, much of it centering
around the concept of “transference,” which we will return to later when
 Structures of the Mind 203

we discuss adult attachment. With his usual audacity, Bowlhy hoped to

clear away the tangle of analytic terminology with a new concept, “the
internal working model,” which, unlike the others, would be grounded in

hard science.
Eclectic as usual, Bowlhy cited evidence that lower animals like the

digger wasp make such internal models in the form of a mental map of

their environment, which enables them to navigate in their world with'

out having to work everything through from the beginning each time
they encounter a familiar situation.^ If this were true for lower life forms,

Bowbly argued, then certainly much more sophisticated models must

exist in humans, including our understanding of how intimate relation-

ships work, how others can be expected to treat us, and how we perceive
ourselves. This line of thinking led him inevitably to the Swiss cognitive

psychologist Jean Piaget.

Piaget argued that we are born with a system of extremely flexible

mental and behavioral programs, which he called schemata, that enable

us to explore and make sense of our environment. It is upon this inborn
framework that all future learning is built. In Piaget’s view, as in Bowiby’s
and later Ainsworth’s, the infant is not a passive creature who is shaped
by his environment, but is constantly exploring, striving to learn, and
struggling to bring the environment under his control.

The baby’s earliest programs, according to Piaget, include sucking,

grasping, listening, vocalizing, and moving the limbs. To learning theo-
rists these were random behaviors. To Piaget they were tools specifically

designed to enable the baby to interact with, investigate, and master his
environment. They were, in effect, sensorimotor blueprints with a be-
havior at one end, a mental representation at the other, and an immense
capacity to absorb and adapt. The infant uses each program to explore

and process information from the world around him. He sucks on a vari-

ety of things, he grasps them, he makes noises, and through these efforts

he constantly picks up new information not only about his environment

but about the impact of his behavior. As he observes the consequences of
his behavior, his inborn programs become modified and more sophisti-
cated — the tools themselves change. Because sucking won’t work on a

glass of water, for example, the sucking program gradually expands

through trial and error to include drinking — and the child not only

becomes more proficient at getting what he wants but has developed an

expanded vehicle for exploration.

It is upon this highly flexible and modifiable framework, which gains

204 The Fate of Early Attachments

in complexity with increasing physiological maturity (enabling the child

to develop new programs like walking and talking) that all intelligence is

built and continues to be built throughout life. And it is through these

mental abilities that the child is able to engage in a distinct class of be-

haviors that Piaget called “exploratory” and that he believed are as

essential a component of our inborn behavioral repertoire as mating and

feeding.^
The with attachment theory were apparent to Bowlby from
parallels

the beginning. Both theories started with the idea that the child is not a
blank slate, that he has inborn mental capacities, structures, needs, and
that he is striving toward various things —
toward relatedness in Bowlby ’s
sphere of interest, toward mastery in Piaget’s.
Bowlby’s encounter with Piaget enabled him to speak in what he
believed were more scientific terms about internal representations.

Through sucking, clinging, following, smiling, crying, and, when older,

by going to mother when in physical or emotional distress or simply

when he wants attention, the child explores his relationships and builds
a model of the way they work. The internal working model, like the

wasp’s internal map, reflects the child’s relationship history, codifying the
behaviors that belong to an intimate relationship, and defining how he
will feel about himself when he is closely involved with another person.

Bowlby saw attachment as essentially complementary to exploration.

When attachment behavior is activated, often because of fear, explor-

atory behavior is shut down. When attachment — in the form of proxim-

ity or felt security — is achieved, attachment behaviors are shut down and
exploration may begin again. This complementarity between attachment
and exploration was observed repeatedly by Ainsworth in Uganda and
Baltimore, by Harlow with rhesus monkeys, and by Margaret Mahler,
who saw children early in the second year constantly returning from their
explorations to spend a moment with mom and “refuel.”^ Thus, Bowlby
wrote, parents who are “encouraging, supportive and cooperative” not
only enable a child to feel securely attached, but also enable him to con-

fidently explore his environment and develop a sense of competence.

When the child became old enough to have insight into his mother’s

feelings and motives, he was able to form a much more complex relation-

ship with his mother, what Bowlby ^called a goal-corrected partnership,

which meant essentially that now mother and child could negotiate and

make plans together, based on their respective needs, a fact that Robert

Marvin would later verify with a number of ingenious experiments."^

 —
Structures of the Mind 205

This, too, would be fed into the child’s model, updating it with a more
sophisticated sense of the complex give and take and the subtle mutual-
ity of relationships.
There were some important differences, however, between the purely
cognitive schemes Piaget proposed and attachment’s internal working
model. Most important, a person’s model of relationships is often a much
poorer guide to reality. Ideas and memories get distorted when love, hate,

anguish, and shame come into play; when we repress memories, cut our-
selves off from feelings, and develop defenses. The attachment models of
anxious children in particular, Bowlby argued, become rigidified and dif-

ficult to update. Secure children may recognize that some people are cold
and unwilling to befriend them, but children whose attachments are pre-
dominantly anxious, who have no experience and therefore no model of
secure relating, may have a hard time recognizing, may not want to rec-
ognize, that another person is able to be steadily loving and available.
The rigid internal models of anxious children may also act as a self-

fulfilling prophecy.^ The person who expects to get rejected gives little of
himself, acts mistrustful, ignores or misreads friendly overtures, appears
superior and standoffish, and others back away from him. Such experi-
ences confirm his early expectations: This is how other people treat me;
this is all 1 am worth.^ The anxiously attached girl may become quietly
hypnotized by fairy tales in which an unloved child, like Cinderella,

achieves the true love she deserves, but her habitual modes of feeling,
perception, and communication may turn such dreams sour in the end.
The internal models of very young children are particularly subject to

distortion, because they readily misinterpret the meaning of their par-

ents’ communications. They feel hated and rejected as a result of un-

timely separations; they interpret outwardly rejecting behavior as proof
that they are not loved. They draw conclusions
— “1 am responsible for
mommy’s drinking; if 1 were a better child she’d be a happier person”
that bear no relation to the facts. Under three, they lack the cognitive

sophistication to think through the implications of what they feel. As

Mary Main notes, anyone will feel unlovable if the person he is most
attached to is rejecting. But an older child, who can recognize the dis-
tinction between appearance and reality, can use his logic to hold out
some hope for himself: “1 may be unlovable, but who knows? Mom’s been
wrong about other things.”^

The false unconscious beliefs about himself to which the small child is

vulnerable can be reopened for consideration and working through at

 .

206 The Fate of Early Attachments

later ages. This sometimes happens when adolescent girls confide in each

other; when adults fall in love, experience a rebirth of trust, and feel lib'

erated to explore previously forbidden internal terrain; or when they

become parents and gain a new perspective on their early life. But built

into the very nature of shameful selhfeelings is a desire to ignore them.

Indeed, we often construct our lives in such a fashion as to keep them

out of consciousness and away from the view of others. So it is not
uncommon, even in adulthood, to be burdened with unexamined and
hateful self'Concepts first incorporated at a young age.
As the child builds up defenses to ward off unmanageable thoughts
and feelings, his working model of relationships is subject to further diS'

tortions. He may, for instance, consciously believe his mother is the most

wonderful and loving mother he knows and unconsciously see her as

insensitive and inconsiderate. He may consciously fear that his brutal

abandon him and unconsciously wish it.^ Other distortions

father will
may he imposed on him by adults. Extreme instances of this were docu'
mented in a study of forty-five children who had lost a parent by suicide
and subsequently became psychiatrically disturbed. It was found that
about one fourth of them had witnessed the suicide or seen evidence of
it hut had been pressured to believe that the parent
had died in some
other way.

A hoy who watched his father kill himself with a shot-gun . .

was told later that night by his mother that his father died of a
heart attack; a girl who discovered her father’s body hanging
in a closet was told he had died in a car accident; and two
brothers who had found their mother with her wrists slit were
told she had drowned while swimming.'^

When the child related what he had seen, the surviving parent typically
memory or discredit it by attributing it to a TV show
tried to ridicule the

or a had dream. They themselves, it seems, were unable to deal with the
reality of what had happened, and thus forced their own disavowal on

the child. Other children see sexual scenes they were not supposed to
see; they are drawn into nocturnal incestuous liaisons they are not sup-
posed to remember during the day; they have hurts they’re not supposed
to have. A little girl is rejected by a father who once actively adored her

because he is overwhelmed by his sexual feelings for her. His horrible

 Structures of the Mind 207

ahout'face is continually discounted by the mother, who not only insists

that daddy loves her hut denigrates the child for thinking otherwise.

Evidence shows IBowlby wrote] that many of these children,

aware of how their parents feel, proceed then to conform to
their parents’ wishes by excluding from further processing
such information as they already have; and that, having done
so, they cease consciously to be aware that they have ever
observed such scenes, formed such impressions, or had such
experiences.

The resulting confusion, mistrust, self-doubt, and guilt can play havoc
with their psychology.
The study of anxiously attached children suggests that many of them
fail to develop cognitive capacities that might enable them to reevaluate
and work through such distorted models. Main has found, for instance,

that insecure six-year-olds have a reduced capacity for self-reflection, for

thinking about feelings, and thinking about thoughts. Secure children are
more likely to be able to acknowledge having more than one feeling at a

time, they can imagine how feelings might change, they recognize that
different people might feel differently in the same situation, and that peo-
ple can say one thing while meaning just the opposite. Anxious children
often lack such awareness. By age ten their options for working through
are further impaired because they have difficulty even recalling the past.^^

The power and persistence of negative models, and the impact they
have on the child’s capacities for self-reflection and complex thought,
help explain why change can he so difficult and life so unfair. But such
models are not completely dominant. Their power is presumably affected
by how extreme is the the child’s anxious pattern with the parent and
whether he has other attachment relationships that are secure. (1 say

“presumably” because these factors have for the most part not been con-
sidered in the literature.) Even a child who has built negative internal
models based on experiences with both parents may retain an island of

secure functioning as the result of repeated experiences with a third per-

son, of far less importance, like a grandparent or baby-sitter. Psycho-
analyst Janet McKenzie Rioch reported such a case five decades ago, in

which a patient was helped in later life by having had a secure attach-
ment to her childhood nurse:
208 The Fate of Early Attachments

During the many hours when she was with this nurse, she was
able to experience a great deal of unreserved warmth, and of
freedom for selhrealization. Her own capacities for love and
spontaneous activity were able to flourish. . . . This, which one
might call her real self, although “snowed under and handi-
capped by all the distortions incurred by her relationship to
the parents, was finally able to emerge and become again
active in analysis.

With his work on internal models and mental processes, Bowlby helped
bring psychoanalytic concepts about inner processes closer to the main-
stream of developmental thinking. A brilliant synthesizer, Bowlby was
proud to have built bridges from Freud to Piaget and to numerous other
areas of scientific inquiry. Ainsworth, like Bowlby, was also committed in
her way to a psychoanalytic view. She had been analyzed in the early
1960s after her divorce, and she had read Freud’s vast collected works
during this time. To her, the bridge from Freud to Piaget made great

sense (even if Piaget himself regarded psychoanalysis as an unscientific

mythology that would soon be dead^^). As developmental psychology
finally began taking interest in Piaget in the 1970s, her Baltimore re-

search, with its powerful empirical base, helped make the bridge from

Piaget to Freud a sturdy crossing for other developmentalists.

But the way he constructed it, Bowlby ’s bridge was a narrow one. His

internal model seemed to lack contact with the complexity, irrationality,

and passion of intrapsychic processes. If it had something to teach his

analytic colleagues, it also had much to learn from them about, for

instance, the processes of identification by which a child could adopt for

himself the most hated aspects of a parent, especially a parent who was
unavailable through emotional distance or loss; the ways in which we
cleave to bitterness, like a hellish security blanket we cannot bear to give
up; the endurance of early paranoia. His internal working model was an
important contribution, both as a step toward giving analytic concepts
firmer grounding in science and as further proof of the formative power
of early experience. But, as 1 suggested in Chapter 8, it glowed a bit too

much with the logic of the day, where goodness if given a chance can
only prevail and where the night’s disturbances are but a dream.
In 1969 Bowlby published his book Attachment, a vast reworking and
updating of “The Nature of the Child’s Tie to His Mother.” This would
be volume one of his attachment trilogy, followed by Separation in 1973,
 Structures of the Mind 209

Loss in 1980, and a revision of Attachment in 1982. In a review of the see-

ond volume, analyst Charles Rycroft would write:

Dr. Bowlby is among psychoanalysts

almost unique in believ'

ing in the principle of economy of hypothesis and in prefer-

ring simple, commonsensical explanations to ones which are

complex and obscure. As a result, his findings are refreshingly
straightforward and tend, perhaps rather boringly, to be pre-
cisely what warm-hearted but naive nonintellectuals have
always thought.

This was, of course, Bowlby s limitation and his great strength, and
Rycroft meant it as praise. It was Bowlby, after all, who had recognized
how often the early loss of a parent or other loved person was the precur-
sor of depression and other clinical states when others had missed it; it

was he who had determined that love is a biological necessity and not, as

so many others had assumed, an accident built on dependency. Time and

again Bowlby took the direct route where others were circuitous and mis-
taken. He was unusual among analysts in choosing to study other disci-
plines and make use of their findings. He was unusual, too, in that he
relied on research and inspired so much of it. Because of these strengths,
his concept of the internal working model caught on in developmental
psychology where analytic ideas had traditionally been shunned. One
attachment paper, published in 1985 by Mary Main and her colleagues at

the University of California at Berkeley, would be particularly influential

in this regard, bringing the concept of the child’s inner life to the fore-

front of academic psychology as perhaps never before.

 The Black Box Reopened:
Marv Main’s Berkeley Studies

By the Minnesota research had been widely reported

early 1980s, the

and extremely influential. Indeed, the model that Alan Sroufe had piO'
neered was now so extensively employed that new data were continm
ously appearing in the journals about apparent consequences of secure or
anxious attachment. Sroufe had spawned an industry.
“People kept trying to do what Alan was doing,” Jude Cassidy recalls.

“Alan has talked about attachment and peer relations, but Alan looked
at it in two'year^olds. I’m going to look at it in three^year^olds. Or Alan

looked at it with best friends. I’m going to look at it with strangers. Or

nobody’s looked at it in five^year^olds, or nobody’s looked at it in the

kindergarten. By the time I came along, it was getting pretty thin. I was
referring to it as the ‘do it to the right, do it to the left’ school of social

research. I had the sense that we needed a shot in the arm — people
are going to he measuring shoe size next! And then Mary Main came
along.”’

Mary Main was one of Ainsworth’s most creative and prolific students.

A graduate of the Great Books program of St. John’s College in

Annapolis, her first love had been linguistics, hut due to poor grades she
 —
The Black Box Reopened 211

had not been able to get into the linguistic graduate program at Hopkins.
“But I was married to a lofty professor who said that if I was interested in
language I should go back to babies, that everything interesting probably
started with babies,” and that caused her to accept a friend’s suggestion

that she try to study with Ainsworth. “It was an arranged marriage,” she
now says of her graduate work with Ainsworth — observing and classify^
'
ing babies in Strange Situations not at all what she originally had in

mind for herself.^ But she would eventually return to language (as we will

see in greater detail in Chapter 24), and when she did, she did so in tour^

de-force fashion.
Main’s cheerful, occasionally breezy style and her girlish voice belie a
consuming dedication we tend to associate with driven scientists. When
Main moved to the University of California at Berkeley in the mid'
1970s, she and graduate student Donna Weston began an ambitious lon^
gitudinal study of middle-class families. Children were assessed at twelve
or eighteen months for security of attachment to both mother and father.

In the spring of 1982, with the children in her longitudinal study now six

years old. Main and her colleagues — graduate students Carole George,
Nancy Kaplan, and a young English assistant named Ruth Goldwyn
brought them in for a two-hour assessment. Forty families participated.
Everything was videotaped.
One segment of the study was conducted by Nancy Kaplan. A Sarah
Lawrence graduate who had recently spent two years working with psy^
choanalyst Sibylle Escalona observing prematurely born infants in the
South Bronx, Kaplan, today a practicing clinician, had hoped by study-
ing with Main to combine her interests in clinical work and research.

“My wish,” she says, “was to try to get at the unconscious, to explore the
sorts of things that all clinicians work with but that are very hard to nail

down empirically.”^ Kaplan showed each of the six-year-olds a series of

photographs. The photos depicted children who were undergoing separa-

tions from their parents. The parents are saying good night at bedtime,
going away for a weekend, bringing the child to the first day of school, or,

in the most extreme case, going away for two weeks. They were asked
what the child in each of the photographs might feel and do in response

to the imagined separation.

Once again the three major patterns emerged — secure, avoidant,

ambivalent — and Kaplan found that for 79 percent of the children she
was able to accurately surmise their original attachment classification

with their mother by how they reacted to the photographs. (Early quality
212 The Fate of Early Attachments

of attachment to father, taken alone, proved not to be significant.) Some

of the children who had been rated securely attached to their mothers at

twelve months spoke touchingly of the sadness, loneliness, or fear associ-

ated with separation, and some had constructive ideas about how the
pictured child might respond. When asked how the child in the picture

feels, one boy said, “Sad, because they are going away. If he had permis-
sion to, he could follow them.” When asked what the child might do, he
responded, “Make sure that his Dad or Mom give him a phone number
that he knows just something goes wrong.”"^ In recognizing that
in case if

the separation was unhappy but not beyond his management, this boy
—
the only child in the study who responded with such exemplary security
— exhibited to an extraordinary degree the balance between attachment
seeking and self-reliance that Bowlby took as the hallmark of healthy
functioning.
The securely attached children were sometimes able to relate the
experience of the child in the picture with their own experiences: “I’ll
tell you about this picture,” said one six-year-old. “I know ’cause when I

started school at my first kindergarten class, boy, I was just like this

girl!. . . 1 mean, I just didn’t want to go to school. At first, I was happy,

but then I got sad I was gonna miss my mom and I just didn’t know
about school and couldn’t go to kindergarten yet. This little girl feels sad

. . . ’cause she doesn’t really know everybody yet.”^

The securely attached children generally took their feelings seriously,

but did not find them too upsetting to talk about. Thus, one six-year-old

said the pictured child would stamp her feet “’cause she’s mad.” Another
simply said that she was “a little angry at her mom and dad ’cause they
are leaving her.”^
Avoidant and ambivalent children also noted the unhappiness and
loneliness of the pictured child but their responses were very different.
Avoidant children often seemed overstressed by the discussion and were
at a loss for what to do. One boy who had been classified avoidant five

years earlier, when asked how the boy in the photo felt, said, “Sad,” in a

high-pitched voice. When asked why, he responded more emphatically:

He FEELS SAD!!! IWhy?] Because his parents are leaving

Iwhimpering voicej. Oh, God, he doesn’t want to sleep all by
himself! He’s a crybaby. He doesn’t want to sleep all by
himself. [What’s he gonna do?l Quack, quack, stay home,
that’s all.
 The Black Box Reopened 213

Another avoidant child said of a pictured boy, “He feels sad because he
doesn’t know if his mom will come back or his dad will come back.”
What can he do? “I don’t know.” This response, “1 don’t know,” or
“Nothing” was typical of the avoidant children.^
It may be recalled that mothers of avoidant children tend to down'
play or be put off by attachment demands. When the children snub
them on reunion in the Strange Situation, such behavior suggests that
in moments of stress, when attachment needs are highly activated,

they’ve learned to shut their attachment behavior down. Their

response, like that of the young child who’s been sent away to the hospi'
tal for a week or more, seemed to be a defense against further disap'
pointment, as well as against the destructive power of their own rage.

At twelve months their defensive strategy seemed to work; they did

manage to cope with the Strange Situation separations fairly well, even
if they did need to put a clamp on their need to be comforted. But now,

at age six, their surface detachment no longer seemed so successful in

the face of attachment stresses. They seemed flooded with anxiety.

What’s more, their inaction had become solidified, as if infusing their
being. They could not even imagine a useful response to a painful sepa'
ration. They believed that an attachment stress could only leave them

miserable and paralyzed.

The ambivalent children seemed very similar at six to what ambiva'
lent children look like at twelve months, especially in the intensity of
their involvement with their parents and their contradictory impulses
toward them, often combining proximity seeking with rage. When one
ambivalent child was asked what to do about the pictured separation, he
responded:

“Chase them.” [Who?] “Mom and dad in his new toy car.”
[Then what?] “And then he’s gonna toss a bow and arrow and
shoot them.”

Another ambivalent boy suggested that the pictured child could “shoot
his gun at him.”

[Then what will happen?] “Then he’s gonna die.” [Who?] “His
dad.”®
214 The Fate of Early Attachments

The depth of the tragedy of such responses is reminiscent of the boy who
killed hismother because he couldn’t bear for her to leave him. In pain
over a temporary separation, some of these children imagine killing off
one or both of their parents and thus losing them for good. This anger,
which arises presumably because someone they love has repeatedly disap-
pointed them, is so overwhelming that it would seem they already feel,

when confronted with separation, as if that person is lost forever.

A fourth category of more disturbed children (see Chapter 17) spoke

not so much of sadness or loneliness but of inexplicable fear. They imag-
ined that they or one of their parents would be severely hurt or killed.
And their responses were often contradictory, strangely repetitive, accom-
panied by aggressive behavior, or expressed in nonsense language. Asked
what the child in the picture could do, one of these six-year-olds said,

“Probably gonna lock himself up.’’ [Lock himself up?] “Yeah,

probably in the closet.’’ [Then what will he do?] “Probably kill

himself.’’^

For some of these children it seemed that any attachment stress quickly

reached annihilating proportions.

What happened next in the Berkeley study was inspired by the work of
James Robertson, and in particular his 1971 film Thomas: Ten Days in
Foster Care, about a two-year-old whom James and Joyce Robertson had

taken in for ten days when his mother was in the hospital. Thomas had
what appeared to be a secure and loving relationship with his mother,
and Joyce tried to keep it alive by regularly showing him his mother’s

photograph. At first he kissed the photograph and handled it tenderly. A

few days later, he seemed to become nervous in its presence, fiddling

with something, his eyes downcast. By the end of his stay with the
Robertsons, he avoided the photograph entirely, moving away from it,

obviously anxious. Main found all this arresting. She reasoned that the

behavior of the mother in the photograph hadn’t changed, so it must

have been something in Thomas himself, something about his feelings
toward her, his sense of his relationship with her, and perhaps his sense
of himself in the context of that rekitionship. In other words, his internal
model had gone through a transformation, what might be called a malig-
nant transformation (although Thomas was gradually able to recover a
more favorable model after his mother returned).
 ^ a

The Black Box Reopened 215

Now, after being shown the separation drawings, the six-year^olds in

the Berkeley study were presented with a Polaroid photograph of them
with their parents taken earlier, when they’d first arrived in the lab. It

was introduced with the words, “But here’s a photograph of yourself and
your family, and you see, you are all together.”^^ Securely attached chib
dren tended to take the photograph readily and smile or show some
interest in it, while the anxiously attached were more inclined to pas-
sively avoid it or actively move away. Some clearly became depressed
while viewing it. “One child who had been playing cheerfully with the
examiner, took on an immediate depressed aspect and bent silently over
the photograph for 12 seconds.”^
Kaplan and Main believed that with these assessments they were tak-

ing an X ray of the children’s internal representations of their attach-

ment experience with their parents, particularly the mother —
representation of how attachment experiences lived within them as a
set of beliefs about themselves and others and about what is possible
for them to do within the context of an intimate relationship. Will
mom or dad be accessible if I need them? Will they respond to my
feelings? Do I have the power to affect them or to do other things that
will be helpful to myself in their absence? It was assumed that the atti-

tudes the children exhibited about the availability of their parents, the
usefulness of their anger, their ability to manage their loneliness, would
be carried into other relationships — indeed, already had been, thus
accounting for the Minnesota findings on how they behaved with peers
and teachers.
Until this time, despite Bowlby’s and Inge Bretherton’s writings on the
child’s internal working model, the differences between anxious and
secure children were largely understood in terms of the different behav-
iors they exhibited in the Strange Situation or in the various longitudi-
nal follow-ups. Main said, in effect. No, the behaviors are not the key;

they are just a manifestation of the way the child has mentally encoded
information relevant to attachment; it is the psyche we want to look at.
The child’s early attachment experiences, she said, cause him to establish
an internal model that organizes and directs not only his feelings and
behavior “but also attention, memory, and cognition,” to the extent that
such mental functions are related to attachment. As a result, people with
different attachment histories not only have different patterns of behav-
ior but different “patterns of language and structures of the mind.”*^
Main believed that the internal model reveals itself in different ways at
 —
216 The Fate of Early Attachments

different ages, but that it is always there and that it is a pervasive force in

each person’s psychological make-up.

The idea that what people say about themselves and their feelings

could be scientifically meaningful had fallen into almost irretrievable

disrepute since the early days of the field in the last century when intro-

spection was believed to be a legitimate research tool. Behaviorism had

so thoroughly disparaged the idea, likening the mind to a black box the
contents of which could never be scientifically investigated, that it

seemed almost inconceivable to base a study on such data. But Main

believed that we reveal our inner life in what we say, and she has devoted
much of her career to proving it.

How does the way a child talks about an imagined separation reflect
his expectations about and close relationships? What can we learn
life

about how an adult approaches love and connection based on how he

talks about his childhood? How does the way a parent remembers his

own childhood become reflected in his behavior toward his own child,
and how is it reflected in the child’s quality of attachment? Nothing like
this had ever been attempted before; and, especially with the emergence
of the material from the adult interviews, which came out gradually over
the course of the next decade, it represented a second revolution in
attachment studies.

When Kaplan had finished with the children and the interviews with
the parents had been completed, the parents returned to the playroom
first one, then three minutes later the other — the total reunion period
lasting about six minutes. Jude Cassidy, a former ballet dancer and now a

student of Ainsworth’s, came to Berkeley to work on the project that

summer. A determined, Florida-born young woman with exceptional
qualities as an observer and a voice that becomes soft and lilting when
she talks about children, Cassidy’s job was to make sense of the reunions.

For babies and mothers in the Strange Situation, the reunion had, of
course, provided stunning information. But these were six-year-olds.
They were going to school. They were accustomed to separation. And, in

any case, they were not likely to wear their feelings on their sleeves.
They were not going to crawl, or cry, or beg to be picked up, or seek com-
fort in any obvious ways. Main acknowledged that she had doubts about
whether anything could be found, especially since the behavior of these
families when reunited all appeared to be within a normal range of
acceptable parent-child interaction. But she gave Cassidy some of the
 The Black Box Reopened 217

tapes, put her in a room with a video machine, and told her to look.
“And so I looked at them. I looked at them a thousand times. That
was all I did that summer, working fulhtime. And at first it really didn’t

seem like there was much there. But I kept looking and kept looking.
Finally, differences just started to emerge.”
The lack of overt variability in behavior meant that Cassidy had to

keep studying the five hours of videotaped reunions, hoping to find some
underlying patterns that would differentiate the relationships. “First I

looked at all the tapes, there were about sixty or seventy, until 1 knew
them really well. Then I found a tape that I just thought. This dyad” (or
mother- infant pair) “was really secure. It looked like a really great dyad.
It made me feel great just to look at them. The child seemed glad to see

the parent, very comfortable, casual, interested in interaction and con-

tact, but in a real calm, subtle kind of way. If the parent asked a question,
the child would do things that we came to call ‘expansions.’ These kids
were not just answering, they would say, ‘Yeah, plus, you know what? I

did this with them and then I did that.’ Or inviting the parent to join in

play. And so, on this first one, I said to Mary, If there’s a secure kid in the

whole bunch, it’s got to be this kid. So she went and looked it up and
said, you’re right, it is a secure kid. Then she said. Try to find some other
kid, somewhere, that you can label.

“So I found a kid who, the mother was asking questions and she
wouldn’t answer. And it was really painful and embarrassing. The mother
was trying to get something going, and the child just would not answer.
So I thought. Okay, this is avoidance at six. And Mary looked it up and,
in fact, it wasn’t avoidance. It had been a disorganized child. So I said,

Hmm. But later found a child who was being very neutral. She was
answering her mother’s questions, but
I

just maintaining neutrality

— ‘fine,

thank you’ — no expansion, no invited play, but not rocking the boat by
being really obnoxious about it. So I said, maybe this is avoidance at six,

this sort of cool neutrality. So Mary looked it up, and, yes, the child had
been classified avoidant as a baby.
“So I continued through the tapes like that. I said, let me go find
another kid who looked like that kid who was secure, who made me teel

the same way. And I did. Then Mary would tell me. Yes, that had been a
secure child, or. No, it wasn’t, it was something else, and we built a sys-

tem based on my mistakes and my correct hits.”

Working in this way, Cassidy eventually labeled every one of the chil-

dren. In 85 percent of the cases the classification she gave the child with
218 The Fate of Early Attachments

mother or father was the same classification the child had received at

twelve months. When I interviewed her eleven years later, it was appar^
ent that that summer, working with Main, had been one of the peak
experiences of her career.
“The thing that was so wonderful was that there were these meaning'
ful parallels. A secure baby wants contact with his mother and wants to
show her that he’s interested in her, that he missed her, that he’s glad
she’s back, and that he wants some comfort. And for a six'year'old, it’s

the same thing, but it’s displayed differently. I think about half of the six'

yeat'olds who had been secure babies at some time or another touched

their mother or their father during the reunion. So they were interested

in physical contact, but it wasn’t. I’ve got to sit in your lap. It was so sub'
tie, and it was so cute. This one little girl, her daddy was sitting on the
floor; he had his legs stretched out in front of him, and his foot was kind
of near her, and she just reached out and took the toe of his great big
size'fourteen shoe and wiggled it back and forth for a second and then
went back to playing. A boy needed to get a cookie, which was on the
other side of this father, and he had to lean across his father, and as he
did, he just put his hand on his father’s knee, like for balance, and then
he reached across and came back. And then it was gone. It was no big
deal. But in that instant he displayed a comfort with his father’s body. Or
one little girl shows a picture to her mother, and she leans in and puts
her hand on her mother’s shoulder, and their faces are side by side, just a
few inches apart. No child classified as avoidant did that.”^^

On the whole, Cassidy found that the anxiously attached children

showed little flexibility in their communication with their parents.

While secure children and their parents treated each other in a relaxed
and friendly way, readily engaged in conversation on a variety of topics,

and fell into an easy intimacy that was not at all clingy, the anxiously

attached children communicated far less, with little spontaneity, and a

very restricted range of feeling and subject matter. The ambivalent chib
dren, not surprisingly, tended to mix intimacy seeking with hostility, and
they often seemed affectedly cute or ingratiating. Avoidant children also
showed a pattern reminiscent of how they’d behaved at twelve months
in the Strange Situation. They tended to keep their parents at a distance,
to keep greetings brief, and to choose strictly impersonal topics ot con'
versation. By keeping busy with toys, they were able to ward off whatever
advances their parents tried to make. The disorganized children were
most likely to try to control or dominate the parent, by acting in an
 The Black Box Reopened 219

either rejecting or humiliating way, or to strangely reverse the roles and

becoming solicitous and parental. One can imagine how such anxious
patterns of relating to parents might become amplified a few years later

in the cauldron of adolescent emotions.

Sroufe’s studies had shown that in their relationships with playmates,
children tend to perpetuate aspects of the relationships they had learned
at home. It was certainly suggestive of an internal working model, but
this study added a new dimension. For what Cassidy was able to show
was that the children, in their relational behavior, were no longer merely
responding to their parents’ leads. With their mothers and fathers, they
were now full partners in the relationship, equally responsible for its tone
in the current moment. The child’s model of relating, as Bowlby asserted,

was no longer owned by the parent or hovering somewhere between

them. It now fully belonged to the child.
While Kaplan studied the children’s responses to imaginary separa-

tions, as well as the family drawings they’d created, and as Cassidy pored
over videotapes of the reunions, Ruth Goldwyn spent her summer at'

tempting to make sense of the interviews that had been done with the
parents. The results of her work with Main on those transcripts would
not only demonstrate the intergenerational transmission of attachment
styles but also provide a way of understanding how attachment experi'

ences reach their final stage of mental and psychological incorporation

in adulthood. Of this, we will speak more later. All told, this single piece
of research opened attachment studies to a new level of discourse, closer

perhaps than it had ever been to its psychoanalytic roots — although it

would still be some time before analysts recognized it.

 —

Thev Are Leaning Out eor Love:

The Strategies and Deeenses
OE Anxiously Attached Children,
AND THE Possibilities eor Change

The behavior of six^year^olds in the Berkeley reunion, like the behavior

of one-year^olds in the Strange Situation, suggested that insecurely
attached children have developed unconscious strategies for dealing
with mother’s neglect, rejection, or inconsistency and with the intolera-
ble furies and hatreds that boil up within them in response. It is haz-

ardous to make assumptions about what is going on in the minds of small

children and even more hazardous to generalize too broadly, especially
when influences are coming from so many different directions, both
inside and out. But certain patterns do suggest themselves when a partic-

ular style of anxious attachment becomes entrenched and thus a key fac-

tor in the person’s developing psychology.

The ambivalent (who represent about 10 percent of middle-
children
class samples, 20 percent of Sroufe’s poverty sample)^ seem to be desper-

ately trying to influence their mother. Many of them seem hooked by

her haphazard, unpredictable style and the fact that she does come
through on occasion. They pick up that she will respond sometimes
perhaps out of guilt — if they plead and make a big enough fuss. And so

they are constantly trying to hold on to her or to punish her for being
 ^ —
They Are Leaning Out for Love 221

unavailable.They are wildly addicted to her and to their efforts to make

her change, they become enmeshed with her in various unhealthy ways,
and later in life they become similarly addicted to other potential
attachment figures, such as teachers in the school years and, in all prob-

ability, romantic figures after that. But through it all they do not believe
they have what it takes to get what they need from another person.
Ambivalent children fret for themselves, and, often, they fret for their
mother: Where is she? What is she doing? Is she okay? Not surprisingly,

some of them become what family therapists have called “parentified”

that is, caretakers for their own parent. Bowlby argued that such extreme
ambivalence is the true source of school phobia, not because, as is com-
monly thought, the child is afraid of school, but because he fears that he
will lose his mother or that she will be unbearably lonely if he dares let

her out of his sight.

According to Main, the predominantly ambivalent child emphasizes
his feelings of helplessness in order to elicit care. He learns to scan the

environment in search of threatening elements that will enable him to

become fearful and thereby get attention. Gradually, the ordinary takes
on a frightening cast, leading in some cases to chronic fretfulness or anx-

iety.^ Meanwhile, he is highly attentive to attachment-related signals, to

any hope that some succor and connectedness may be in the offing or to

any danger that it may be removed. This extreme attentiveness saps

energy that might have been available for play or work, compromising
intellectual development and giving the personality a dependent and
highly emotionalized cast. Ambivalent children are not turned off. Their
longing for connection is always on “high.” But they often alienate oth'
ers with their impulsiveness; and, in those relationships where hope is

kindled, they may alienate them with their hypervigilance and clinging.
Attachment theory does not tend to address itself to the cauldron of

rage and aggression, including in some cases sadistic and masochistic

aggression, hatred and envy, self-hatred and persecutory anxiety that
must churn within these children. This is partly a legacy of its coming
of age within developmental psychology, where the emotional life has
not traditionally been examined in depth, and partly a legacy of John
Bowlby’s limitations in this realm, as exemplified by his split with
Melanie Klein. Klein’s main concern was studying just such churning
passions and the relationship fantasies that accompanied them, which
she saw as prominent in every small child. One of her key early insights

was that extreme passions are inherent in our make-up and a natural
222 The Fate of Early Attachments

response to feelings of deprivation or persecution, real or imagined.

Much to Bowlby’s horror, of course, Klein ignored whether they ivere real
or imagined and seemed to care little about the environmental condi'
tions that might promote such violent inner storms or stand in the way
of the child’s mastering them to the extent that love could prevail. But

Bowlhy’s political differences with Klein notwithstanding, he was, as 1

have said, not drawn to this subject matter. He seems to have successfully
lidded his own cauldron and was not eager to get reacquainted with its

contents. Not that Bowlhy entirely ignored aggression: He knew it was

important. The reader may recall that years earlier Bowlhy had discussed
how much the child needs to be able to express hatred and aggression
toward the parent (see Chapter 4). Being allowed to have these feelings
without being overwhelmed by guilt or anxiety helps the child to accept

the ambivalence that is part of every close relationship and gives him the
confidence that he can control his negative impulses, that they do not
have to destroy him or those he loves. Although this early insight of

Bowlby’s has not been adequately applied to anxious attachment (for an

exception, see Kobak’s thinking mentioned in the following chapter), it

seems obvious that an avoidant or ambivalent child would lack just such

confidence. How can he satisfyingly voice his belligerence to a parent

who is too dependent on the child’s glowing love, who cannot tolerate

rejection, who feels turned off by need, who caves in to depression when
the child is tantrummy or oppositional, who desperately needs the child

to be a perfect narcissistic reflection of herself?

There is another implication here, too, perhaps especially for the

ambivalent child, whose hurt and rage and hatred are so volatile and so
quickly unmanageable: He never develops the sense that mom is there to

contain his overwhelming emotions; that he can have a tantrum; that he

can hate her and feel as if he and mom are through, hut that she will he
soothing and convey the sense that the tantrum will soon pass without
causing permanent damage and that even his wish to annihilate her will
not have devastating consequences. In other words, even if his extreme
negative feelings are too much for him, they are not too much for her;

she can (in Winnicott’s word) “hold” them, and through his relationship
with her he will learn to manage them one day himself.
What then becomes of the amhfvalent child’s rage, a rage that already

at age one in the Strange Situation was so disorganizing that the child
could not take in the comfort that the mother offered and that the child
himself so desperately sought? Studies do not exist to answer this queS'
 They Are Leaning Out for Love 223

tion, and even if we would probably find so many variables

they did,
within the ambivalent syndrome as to make a single answer impossible.
But work with patients in psychotherapy does offer some clues regarding

the internal processes of at least some of these children. In these paradig-

matic cases, where the mother behaves in just the way expected by
Ainsworth and Main, the relationship with the mother herself probably
remains stormy, because the child is still desperately trying to get her to

be what he needs her to be, which means in many cases, 1 think, not just

getting her to be available, accepting, attuned, but also to contain his

distress, to perform the holding function that the child has still not
learned to perform for himself even though he may be technically capa-

ble of it.

In many cases grisly power struggles will follow, leaving this ambiva-
lent child burdened with unconscious fantasies of mutilation, dismem-
berment, retaliation, and desertion — the sorts of fantasies that disturbed

young children revealed to Klein in play therapy. These unconscious fan-

tasies about what happens in close relationships will haunt the growing
child’s sense of self, making him feel shamefully unworthy of being close
to others without quite knowing why. And they stand ready to corrupt

future intimate relationships, for the fantasy does not just relate to me
and mommy, but to me and anybody with whom am close
1 or by whom 1

long to be loved. Because most ambivalent children are so anxious about

abandonment, their aggression toward others is, 1 suspect, most apt to
come out in disguised or passive-aggressive ways, with extreme outbursts
still being reserved for mom or for those moments with highly valued

others when there is too strong a whiff of betrayal. Meanwhile, 1 suspect

that many ambivalent children remain prey to persistent feelings of envy

and resentment which further foments their rage and their sense ot

themselves as poisonous. (We’ll talk more about the ambivalent style in

adulthood in Chapters 24-26.)
Ambivalent attachment is a very broad category and can take up resi-

dence in the personality in an endless variety of ways, but consider the

psychological dynamics in some adolescent girls who suffer from ano-
rexia. Often, the anorectic stance grows directly out of the impossible
power relations the child has had with the mother, including feelings

of persecution, neglect, and cruelty. In many cases the anorexia devel-

ops into an angry refusal not only to take in food but to take in any car-
ing, any emotional nutrition, from another person, even though, deep
down, that is what is most wanted. This refusal, so reminiscent of the
224 The Fate of Early Attachments

ambivalent baby’s reaction in the Strange Situation when she seems

unable to absorb her mother’s efforts to comfort her, is, among other

things, a way of hating, a subtle but powerful form of aggression, that can
last even after the eating disorder itself has passed. Any therapist who
has tried to jump through hoops to nurture a patient like this only to be
thwarted by confused grimaces or shrugs, or who has experienced a
Sisyphean sense of losing the very connection with the patient that he
confidently thought had been built up over previous months or years,
knows eventually that the anorectic is not just starving herself, she is

starving those who love her, too. By refusing to connect — or to stay

connected to her own connectedness — she is acting out, albeit uncon^

sciously, a form of sadism.
When 1 have seen women in treatment who 1 assume have had am-
bivalent relationships with their mothers, they usually have a strong
identification with their mother as well. The identification is another
aspect of the child’s, and later the adult’s, fantasy^emotionahrelational
life, with the result that many of these women feel doomed to the worst

aspects of the mother’s destiny, including in some cases a bitterness

toward men or toward fate in general.

With his bids for loving contact repeatedly frustrated and sometimes
angrily rejected, the child with a history of avoidant attachment to his
mother (about 20 percent of middle'class samples) finds himself in a
special box. He doesn’t feel he can be openly angry with her, despite the

fact that anger, according to Bowlby, is the natural response when a

child’s attachment needs are thwarted. Experience has taught him that
his anger will only cause her to become more rejecting. And so he has
learned to turn himself off. At the slightest hint of pain or disappoint'
ment, he shuts down his attachment system and experiences himself as

having no need for love. Unlike the ambivalent child, whose attachment
antennae are always up and receiving and who seems to have no defenses
to ward off painful emotions, the avoidant child. Main believes, has

made himself deaf to attachment related signals, whether they are com-
ing from within himself or from someone else. He avoids any situation
and perhaps any topic that has the potential for activating his attach'

ment needs.
Even who seems to have accepted life on the
later on, as a child

edge of human connectedness, who seems to many observers to prefer

detachment, the prospect of further rejection is too terrible to risk. The
 They Are Leaning Out for Love 225

predominantly* avoidant child cannot be warmly affectionate with his

mother or go to her when in need. But by keeping his attachment system
dampened, he is at least able to stay near her without risking more pain
or ruining the connection with his disappointment and anger. Thus,
despite appearances, the strategy of the avoidant child still seems to serve
the purpose of preserving proximity. Psychologically, he is firmly in his

mother’s orbit, his thought, feeling, and behavior shaped by the claims of
that relationship, but, like Jupiter or Uranus, he abides at a distance that
affords him little warmth.
This constricted way of life, in which positive displays of feeling are

curtailed, can often be managed, especially if the avoidant child develops

solitary interests. But when he’s most distressed and needing care, his
efforts to minimize or turn off the strong, biologically based attachment
needs emanating from within can be a terrible struggle, which leaves him
stressed to his limit. Now the rage that he feels toward his rejecting
unavailable mother is no longer so easily repressed. He becomes taut; his

trigger is held by a thin hair; and it is possibly at such moments that he is

most likely to unleash his anger at other children.^

To the outside world, the avoidant child seems to say. Who needs you
— 1 can do it on my own! Often, in conjunction with this attitude,

grandiose ideas about the self develop — I am great, I don’t need anybody

— which have little in common with secure independence. For some

parents it is useful to promote such grandiosity in the child. If the mother
can convince herself that the child is superior to other children because
of his amazing autonomy, she can feel reassured about her own lack of
nurturing attention: This kid is special, he barely needs me, he’s been
doing his own thing practically since he was born.
not hard to imagine the fury that engulfs a child whose attach'
It is

ment needs are so profoundly frustrated. We know that the avoidant

“1
child seems to abandon his overtly rageful stance in favor of a defiant

don’t need you!” But when anger toward a parent is stifled, as it often

is, for instance, when a parent dies, the child tends to turn the anger

against himself, creating a stubborn pocket of guilty or shame^ridden

*1 use thisword predominantly to emphasize that there are degrees of anxious attachment. In general-
izing about ambivalently and avoidantly attached children, am talking about children who are very
1

solidly in an anxious category. Even then it should be remembered that

all children are different,

each with special vulnerabilities that may pull them into more disturbed realms and special
strengths

that may pull them out of an expected disturbance. And they may have
other attachments with

other adults that gives them alternate modes of relating.

226 The Fate of Early Attachments

depression. The depression may be masked by compulsiveness, aggres-

sion, or other symptoms. Meanwhile, the “I don’t need you!,” begun as
a rageful, wounded pout, now becomes an unconscious stance toward
people in general, even loved people. This too is a form of selhstarvation
and is probably intensified by an identification with the depriving par-
ent, such that the avoidant child now becomes the agent of his own
deprivation.
As with the ambivalently attached, it is hard to imagine the avoidant
child or adult being able to express anger while simultaneously retaining
a feeling of love and connection. We still have a lot to learn about why
various children become avoidantly attached to their mothers, but to the

extent that the relationship fits the picture that emerged from the
Baltimore study, we can assume that in the avoidant person’s fantasy, to
express anger over one’s hurts or unmet needs is to be rejected, swatted

down, disparaged. His anger is, therefore, likely to remain unexpressed

(except in outbursts of temper), because the consequences of expressing
it are too terrible. Instead there may be an unspoken withdrawal “I
—
don’t need this person any more; he’s history” which may in turn be —
covered by an outward amiability that suggests nothing has happened.
This is not to say that avoidant anger can’t also be cold and dooming
(perhaps like the parent’s) or even murderous, but it will rarely be
accompanied by a sense that anger and love can go together: that I can
hate you and still know I love you, that 1 can be confident even as I’m

fuming that things will be right with us. The avoidant personality thus
feels that his anger is always putting his relationships in jeopardy. In a
sense, it could be said that he is in hiding from his own attachment
storms and the inner representations that go along with them.
The avoidant child may be afraid to get too close. He may unthink-
ingly wander away from his connections, unaware that he is starving
himself in the process. Or he may become openly rejecting, especially if

he has been rejected by the parent with whom he is most identified; and
his rejecting style may be sadistic or exploitive in the manner of that par-

ent. As with ambivalence, the ways in which avoidant rage can get
played out, both in fantasy and behavior, are endless, depending on the
unique set of circumstances with which the child is presented both in
infancy and later on.

Given the distortions of thought, feeling, and relationship style inherent

in anxious attachment, there is understandably a great deal ot concern

 They Are Leaning Out for Love 227

about the tenacity of attachment patterns. That they often seem to per^
sist through childhood and into adulthood is hardly surprising,^ if for no
other reason than the fact that the environment most children start out
with doesn’t change. Early anxious attachment can be assumed to arise
from several sources: trying circumstances, such as an overtaxed, under-
supported mother not having the time or the peace of mind to be sensi-
tivelyand consistently available to each of her children; ignorance,
which might cause otherwise caring parents to let a baby cry for pro-
longed periods, to leave him repeatedly or for too long a time before he is
able to handle it, or to become prematurely concerned with training for

independence; unhappy events in the child s life (deaths, separations,

sibling rivalries), which might create emotional problems the
parent

cannot handle; some innate need in the child that the parent is unable
its way
to satisfy; and, finally, parental psychology, which easily works

into and complicates the other conditions. Indeed, what

begins with

ignorance, unfortunate circumstances, or infant need is often prolonged

by psychology, as the parent reacts or overreacts to the child’s anxious

pattern.
Circumstances and ignorance, of course, can change much more read-
ily than parental psychology. But, even if the
miraculous happened, and

parents with all their deeply ingrained psychological baggage were

sud-

denly liberated and able to let their love flow without the anxieties, con-

flicts, and poor habits that have been built up and reinforced
sore points,
the
over the course of a lifetime, the child might have a hard time seeing
change. The avoidant child doesn’t want to be tempted to open
himself

to hope or trust when he’s worked so hard to close himself down and
when so much is at stake in terms of the agony of renewed rejection.
Indeed, he may feel the temptation itself is so pernicious, arousing

agonies of renewed ambivalence, that the person who seeks his trust

may become the object of his sorry rage. The ambivalent child, mean-

while, whose care has been inconsistent or chaotic, cannot believe that
a caring gesture is any more than a passing fancy.
So he is likely to keep
testing and testing, keep mixing clinging with hostility
and unreasonable
demands, perhaps driving away the parent who wants to initiate some-

thing new.
Throughout their early years insecurely attached children are believed

amenable to change. Avoidant children, for example,

to be relatively
will seek attachments with teachers and other adults, and, if they are
lucky, they will find a special person who will
provide them with an
 .

228 The Fate of Early Attachments

alternative model of relatedness. Research has shown that if they are

securely attached to their father (or to another secondary caregiver), that
will he the best insurance in helping to overcome insecure attachment to
the mother. But even if it’s only an uncle they see occasionally, the
knowledge that he cares will keep a different model of relatedness alive
in them. Indeed, studies of resiliency indicate that a child’s having had
such a person in his life can make a significant difference in his ability to

believe in himself and to overcome adversity.^

But it may be hard for the insecurely attached youngster to find such
an alternate attachment figure because the strategies that he has adopted
for getting along in the world tend to alienate him from the very people
who might otherwise be able to help. The behavior of the insecurely
attached child — whether aggressive or cloying, all puffed up or easily
deflated — often tries the patience of peers and adults alike. It elicits reac'
tions that repeatedly reconfirm the child’s distorted view ot the world.
People will never love me; they treat me like an irritation; they don’t
trust me; or, I always feel that 1 need them more than they need me.
But if adults are sensitive to the anxious child’s concerns, they can
break through. Sroufe believes that teachers in particular can be trained
to bypass the child’s resistance. He cites the example of an avoidant
child named Eddie. Eddie’s “acts of hostile aggression toward weaker
children, his devious behavior, and his blunt, matter^oTfact noncompli'
ance (‘No way, Jose!’) infuriated’’ the teachers. According to Sroufe, his

malicious, antisocial behavior, his apparent pleasure at others’

distress, and his fearless contesting of wills with the teachers
had all the marks of incipient sociopathy. His elaborate devi'
ousness and his swaggering style seemed to confirm that he
was beyond reach. But in fact, of course, no four year old can
genuinely be a sociopath. And beneath the swagger was a des-
perately needy child. . .

Sroufe instructed the teachers not to punish Eddie but to be patiently

available to him despite his provocations.

The teachers quickly learned to see opportunities for closeness

with [Eddie] and explicitly disconfirmed his feelings of low
self-worth by not rejecting him. As often as possible, the
teachers prevented [Eddie] from engaging in hostiie behavior.
 —
They Are Leaning Out for Love 229

And when it was necessary to separate [Eddie] from other

children, which was often the case early in the term, a teacher

would stay with him.

Eddie soon stopped being a behavior problem, formed a strong attach'

ment to one of the teachers, “and made remarkable progress toward
learning to meet his basic needs for closeness.”®
How much of this Eddie will be able to carry over to other relation'
ships is hard to say. But the difficulties involved are suggested by another
avoidant Minnesota child with whom the same strategy was applied.

About fifteen weeks into nursery school this little girl, who had become
very close to one of her teachers, had a dream. “In the dream,” Sroufe
related, “her teacher hurled her against the wall. She told the teacher the

dream and the teacher said to her, ‘I would never do that.’ And it

absolutely shocked me — I was in an observation booth hearing this

the little girl asked the teacher, ‘Why wouldn’t you do that?’ Eor her to

understand a nurturant, responsive relationship was beyond her mind at

that point.

One way or another, it would seem important to reach insecurely

attached children by adolescence, because that’s when it is believed their

patterns become more firmly set. Even then they can still be changed;
there is still the possibility of psychotherapy, not to mention other vital

relationships, and the emotional flux of the adolescent years sometimes

opens children up in new ways. But clinical experience strongly suggests

change is most easily accomplished during the earlier years, when a
steadfast parent or an available teacher can turn a child around.
In childhood, a strong component of provocative behavior, like Ed'

die’s, is a plea for the adult world to be different. “Love me, show me your
love, change,” the child is saying. The child wants to have his pain

soothed, he wants the relationships with those he loves to be whole

again, and he is ready to forgive. This is still true for the adolescent,
and

even for the adult, but much more committed to

with age we become
keeping things as they are. As we will see later, when exploring adult
attachment, bitterness and resentment become hardened into a new kind
of security, and the hope for change is dimmed. A small child like Eddie

may seem difficult but is wonderfully open and flexible by comparison.

Parents, of course, have great difficulties with the negative patterns

their anxiously attached children develop. The ofTputting communica'

230 The Fate of Early Attachments

tive styles described by Jude Cassidy among the middle^class six^year-olds

in Berkeley, not to mention the more disturbing patterns seen among the
underprivileged Minnesota children, can be tormenting to parents. Some

may convince themselves that nothing serious is amiss or rationalize

what they see as an inevitable part of growing up: Kids are difficult, they
don’t get along with parents, and that’s just the way it is. Others only see
the child’s contribution, are unhappy with his glumness, artificial or
cloying sweetness, hostile defiance, withdrawal, or manipulation, are
hurt by his lack of open affection, and make things worse by pulling away
or coming down on him for it. Indeed, as I’ve suggested, one of the
tragedies of the development of anxious attachment in the first year is

that it generates behavioral styles in the child that not only disappoint,
annoy, or infuriate his parents but that pushes the parents’ buttons,
bringing into play (or further into play) disturbed aspects of their own
psychology.
Virtually every attachment researcher has noted the melancholy con^
sistency with which parents of insecurely attached children misinterpret
their behavior. According to Sroufe they often feel depressed and dis^
couraged because they believe their kids don’t love them. Nothing, of
course, could be further from the truth. Virtually all children, even

abused children, love their parents. It’s built into the nature of being a

child. They may be hurt, disappointed, caught in destructive modes of

being that ward off any possibility of getting the love they yearn for, but
to be attached, even anxiously attached, is to be in love. Each year the
love may become a little more difficult to access; each year the child may
disavow his wish for connection more firmly; he may even swear off his

parents and deny that he has any love for them at all; but the love is

there, as is the longing to actively express it and to have it returned, hid-

den like a burning sun.

Even a mother who has sought therapy, who’s found a stable mate,

who’s overcome distracting financial distress — who has in one way or

another become ready or able to take a more responsible, consistent, nur^
turant role toward the child — may find it hard getting through to the
child who has adopted such survival strategies. She may find it difficult,

for example, to convince him to give up his angry, cut-off stance and be
open to receiving love from her again; or to let go of the clinginess, the
guilt, and the power struggles and trust that mother has changed, that
she will not neglect him this time, that they can be separate people and
she will be there to meet his needs. Getting this message through takes
 They Are Leaning Out for Love 231

much patience and consistency, until the child builds up a new set of

expectations, a revised model of the way love works.

When the child actively rejects the parent’s advances, the parent’s job
is complicated. For he must not only keep faith in the possibility tor an
open and loving connection and newfound responsiveness
persist in his

in the face of painful rejection, but he must also deal evenly and fairly

with destructive aspects of the child’s behavior. The mother or father

whose own emotional problems were responsible for the anxious pattern
to begin with finds both ends of this task difficult. To keep loving in the
face of rejection is contrary to the parent’s own internal model. To want
to hug a distressed toddler, only to have him squirm away, to speak lov-
ingly to the six-year-old, only to see him look at the ground and fidget
impatiently, seems beyond endurance. By the same token, to be respect-

ful and to steadfastly demand respect; to discipline a child in a moderate

way; to neither indulge, seduce, manipulate, nor explode — all this is

contrary to emotional habits built up over a lifetime.

1 recall an incident from my own childhood. 1 believe 1 had an anx-
ious attachment to my mother, which was perhaps mainly ambivalent
but had certain avoidant features and looked more and more avoidant as
1 got older. My mother,
in turn, was, I think, avoidantly attached to her

mother, a hardworking but somewhat distant woman who attended to

everyone’s physical needs but was not especially warm. In those days my
friends and 1 played a game called Skelzy, which required shooting
weighted bottle caps around on a pattern we chalked on the ground. We
weighted the caps by melting bits of crayon into them, usually hy placing
them on a hot radiator. One afternoon my mother, who was busy else-

where in the apartment, had left water boiling for rice in the kitchen,

and I saw an opportunity. 1 placed the bottle cap filled with crayon bits

on the lid of the pot. To my chagrin, the lid tipped and everything fell in.

I called my mother in a fearful and guilty state. I fully expected a blast of

temper. Although I could hardly have expressed it this way at the time, I

felt reemerging from within me, as I awaited her response, a familiar

aspect of had been out of mind for a while but now seemed
my self. It a

deeper truth about me than whatever I’d been feeling a minute before: I

was an ugly problem child.

My mother has since told me that she felt guilty about her temper and

the hurt it might have caused. It seemed on this occasion that she

worked mightily to restrain it and be patient. In any case, I was surprised

to encounter a warm, slightly indulgent reaction, as if my effort to make

232 The Fate of Early Attachments

a Skelzy cap in this way was amusing if impractical and certainly a forgiv-
able error coming from a child she loved. 1 was lovingly sent on my way.

What 1 did next seemed like nothing so much as a confirmation of my

badness. 1 decided to try again. Again the cap and crayon bits fell into

the rice water, and this time my mother obliged me with the reaction 1

had originally anticipated.

This, 1 believe, is typical of the way in which a relational style is per^

petuated even when the parent makes a stab at something new. My

mother’s first reaction must have struck me as needing further testing. 1

had to know whether 1 could trust the change in her, whether it was
safe to see the relationship in a new way. Besides, it would be nice to

get that indulgent smile again. My mother thought I was a wonderful kid
— smart, independent, competent — and would have been incredulous at

the thought that beneath my confident smile I sometimes felt needy and
unhappy and completely uncertain about how or whether to approach
her. Meanwhile, she had her own demons to stmggle with, including this

irrational temper she could not control and didn’t know where it came
from. Now, feeling under assault by a taxing child who seemed bent on
ruining her afternoon, her temper was reemerging. If she had indeed been
trying something new, she had underestimated how much work she would
have to do to get me to see her and our relationship differently.

The avoidant child does not tend to come to his mother when in diS'

tress. This is another way in which the changing mother has a difficult

job. She may want to respond sensitively now, but she has fewer opportU'

nities to do so and is likely to be rebuffed for her efforts. The child does
not have a good feeling about opening himselt up to his mother when he
is in pain. He does not have a good feeling about letting her into his
domain of vulnerability. He associates such experiences with rejection

and shame.
Nevertheless, I do believe that the avoidant child offers his parents

opportunities for new beginnings. He is hurt by a playmate and makes an

oblique mention of it. The mother who has not tended to the child’s
hurts in the past has an opportunity to question him now in a sensitive

way and to sympathetically speak his unspoken feelings — suggesting for

the first time that they are not shameful hut a natural human response
that can he shared with another person. The child may not seem to like

the experience or to accept whatever words are offered, hut a second

opportunity may arrive after a shorter At some point the
interval.

avoidant child, trusting in the mother’s availability, may spill his hurt
 They Are Leaning Out for Love 233

and insecurity in a shocking way: The adolescent, having fallen in love

and been rejected, for instance, may reveal that he feels thoroughly

dependent and weak. This is a challenge for the mother to hear, for these

are the qualities that rankle her. But it is an opportunity — not only to

lovingly explain that there is nothing wrong with wanting to be loved or

to be hurt when his love is spurned, but also to give him a rewarding

experience of depending on her, an experience that could affect his

model of what relationships can be about.

That these opportunities can be assisted and promoted by psychotherapy

is strikingly revealed by a case reported by child psychotherapist Juliet

Hopkins. Hopkins, a quietly attractive woman with soft, penetrating

eyes, is Bowlby’s niece, and, like him, she too found a professional home
at the Tavistock Clinic. Trained to work and think psychoanalytically,
she finds that attachment concepts add a valuable layer of meaning to
her work, as is apparent from her report on a young patient named Clare,

who was referred to the clinic when she was six years old.
Clare’s mother had not wanted her. She had found her physically
repellent as a baby, prop fed her, and confined her to a playpen until she

began nursery school. The mother was not totally rejecting, however,
and did gradually come to enjoy talking with Clare and playing with her.

She believed that Clare resisted cuddling.

As soon as Clare could walk [Hopkins wrote] she walked away

from her mother and was liable to get lost. She had always
been stoically independent, never asking for help except
when she had hurt herself. Her mother described her as “acci-
dent prone” and felt that she had accidents to gain the atten^

tion which she had no other means of seeking. Even when she
had hurt herself she did not cry.

Clare had the uncanny ability to reproduce her mother’s way of talk'
ing and gesturing, as if, Hopkins surmised, by becoming her mother she
could do without her. When she started treatment, she did the same with

Hopkins, interpreting her own behavior so that she would not need the
therapist. “Once, when 1 spoke to her of her hidden wish to cry, Clare

explained, ‘I never cry because if I started I would never stop.’ I under'

stood this to mean that she feared there would never he anyone to coni'

fort her.”
234 The Fate of Early Attachments

Clare was preoccupied by lepers, and her nightmares were inhabited

by them. She believed that they were contagious and that “if they touch
someone they She nevertheless thought “they could be cured by the
die.”

laying on of hands.” These fantasies spoke volumes to Hopkins about

Clare’s extreme, avoidant history.

As therapy proceeded Clare became aware that she felt herself

to he a leper whom no one wanted to touch because she

would kill them, and she also became aware of her longing to
cry and to be comforted. In her session she was tortured by the
longing to touch and be touched by me which conflicted with

her terror of it, and by her longing to cry which she fiercely

resisted because it was “so silly and only babies do it.”

After nearly a year of therapy an accident occurred that Hopkins

believed was intentional (if unconscious) on Clare’s part, allowing her
some of her longings into action.” Clare fell heavily from
“to translate
Hopkins’s desk onto the floor. “As she lay there she raised her arms
beseechingly towards me and burst into tears. I picked her up and she
sobbed for some minutes on my lap, though keeping her head averted
from me.”
The next day Clare’s mother told Hopkins that Clare had come to her

in distress, seeking comfort for the first time. “Mother, who had been
greatly helped by her own therapist, was ready to respond to Clare and
thereafter Clare continued to turn to her mother when upset and began
... to confide her worries to her. She became cuddly and, before long,
very clinging and demanding, and no longer accident prone.

If reaching and redirecting anxiously attached children is difficult as they

get older, the problem of repair is especially thorny for abused children,

because the messages they get and the working models of relatedness
they develop are more confused and more sealed off against intrusion.
“An abusing mother,” says Pat Crittenden, “tends to he fairly coercive

and demanding, even hostile, hut to come across almost sickly sweet.

She is unlikely to scream and yell at her child. She is far more likely to

paste a smile on her face and with gritted teeth demand that her child do

something. The child then learns to associate a positive expression of

feeling with a really negative experience. And so when he goes off to

school, or meets other members of his family, or maybe later meets a peer
 ^

They Are Leaning Out for Love 235

or a potential lover, he will misinterpret positive expressions of feeling.

He will assume that people who appear to be nice are being coercive.”^
If unanswerable attachment alarms go off for the avoidant child whose

mother is averse to holding and comforting him when he wants it and

makes him feel rejected when he reaches out to her, one can barely imag-
ine the painful dilemma faced by the child whose object of love thrashes
him, often at just the moments when he is most in need of connection. It

is hardly surprising that these children seem to fall into a fourth attach-

ment category, not part of the original Ainsworth classification system.

Labeled “disorganized/disoriented” by Mary Main, they display behaviors

typical of both the avoidant and the ambivalent baby.* In the Strange

Situation they seek proximity with the mother in a disoriented, herky-

jerky, semiparalytic way — they approach her backwards, or they freeze

suddenly in the middle of a movement, or they sit for a while and stare

into space. As they grow older, their mode of relating to others becomes
contradictory and infuriating, such that if they are eventually removed
from an abusing home, they fail in one foster care family after another.

Because love for them is fused with aggression, and because they are so

hardened against themselves — and therefore against other wounded

beings in pain — the rate at which they repeat the misdeeds done to

them with their own children is high.^^

Selma Fraiberg, the child psychoanalyst whose book The Magic Years

helped generations of parents to understand the process of early develop-

ment, attempted to address this problem by inventing a form of treat-
ment for mothers and their babies where neglect and abuse had already
become serious enough to have come to the attention of community
clinics, agencies, or courts. Into Fraiberg ’s Ann Arbor clinic came bleak
and cynical young mothers, who showed no reaction to their baby’s cries
or no interest in handling them. They often revealed such indifference
toward their infants that it was painful for the therapists to observe.

Eventually they were asked to recount their own experiences as children.

After repeated invitation, they told their stories, heartrending recita-

tions of beatings, abandonment, psychologically cruel punishments, and

painful disregard; but they told them flatly, as if reading from the phone

book. At one point a sixteen-year-old mother who often forgot to buy

*While some studies suggest that a significant portion of these hahies have been abused (Cassidy and
Berlin, 1992), Main also found that many of the mothers experienced tragic early losses that they
had never been able to properly mourn. The result, she believed, was a pervasive sense of anxiety

and fearfulness that was subtly communicated to the baby (Main and Weston, 1981; Main, 1991).
236 The Fate of Early Attachments

milk for her baby and fed him Kool-Aid or Tang cried, But whats the
always kept things to myself. I want to forget. I don’t
use of talking? 1

want to think.

As the mothers told their stories, the therapists intervened to speak

the missing feelings: How hard it must have been for you to be all alone,
how terrified you must have felt; how horrible not to be heard by any^

one. Gradually, “grief, tears, and unspeakable anguish” for themselves as

cast'off children began to emerge, and as this happened, the
wounded
mothers reached out to their babies. One mother, while pouring out her
long disavowed grief, picked up her baby, “held her very close, and
crooned to her in a heart-broken voice. Another mother, angry and
tearful as she reexperienced emotions she’d sealed up years earlier, spon-

taneously went to pick up her little boy, “enclose him in her arms, and
murmur comforting things to him.”^^ It was as if the mothers’ attach-

ment feelings, dammed up for many years, were flowing again.

Psychologist Alicia Lieberman studied with Ainsworth at Johns
Hopkins in the mid-seventies before going on to work with Fraiberg, first

in Ann Arbor and later at the Infant-Parent Program at San Francisco

General Hospital where she still works today. Lieberman, who was raised

in Paraguay and whose enthusiastic, refreshingly open style of speech

carries an Hispanic accent, saw in Fraiherg’s work an opportunity to

combine her attachment research and psychotherapy. In

interests in

recent years, she has taken Fraiberg’s model and applied it to a non-
abusing group of families whose babies had been assessed as anxiously
attached.
Lieberman reported the case of a mother who often provoked
In 1991
her eighteen-month-old by teasing him and taking away his favorite toys.
Typically, the baby threw a tantrum on such occasions, hut the mother

did nothing to alleviate his distress, explaining her apparent indifference

by saying he should he able to take care of himself. Lieberman was even-
tually able to ask the mother if that was the way it was for her as a child.

The mother gradually recalled that throughout her childhood she had

been “mercilessly teased and physically abused by her psychotic older

sister.” If she asked for help, her mother would say; “You girls work it

out yourselves.”
Lieberman spoke to her tenderly “about a young child’s needs to he

believed and protected when she asks for help.” At this the mother
began to cry profusely: “Nobody listened to me. I was so scared and I had
to take care of myself.” Lieberman asked, “Do you think that is why you
 They Are Leaning Out for Love 257

feel that Andy should get out of his tantrum all by himself?” The mother
seemed surprised and thoughtful. She said, “Maybe; 1 need to think
about it.” She then became considerably more tender toward her son.

During the next session another tantrum occurred. The mother, for the

first time, embraced him while he cried.

Bowlby could only have been pleased by such developments. For they
brought full circle his own clinical work in the London Child Guidance
Center of five decades earlier. But now, not only were parents being
treated along with the children, but attachment principles were inform-
ing the work and also being used to evaluate the results. In a study of one
hundred mother- infant pairs who had been assessed in the Strange
Situation at twelve months, Lieberman found that after one year of this
type of treatment the anxious pairs were virtually indistinguishable from
those who had been rated secure.
Reading all this, some parents may understandably wonder. Should 1

get a Strange Situation done on me and my kid? As a general rule, most

attachment workers would, 1 think, recommend against it, preferring

instead that parents simply make an effort to be more sensitive to attach-

ment issues or speak to a knowledgable therapist if they are worried.
Knowing a child’s quality of attachment may relieve some parents and
spur others to take action, but the relief may yield an ill-founded compla-
cency (even securely attached parents and children have problems),
while the action may be compromised by onerous levels of parental guilt.

Besides, the Strange Situation was devised as a research tool, and its

power is based on percentages. Some infants who receive sensitive care

look anxiously attached (they may have had a bad day), and some who
have neglectful parents look secure (they may have some irrepressible
positiveness that is not yet understood). Sroufe has been asked by the
courts to help settle custody cases by putting the child through a Strange
Situation with each parent, but he has steadfastly refused.
 Ugly Needs, Ugly Me:
Anxious Attachment and Shame

In the studies coming out of Minnesota and elsewhere, efforts were

repeatedly made to measure the selLworth of securely and
anxiously
unmistakable,
attached children at various ages, and the differences were
with secure children scoring higher in every measure of
selhesteemd

Attachment theory held that getting love, reliably and consistently,

makes the child feel worthy of love; and his perception that he can
attain what he needs from those around him yields the
sense that he is

an effective person who can have an impact on his world. Sroufe’s

find'

ings seemed to confirm this.

The child whose needs are not met, who feels ineffective in his efforts,

or who, worse, is rejected or put down in various ways, according to

Bowlhy, builds up a negative set of assumptions about himself. He is not

worthy of love or respect. He is, in effect, ashamed of what he is. In

attachment writings, the word “shame” is rarely if ever used; indeed,

attachment studies and shame studies have been largely unrelated fields
of inquiry. But there can he little doubt that shameful feelings about the

an important component of relational insecurity.

self are

Shame rises up from several sources. The anxious child must inevit'
ably feel that there is something wrong with him and something wrong
 Ugly Needs, Ugly Me 239

with the immense love that flows out of him toward his parents and is

somehow not accepted.^ The very fact of not being attuned to, so central

feature of anxious attachment, is in itself a shamednducing experience,

and some studies have found what looks like an early expression of

shame in infants whose mother does not appreciate, echo, or elaborate

the feelings they express. (For more on the subtleties of attunement, see

Chapter 23.) The anxiously attached child may also feel, at some level,

misshapen by his unwanted feelings. He is mistrustful, he is bitter, he is

retaliatory, he is violent, and, perhaps, worst of all, he is very, very bad

for harboring a secret hatred of his mother, a hatred for which in all like'

lihood he has been made to feel guilty and monstrous whenever he has
dared to show it. Greed, natural in infancy and, in most children, modi'
fied through gratification, may become a perpetual disgrace or a dirty
secret for him. Rage that he has not been able to satisfyingly express

turns inward, such that he, rather than the parent, becomes the villain.

For any or all of these reasons, he may come to hate himself and to want
to keep the secret of his defectiveness from the world.
Laura, a sixteen'year'old treated for depression and overeating by
Juliet Hopkins at the Tavistock Clinic, had, like Hopkins’ other young
patient, Clare, a mother who had shunned bodily contact with her diir'

ing her infancy. If Clare saw herself as a leper, Laura came to see herself
as a tortoise and her little sister (who had been caressed as a child) as a

cuddly bunny. Laura had dreams in which she was alone on a desert cov'
ered with a revolting skin disease. She had a concern for India’s Un'
touchables and had set up a collection box for them. “I have found it

common,” Hopkins observes, “for physically rejected children to drama'

tize or draw themselves as physically repellent or unstrokeable creatures,
like tortoises, toads, crocodiles and hedgehogs.”^ This sense of deformity,

degradation, or worthlessness is a central feature of shame.

Shame theorists believe that neglect and rejection, especially if early

and extensive, generate a selLfeeling in the child of ugliness and undesit'

ability. Parental rejection can be global, it can be limited to specific

aspects of the child that the parent dislikes, or it can he a subtle combi'
nation of the two. Is the child accepted and valued for what he is — slow,

placid, squirmy, delicate, dark'skinned, plain? Is he allowed to be dirty,

funny, needy, silly, proud, aggressive, weak, defiant, creative, uncertain?

To the extent that such qualities arouse anxiety in the parent and are
responded to with coldness, punishment, or ridicule, they will become
sources of shame.
240 The Fate of Early Attachments

A defensive parent’s own limitations can also translate into shame for
the child. A mother who selhcentered and ungiving may ward off her
is

own feelings of guilt by finding fault in her daughter. She perceives her
little girl as overly demanding and repeatedly scolds her for being selfish.
and despises herself for it.
The child comes to believe that she is selfish

Her natural self-assertion is compromised as she comes to feel that she

should not take, should not ask, should not calculate in her own
behalf,

for any of these things may exhibit the hated quality.

Whats more, she
finds that if she is restrained and solicitous, her mother likes and

approves of her. Nevertheless, her unmet needs keep rising to the surface
her
and she acts them out in ways that cause renewed displeasure in
mother and renewed self-hate in herself.

As she gets older, the girl compensates for her supposed defect with

rigid displays of generosity. She remembers everyone’s birthday, shes

always ready with a compliment, she seems content to settle for second
best. No one must ever know what she truly is; no one must
ever see that

clawlike third hand reaching out of her pocket with “Selfish!” written all

over it.

Because this inner dynamic proceeds largely outside of awareness, the

shame image often persists into adulthood in a strangely unevolved form.
If not understood or worked through, it retains the terrible charge of

parental rejection. The girl becomes a young woman who unconsciously

believes not just that she suffers from a troubling flaw, but that she is
revolting and untouchable and that her selfishness is a deformity that
makes her unfit to live among other human beings.

People differ in the degree to which they defend against shame. Some
obsessively avoid it by restricting their lives and narrowing their con-

sciousness. Through an addictive or compulsively busy life-style, un-

wanted self-images can be kept from impinging upon awareness. Others

are more aware of their shame and tormented by it, sometimes to
the

point of depression. Perhaps the best evidence that these two styles of
living with shame are associated with avoidant and ambivalent attach-
ment comes from a study of six-year-olds by jude Cassidy. Cassidy found

that securely attached children have a strong feeling of self-worth and

competence, but when pushed were able to acknowledge imperfections.
In other words, they seemed ta he neither tormented by shame
nor

rigidly defending against it. Avoidant children, in contrast, persistently

portrayed themselves as perfect and refused to admit to any shortcoming;

while the low self-worth of ambivalent children was prominent and
 Ugly Needs, Ugly Me 241

undisguised.^ This study suggests that from an early age quality of attach-
ment may be connected not only with the degree of shame formation but
with the development of fundamental dysfunctional personality styles.

Shame formation does not end with infancy but persists throughout
childhood. In any family the ways in which differences get worked out
are particularly fertile breeding grounds for shame, for they tell the child

a lot about his rights, his dignity, his worth. Is he allowed to feel he’s still

okay when saying no, when complaining, or when expressing the normal
hostility or aggression that children inevitably feel at times toward their
parents? Can the parent set limits or punish without spilling disgust or
rage all over the child? Chicago analyst Michael Basch, who has inte-
grated shame and infant studies into psychoanalytic thinking, has argued
that these highly charged moments are critical, for, as he puts it, “Shame
is the response to emotion that is not being dealt with effectively.”^
In a telling parallel, one that recalls Bowiby’s early thinking on the
subject, Roger Kobak has held that secure attachment is linked to the
child’s ability to successfully express negative emotions.^ A secure child
can cry, shout, or fall silent and know that he will be responded to in a
meaningful way. His negative emotions are thus an effective part of his
repertoire and do not become ugly to him and have to be disavowed. He
can say, “I hate you,” or “You don’t love me,” or “You’re mean,” or “You
like the baby better than me,” with confidence that he will not be met
with an icy stare, the silent treatment, a slap, an order to go to his room,
or various forms of guilt: “How can you say that?” “Don’t you know that

hurts mommy’s feelings?” “You ingrate!” “You nasty, selfish, rotten little

boy!” “You apologize!” None of this means that the parent must tolerate

destructive acts of aggression on the part of the child, like biting or hit-

ting. But rather that such behaviors can be curtailed without stigmatizing
the feelings behind them and ultimately the child himself.*
Evidence suggests that mothers of secure children seem much less

threatened by the child’s negativity. Their early sensitivity to the infant’s

signals reveals itself in later years in their ability to spot negative emo-
tions even when they are unspoken and to help the child to give them
voice. Robert Marvin found that in Strange Situation observations done
with three- and four-year-olds, mothers of secure children were much

*A parallel aspect of secure attachment is no doubt

related to the giving and receiving ot authentic
positive emotions such as love, joy, appreciation, and enthusiasm. Many parents, ot course, have
trouble participating with the child in these dimensions, too. See Chapter 23.
242 The Fate of Early Attachments

of their display of concern upon returning, to ask

more likely, as part

about the child’s anger at her for leaving.® The mother of the insecure
child does not behave this way, and he, in turn, may lack the confidence

that he speaks of his hurt or anger she will be receptive. Karin

if

Grossmann similarly found that mothers of avoidantly attached infants

tend to ignore the negative feelings their children express during
play,

only giving them friendly attention when they are in a positive

mood.^
with
Parents of anxiously attached children seem to have difficulty
at the Barnard
their own negative emotions as well. In a small study

College Toddler Center in New York, Lawrence Aber and Arietta Slade
found that

mothers of secure children were far more likely to express

anger directly to their children; furthermore, they reported

that when they became angry they would say so: “Mommy
gets angry when you do that,’’ “I’m angry at you, and so on.
Mothers of anxious (predominantly avoidant) children rarely

reported that they got directly angry at their children nor did
they report that they were likely to describe these feelings
to the children when negotiating conflicts Interestingly,

these were also the mothers who experienced their children as

most aggressive and who reported that their children would

bite and hit them.^°

Unleashed anger, expressed with venom, is abusive to a child and will

leave him feeling like a worthless thing. But unexpressed anger is damage

ing, too, for it leaks out in insidious ways — a look, a tone, an abruptness
of manner — which the child experiences as rejection. Sometimes the
subtler it is the worse it is, for both the sense of rejection and the shame
get planted at a deeper, less accessible level. The child does not burst

into tears or seek condolence from the other parent. There is no memory
of parental meanness, which might be available for the child to work
through — in play, in ruminating about fairy tales with witches and

demons, or even consciously at a later date. There is no concept like,

Mommy doesn’t want me to be this way, no opportunity to keep the

unwanted aspect of the self alive' through some form of rebellion. There
is only the subterranean certainty that he is no good and the relentless

supression of some dynamic aspect of the personality that might other^

wise reveal one’s deformity to the world.

 —
Ugly Needs ,
Ugly Me 243

Findings about parental attitudes toward negative emotion give us not

only a better understanding of shame formation but are suggestive of
some of the critical factors that may be at work in the continuation

and creation — of secure attachment in later childhood. To be under^

stood instead of punished, to express anger and not be rejected, to com^
plain and be taken seriously, to be frightened and not have one’s fear
trivialized, to be depressed or unhappy and feel taken care of, to express a
self'doubt and feel listened to and not judged — such experiences may be
for later childhood what sensitive responsiveness to the baby’s cries and
other distress signals are for infancy.

The first emotions that the parent deals with are the attachment emO'
tions themselves — the baby’s desire for connection, his need to be taken

care of, his need to be responded to when hungry, when in pain, when
wanting love or attention. Many parents, because their own dependency
needs were rebuffed as children, still live in unresolved pain over them.
But the pain is not felt, it is dissociated, and they structure their lives to

keep it that way. An avoidant boy who is ashamed of being needy may
grow up to be a man who is a caricature of independence, unable to ask

for help or closeness or even to feel those longings within himself with'
out risking the disintegration of his self-respect. Will he be warmly
receptive to moments of clinginess or dependency in his own little boy?
Clinical evidence suggests that parents cannot tolerate seeing their
unmet needs expressed by their children, and they cannot tolerate the

anger and distress the child expresses when those needs go unmet again.

They either overreact or become dismissive, with the result that the

child’s attachment feelings — as well as his anger and distress — are either

walled off from his consciousness or revved up to the point where they
overwhelm him. His ability to communicate his attachment-related
feelings is gradually shrunken and distorted until it demands misinterpre-
tation. This fairly well describes the avoidant condition, further compli-
cated by shame.
Two incidents from my own childhood may help illustrate this point. 1

grew up in a large city housing project with hundreds of kids. One winter

afternoon, when I was seven or eight, the ground was covered with snow
and all the boys were out in force playing and fighting. Gradually, as it

got darker, one kid after another went home. When none of my own
friends were around anymore, 1 wanted to go home myself, but my
mother was still out and the apartment locked, so I kept hanging out
244 The Fate of Early Attachments

I was the last kid. 1

with the others. Finally, it was completely dark, and
felt lonely and bleak. I thought of going
to visit my best friend’s family. 1
the happy
went to their building and up to their door. 1 could hear
sounds of family life coming from inside. But I could not bring myself to
boy, so
could not present myself to them this way, a needy
little
knock. 1

pathetic and alone.

The terrible thing for me in all this was not that my mother stayed out

and that was stranded. That was painful, but it only happened
late 1

was not a pattern. But the experience caused my

attachment
once; it

to surge dramatically, and, because of the way

attachment issues
needs
were dealt with in my home (and in my parents’ homes when they were
where autonomy and “strength” were emphasized and signs of
children),
unde-
dependency either ignored or found revolting, I felt pathetic and
sirable when my need for comfort, security, and care surfaced.

The second incident occurred a few years later was perhaps eleven — I

QQYv —
when my parents were going to dinner at my uncle and aunts and
1 was being left home alone. had no interest in going but for
1
some rea-

son, which 1 could not understand at the time, whined and complained
1

continuously and demanded to be taken. Finally, a call was placed and 1

was brought along. In retrospect, 1 think this was another moment

when
my attachment needs were surging, perhaps because of something that
had happened in school earlier that day, and as a result 1 did not want
to

be left had no experience negotiating such feelings, either

alone. But 1

with myself or with my parents. Another child might have

conveyed
talk about what
that he was down and needed comforting. If he could
might
was bothering him, be told that he was loved, be kissed or held, he
not have felt so alone when his parents left. But 1 could not ask for these

things and may have had difficulty tolerating them if they had been

offered. Instead, I ended up with a victory I didn’t want and a lousy feel-

ing about myself —

whiner and complainer, a big baby, needy and
a

greedy, who upsets the plans of others by making a fuss

over what he

doesn’t even want and is in no better mood after he gets it.

1 think repeated incidents of this sort give one a strong, if disavowed,
impression of who and what one is. Most of us repress a shameful self-

image because it is so painful, often building up a contrary, conscious

self-image, which we broadcast -to the world through various false dis-

playsand which we devote considerable psychic energy to maintaining.

But the negative self-feeling remains, and the fact that it is unconscious
 Ugly Needs ,
Ugly Me 245

gives it tremendous lasting power, if for no other reason than it can never
be examined anew, can never be felt, shared, or worked through.
The defenses that the avoidant child erects — as well as the manipula-
tive and devious ways he develops for getting his needs met — and the
anxious displays of dependency to which the ambivalent child is prone
are both causes for additional layers of shame, as first family members,
then neighbors, and then teachers and schoolmates all begin to react to
him as if he were annoying, hateful, inadequate, or odd. Aggressive, defi-

ant children do not like themselves for being that way. At some level of

consciousness they view themselves as angry, spiteful, unlovable beings.

They don’t know that they are also sad and desperately hungry for love.
Necessity has forced them to disavow such feelings and often to repress
the memories that go with them. They thus have little basis for sympa-
thizing with themselves. Meanwhile, the seeds of self-disdain are nour-

ished by the punishments, harsh words, and overt exasperation of others.

Clingy, panicky children, who are pitied or ridiculed, can only devalue
themselves as well. Their fundamental experience is rejection. The in-

evitable accompaniment is an ever-present sense of inferiority. If they

are also victimized, as some ambivalent children are, their self-worth,

already compromised, suffers further depreciation. For the ambivalent

child who is enmeshed with his mother, a different sort of shame may
strike later, in young adulthood, when he finds himself lacking in self-

reliance, overly dependent on his mother, and plagued by insecurity in

his relationships.

But what about the child whose degree of early neglect or rejection is

more extreme? Or the child who has been subjected to ridicule or severe

punishment whenever he complains, protests, or acts angry? Or the child

who is violently rebuffed when he exhibits the annoying and manipula-
tive styles of expressing need that follow in the wake of early rejection?

For him the feelings of shame that attach to the identity are proportion-

ally worse: I get treated like shit because I am shit; I come from hateful

people and I am like them.

This tightening grip of shame is apparent to anyone who has worked
with such children in therapy. 1 recall an adolescent boy I’ll call Juan
whom 1 treated some years ago. His father had left the home when Juan
was very young and showed little interest in him thereafter, a blow of
inexpressible enormity. His mother was controlling and fault-finding and
seemed to favor his younger brother. In his behavior — isolated, sheepish.
246 The Fate of Early Axtachments

occasionally bullying (he enjoyed being called

the Gorilla and liked to
inability to ap-
knock down other skaters at the rink); his complete
especially when in need (he sat alone in the
lunchroom
proach anyone,
being transferred to a new school and failed to make a sin^
for weeks after

gle friend in the four months he was there) — Juan was in every respect

attachment history. Al-

an example of a boy with an extreme avoidant
under a layer of
though Juan hid his shameful sense of defectiveness
grandiosity, his low self-worth was apparent in
the way he hung his head
always be rejected,
as he lumbered about, in his conviction that he would
in the fact that he pinned all his hopes for
love on an unlikely scenario
of life. On one occasion 1
of stardom but otherwise expected very little
his first thought was
asked Juan to draw a picture of an animal. Although
garbage-eating
a tiger, he settled on a pigeon, which he described as a
as he did so, of the hor-
bird, a flying rat. He then told a story, snickering
boys did to this pigeon, ending with its being killed
and
rible things that

tossed off the roof. It is not difficult to imagine the various ways in which

such a model of the self will get played out in later years.

The late psychologist Helen Block Lewis, one of the pioneer explica-
tors of shame, saw in the Bowlby-Ainsworth theory a means of ground-

ing the shame concept in evolutionary design. The ability to vicariously

experience the feelings of another, she said, “is the foundation of attach-

ment (on both sides), and the price we pay for it.’’ Shame is the vicarious
(and losses) are
experience of the rejection. To the infant, separations
experienced as rejections and therefore as blows to the self.
We would
not be human if this weren’t so.

Lewis believed that shame was always accompanied by

what she called
resistance of
“humiliated fury” and that this helped account for the angry
the ambivalent child in the Strange Situation.
The avoidant child, on
a pattern of reaction
the other hand, “may he showing the forerunner of
that involves bypassing the shame of being rejected. They do not directly

express humiliated fury. Rather, they behave as if they were turning the

tables on the rejecting mother by rejecting her.”^^

research in the
All told, the accumulating evidence from attachment
1970s and 1980s and the theoretical work it stimulated
was yielding a
lasting, core
powerful sense in the field that quality of attachment is a
issue for the personality. It seemed to encapsulate
something vital about
turn, related directly
the nature of one’s earliest relationships, which, in
to one’s feelings about oneself and one’s
expectations of others. When
 —
Ugly Needs ,
Ugly Me 247

combined with the knowledge gained from decades of clinical experi'

ence, and with the work being done by Main and her students on inter'

nal representations, it seemed plain that attachment quality must affect

how one approaches human connections at later stages of life and in

periods of crisis.

Clearly, quality of attachment was not everything. It did not govern

the development of one’s intellect or ability to play, which are affected
more by other aspects of the parent'child relationship. And it was cer-
tainly only a fraction of one’s psychological life, which rested on numet'
ous other factors in one’s makeup and development. In the emotional
complexity of any single person, quality of attachment, assessed at age

one and only to the mother, left out much of the complexity of any indi'
vidual life, and it left out much of what arises later in childhood that
makes a lasting impact on one’s psychology — negotiating a growing sep'
arateness from mother in late toddlerhood, coping with the sometimes
tormenting or traumatic conflicts of the oedipal period, making and loS'

ing friends, finding oneself rudely replaced as the baby of the family by a
younger sibling who co-opts everything that we thought rightfully ours,

and so on through all the challenges of adolescence. Attachment cate'

gory said nothing about the child’s inherent resiliency, his luck in finding
other, alternative attachment figures — or his determination to do so
or his ability to hold on to and nurture a kernel of positive experience
with a teacher or relative. It said nothing about the choices he might
make in adolescence or adulthood — in marriage, in career, in habits or

pastimes — or the misfortunes and bad influences that may befall him.
Nevertheless, to a growing number of developmentalists, the quality of
early attachment to the primary caregiver stood out, like the key in
which an otherwise complicated piece of music is played, imbuing the
personality of many children with a characteristic inflection that is prC'

sent from movement to movement.

 A New Generation oe Critics:
The Findings Contested

To many young developmental psychologists, reading the work of Ains^

worth, Sroufe, Main, and their students and to others who
heard about

attachment principles through popular authors and popular books

b\

experts like Selma Fraiberg, Penelope Leach, or T Berry Brazelton, one

of the beauties of the material has been its great common sense. It seems
their biologic
only natural that infants would seek attachments as part of
makeupj that our earliest relationships become a part of us, that
cal
an internal working model would account for the types of
something like

relationships we develop later in life.

intuitively pleasing, that’s what’s getting in the way,” says

the
“It’s

internationally renowned developmentalist Jerome Kagan. Because it

makes intuitive sense people are assuming it’s right. But most of the time

intuition is wrong. I mean, intuitively the sun goes around the earth,

right? Intuitively, the earth is flat, right? Why is psychology the least

advanced science? Because our intuitions aren’t very good.”^

Kagan, a hugely influential Harvard psychologist who eschews ideO'
logical labels (“I’m part of the reasonable school”), is the
author of The

Nature of the Child, which casts a critical eye on such popular assump^
 A New Generation of Critics 249

tions as “a mother’s love for her infant is necessary for the child’s future
mental health” or “that the events of infancy seriously influence the
future mood and behavior of the adolescent.”^ At sixty-four Kagan is

ebullient, enthusiastic, and likable. He is combative in his views, enjoys

making others defend theirs, and seems keenly aware of the impact he
has as a speaker. He occasionally sports an ironic halhsmile as he raises
his eyebrows, cocks his head, and scrutinizes an opponent’s arguments for

a weakness. Kagan is also prolific, and his work has been widely hailed.

His accessible writing style, his breadth of knowledge, his ability to inte^
grate insights from anthropology, sociology, and philosophy, not to men-
tion developmental psychology, into a grand and persuasive theme have
made him a formidable and at times maddening antagonist to the attach'
ment community.
By the early 1980s Kagan had emerged as the spokesman for the grow'
ing body of attachment critics, rejecting almost every assumption upon
which attachment theory was based. He does not believe that early eX'

perience is particularly important. (“I was rejected by my parents” has

become, he charges, “a rationalization for distress and incompetence.”^)
He spurns the idea that early experience forms any lasting psychological
structures, such as Bowlby’s internal working model. And even if early

experience did matter, Kagan doesn’t believe that the Strange Situation
measures it. (“Is it reasonable that a history of interaction between
mother and infant comprising over a halBrnillion minutes in the home
would be revealed in six minutes in an unfamiliar room?”"^) He sees
secure and insecure attachment, at least as those terms are currently
used, as misnomers and believes that what’s actually being measured are

aspects of the child’s behavior that reflect either its inborn temperament
or its assimilation of parental values. Finally, he believes that attachment
theorists are too concerned with what they call security and have
allowed their own values to skew their work. (“In the forties and fifties,
the children now called attached were called overprotected and that was
a bad thing.”^) His position on attachment is complicated because he
attacks it from several different directions, which are not always consiS'
tent with one another, but all of which he argues with great verve and
authority.

Kagan believes, for instance, that the modern concern for attachment
and for the general impact that parents have on their children is an
ephemeral aspect of this time and this place and should hardly be the
basis for a general psychology of development. Because of the uncertain'
250 The Fate of Early Attachments

ties inherent in modern society, Kagan argued in 1978, we have latched

in order to regain
on to the idea of the importance of early experience
some sense of control.^ In the process, motherdove has been elevated
to

something of an icon:

Every society needs some transcendental theme to which

citizens can be loyal. In the past, God, the beauty and utility

of knowledge, and the sanctity of faithful romantic love were

among the most sacred themes in our society. Unfortunate'
ly the facts of modern life have made it difficult for many
Americans to remain loyal to these ideas. The sacredness of

the parent'infant bond may be one of the last unsullied

beliefs.^

By taking the debate to a higher plane, where the cherished beliefs of

our culture may be seen as fleeting, Kagan was able to strike at the foun'

dation of attachment thinking. He objected to Bowlby’s claim that the

“loss of a loved person is one of the most intensely painful experiences

any human being can suffer.” He dismissed Bowlby’s belief that “intimate

attachments to other human beings are the hub around which a person’s
life revolves, not only when he is an infant or a toddler, hut throughout
his adolescence and his years of maturity as well, and on into old age.”^
According to Kagan, such is not the human condition but merely the

way it is for certain people, raised certain ways, in certain cultures. A

high'caste Hindu, he argued, who has trained himself to avoid attach'

ments of all kinds, would hardly feel this way.* His dismissal of Bowlby,

like many of his arguments, had an Olympian tone that must have

caused teeth to gnash from London to Berkeley:

Although many nineteenth'Century observers would have

understood and probably agreed with Bowlby, few would have
written three books on the theme of attachment because, like

the blue of the sky, the idea was obviously true. Bowlby’s coO'
elusions are newsworthy in the last half of the twentieth cen'

*This strikes me as somewhat facile. That the Hindu monk must struggle mightily to extricate him-
self from feelings of attachment suggests that attachment is indeed a basic and powerful
element of
human nature. In Christopher Isherwood’s Ramakrishna and His Disciples, a Hindu holy man speaks

of the unexpected personal torment even he feels over the loss of a loved one and voices renewed
compassion for the emotional pain experienced by ordinary folk when they suffer a loss.
 A New Generation of Critics 251

tury because historical events have led many citizens to ques'

tion the inevitability of maternal devotion to the child and of

the child’s love for the family. Parental abuse and adolescent
homicide of parents have undermined the nineteenth^century
faith in the naturalness of familial love. Modern citizens have
begun to question the universality of deep affection and con-
tinned loyalty, whether between adults or between parents
and children, are saddened by the conclusions implied in that

inquiry, and are eager to hear a wise commentator on human

nature assert that the love between child and parent is an
absolute requisite for psychological health.^

Kagan acknowledges that “the concept of attachment remains useful

and should not be abandoned, and yet it is plain from much else he
says that he thinks little of the concept and less yet of the work that’s

been done in its name.

As for quality of attachment, Kagan sees it simply as a matter of
contemporary mores: “My view is, if you’re attached, you are motivated

to adopt the values of your parents. If your parent values autonomy,

you’ll be autonomous; if your parent values dependency, you’ll be de-
pendent.”^ ^ He argues that the children whom Ainsworth has labeled
securely attached become upset when left alone in the Strange Situation
not because they’re securely attached but because they’re less able to deal

with uncertainty. They’ve been trained for dependency, and are showing
its ill effects.

Like Bowlby, Kagan is a great synthesizer, but despite his acumen, it

sometimes seems that Kagan has assimilated just enough attachment the'

ory to discredit it, rather than to wrestle fully with its meaning. His writ'
ing on attachment shows a number of errors and misunderstandings,

more typical of the fault finder and antagonist than the reasoned coni'
mentator, and that may have hindered his ability to be heard. His diS'

taste for attachment thinking is such that when 1 called him tor a first

interview in 1989, for a piece on attachment for The Atlantic Monthly, he

was incredulous that 1 on such a
would wish to write a popular article

theme. In 1982, according to Sroufe, he convinced the W. T. Grant

Foundation to stop funding attachment and similar research on early
development.^^ But regardless of cracks in some of his arguments and the
ire he arouses, his views cannot he ignored.
A researcher who studies inhorn temperament and who strongly
252 The Fate of Early Attachments

believes that genes contribute to much of what we become, Kagan argues

that too much attention has been wasted studying the effects of early
environments. He cites studies demonstrating that even maternally

deprived children, raised in institutions, could make surprising recoveries

when given a superior environment and adequate treatment. Research

by Stephen Suomi, Harlow’s successor at the University of Wisconsin,
seemed to support these findings, showing that severely deprived rhesus
monkeys who were frightened and unsocialized could become virtually
normal if placed in the company of a group of younger, “therapist” mon^
keys, where they could gradually learn social skills. Such stories of
rehabilitation caused many in the field to claim that Bowlbyism had
been cut off at the legs, for they assumed that attachment theory was
based on the assumption that early inadequate care always caused irre^

versible damage.
This misconception, which dogged Bowlby since he first wrote about
maternal deprivation, was largely, 1 think, the responsibility of attach'

ment workers themselves, and especially the ardor with which they
reported the predictive power of Ainsworth’s classifications. To predict,

of course, does not mean to determine. It does not even denote a causal
relationship. (The rising sun fairly consistently predicts an increase in
automobile traffic on certain parts of the planet.) But this was more and
more the implication.
As Bowlby saw it, anxious attachment in the first year did not neceS'

sarily breed future difficulties, but it was a liability for the child, espC'
dally since the future is in some way built upon the foundation laid in

the past. The fact that the parental psychology that may have con'
tributed to anxious attachment in infancy and the parental behaviors
that went with it usually remained in some form throughout childhood
increased the liability. That parents often had negative reactions to the

behaviors of their anxiously attached children, and that parent and child
became involved in an escalating pattern of negative interaction as a

result, made the risks of disturbed emotional and social development that
much more severe. Finally, the fact that the child built a model of relat'

ing based on such experience and tended to become committed to that

model, perpetuating it at home an.d replicating it with others, made

change that much more difficult. Yet things do change, and the com'
plexities multiply to such a degree that no hard predictions can be made
about the future.
 A New Generation of Critics 253

The Minnesota studies had shown that after the first year, mothers of
anxiously attached children continue to respond to them inadequately,

so that at subsequent stages of their development — when they need to

establish a healthy autonomy, manage their frustration in dealing with

challenging situations, know what to do when their feelings were hurt by
a playmate — they were still not getting the sort of parental care they
needed. What’s more, qualities in the children themselves were now
adding to the impediments, frustrating parents when they wanted to be
helpful. This was evident as early as one year, when avoidant children
would not make themselves available to maternal soothing (by acting as

if they didn’t need it) and when ambivalent children drove their mothers
nuts by alternating a momentary cuddle with arching away and kicking,

such that nothing the mother did seemed to alleviate their misery. Given
all this, it was logical to see anxious attachment in the first year as hav'
ing the potential to set the tone for much of the child’s emotional and
interpersonal life.

But the question of causality was a murky one, and in the enthusiasm
of attachment presentations — where one naturally wanted to give the

importance of one’s findings as bright a shine as possible — the implica'

tion was often left hanging that maternal rejection in the first year in

and of itself caused dependency or bullying in the sixth, and did so

regardless of what might happen in years two through five or in the

child’s other relationships.^'^ If Kagan and other critics made the first year

sound inconsequential, attachment theorists and their followers in the

press sometimes made it sound insurmountable.
Sroufe’s own early reports could have dispelled such notions, for

unlike children in middle^class samples who showed remarkably consis'

tent attachment styles not all of the children in his high-risk sample
maintained the same attachment classification between ages twelve and
eighteen months. Change occurred rarely, but when it did, it usually

reflected the upheavals, both negative and positive, to which these fami'
lies were prone, like the disappearance of a father or the single mother’s
forming a stable partnership with a new man. In another Minnesota
study, in which the mother’s behavior was assessed alongside the child’s,

the quality of the relationship clearly continued to play a critical part in

the quality of the child’s adaptation. A number of children who had been
securely attached as infants exhibited behavior problems at two and a
half. It turned out that their mothers no longer behave like the mothers
254 The Fate of Early Attachments

of secure children. They were neither warm nor encouraging, were ineh

fective teachers, failed to provide suitable playthings, and by the time

their children were four sometimes confessed to feeling confused and
bewildered.
Similarly, some children with anxious attachment histories did not
display the behavior problems that were typical of anxious children in
the poverty sample. The mothers of these children respected their striv'
ings toward independence, allowed them to explore without being intrm

sive, were warm and supportive teachers, and provided the children with
a healthy, stimulating environment. In most respects, they looked just

like the mothers of securely attached children. And perhaps that is what
they now were.^^
Sroufe not only believed that fundamental changes can take place
after the first year, he had to repeat this point again and again in an
effort to shake off the charge that he was a first^year determinist.

But as far as critics like Kagan were concerned, such protestations only
enhanced theirown arguments. For if attachment status can change as

care changes, then why make such a fuss over the first year? Why pursue
all this data showing correlations of year^one attachment classifications
with later emotional health? Kagan and others found the idea of an
internal working model, which acts like a flywheel, tending to keep a
child on his established course, at best unproven, at worst ludicrous.

They maintained that the disturbances caused by an inadequate or

rejecting environment lasted only as long as the child remained in that
environment.
It was not that Kagan believed a child was infinitely malleable. Quite
the contrary. But he argued that those characteristics of the child that
are most enduring are the ones he brought with him from birth.

Environmental influences come and go, but heredity lasts. In stark con^

trast to Bowlhy, who believed that present adaptation is always a weave

of developmental history and current circumstances and that early expe^
rience is especially formative, Kagan saw early experience with parents as

having much more transient effects. “A particular quality in a young

child,” he wrote in 1978, “is not likely to persist unless the environment
continues to support it.”*^ Or, as he put it on another occasion, experi-
ence was like material on a tape; if it could he recorded, it could he
erased.’^

If it were true that the effects of early experience only endured as long
as the environment remained unchanged, then except for the fact that
 ^

A New Generation of Critics 255

no one wanted to see little children suffer or feel insecure, there was less

urgency to one aspect of what Bowlby and Ainsworth had been fighting
for: loving, responsive parenting in the early years. It would be nice if it

were there, and everyone would like to have it so, but infancy is just one
ofmany phases the child will go through and his parents just two of the
many relationships he will have. This point of view had policy implica'
tions, too — for orphans, hospitalized children, children being fought
over in custody battles, and babies in poor-quality day care, to cite some
obvious examples. Certain conditions might cause insecure attachment
or even psychopathic behavior, but that is less cause for concern if the

disturbance will pass as the child grows older and has new experiences.

For example: Should a two-year-old, or even a four-year-old, who has

been living happily with foster parents that want to adopt him be re-

turned to his biological mother who hasn’t seen him since early infancy

but has now gotten her life together and wants her child back? Fieart-
rending cases like this come up in the courts repeatedly. Some, including
most attachment thinkers, argue that the interests of the child demand
that he stay with the foster parents, that it would be criminal to rip him

away from the people to whom he has formed his primary attachments.

In 1989 Ainsworth supported foster parents in just such a struggle, argu-

ing that it would be heartbreaking for the child and perhaps emotionally
damaging for him in the long run to be extracted from his family and
returned to parents he didn’t know. Others insist that children will adjust

to such painful changes without lasting damage.

Sroufe tried to counter the argument that the effects of early experi-
ence could be virtually erased by examining children whose level of
functioning had shown marked declines or improvements. Some chil-
dren in the poverty sample, for instance, had had secure attachment his-
tories but deteriorated in the preschool years to the point where they

could not be distinguished from children with anxious attachment histo-

ries. Sroufe found that these children were more likely to rebound later
than children who had been anxiously attached from the beginning.
He also found that children who had been anxiously attached as infants
hut who were functioning well in later years, such that they could barely
be distinguished from secure children, tended to suffer from underlying
problems with their self- esteem. “Even when children change rather
markedly,” he wrote, “the shadows of the earlier adaptation remain.”^
He believed that in times of stress those shadows were most likely to
appear — and he, too, was able to cite monkey data, indicating that
256 The Fate of Early Attachments

Suomi’s rehabilitated macaques do indeed relapse under some condi'

tionsS^
The experience of adults in psychotherapy, who have to struggle

mightily to overcome the relational styles, attitudes, expectations, and a

sense of self learned in the home; the proneness to depression in times of
loss or separation of those who have suffered the early loss of a parent;
such things would seem to attest to the enduring power of early models.
But that does not mean the shadows of the past always linger. What is

more, we still know remarkably little about the effect of other attach'
ment figures, most prominently the still largely overlooked father, where
the relationship may be good enough to overcome insecurity with the
mother, at least to the extent that other aspects of development to go
forward relatively unhindered. Nor do we know much about changes in
attachment status. We don’t know if there is a tendency for the number
of secure children to decline as parents are faced with more and more
complex challenges — toilet training, the terrible twos, the child’s groW'
ing assertion of autonomy, oedipal conflicts, relating to peers, struggles
with schoolwork, the complexity of adolescent emotions; or if many chih
dren who had been insecure as infants become securely attached later as
parents’ life situations change or as hidden strengths — of parents or
child — come into play at later ages. We do not know, partly because, as
Ainsworth has noted with some chagrin, no one has done for later peri'
ods of childhood the sort of naturalistic longitudinal study in the home
that she did for the first year.^^

But certainly evidence exists for change in both directions. Indeed, in

Sroufe’s sample, although impressive correlations between infant attach'
ment status and current functioning hold good for later ages — especially
regarding dependency (where the correlations to early attachment status
were as strong at fifteen as they were at four and a half^"^) and peer com'
petence — some of the correlations do decrease steadily as the children
get older, suggesting the persistent impact of developmental and envi'
ronmental changes. Critics may seem cavalier in dismissing the power of
early experience or suggesting the ease with which its effects can be
reversed. But whether something — in the form of an underlying vulner'
ability or strength — must always remain of the earlier attachment expe'
rience is still a matter of legitimate debate.

With the development of a method of assessing adult psychology viS'

3'visattachment (see Chapter 24) and the growing up of the first chih
dren put through the Strange Situation, it has become possible to address
 A New Generation of Critics 257

this question of continuity in a new way: by comparing attachment sta-

tus in infancy with attachment status in adulthood. Results from longitu-
dinal studies in Germany have been inconclusive, revealing major

reversals of attachment status and no correlations. But when Judith

Crowell and Everett Waters, who re-examined the grown children, age

20 to 22, from the first (middle-class) Minnesota sample, they found

something very different. Sixty-nine percent retained in adulthood the

Strange Situation classifications, secure or insecure, they had as infants
(ambivalent vs. avoidant was not assessed). When major life events that
might be expected to disrupt attachment status in childhood were
accounted for (such as the death of the mother when one child was
four), the correlation rose to 77 percent. All this is, of course, still pre-

liminary.
But if we put aside for the moment the various empirical uncertain-

ties, what stands out as important here? To me it is this: First, that there

do seem to be patterns of attachment, secure and insecure, and that they

matter a great deal, such that we would naturally want our babies to have
a secure attachment to us if we possibly could and that we would want
that security to persist. Second, that these patterns show up as early as

age one and that they grow out of the quality of the parent-child rela-
tionship. And, third, that they continue to be an issue for people later in
life Chapters 24-26), such that they can be seen as representing an
(see

aspect of our psychological structure. Thus, no matter what damage we

might do as parents, and we are bound to do some, perhaps much, know-
ing that our children feel confident of our love and availability at least
insures that they have a good foundation for self-esteem and non-
avoidant self-reliance and for building secure and loving relationships
with others. This in itself seems worth knowing, regardless of whether
the early pattern is destined to endure or how easy or difficult it is to

change. My own clinical experience and training, like that of many oth-

ers, also tell me that anxious attachment to the primary caregiver that

begins in the first year and persists thereafter is the worst case for any

child, especially if there are no mitigating attachments to others. For it is

our earliest wounds that are most deeply unconscious, that are almost
unknowable, unattached as they often are to memory or language, and
thereforethe hardest to question, to symbolize, to verbalize, or to
change. Change, of course, is possible, especially if there is a healing
source of love elsewhere or if the parent makes an early and radical shift.

There is no reason to believe the primary issues of early development are

 258 The Fate of Early Attachments

closed at the end of the first year. Whether they are understood in terms
of narcissism or infantile paranoia or secure attachment, all of early
childhood will be important in determining how they get worked out,
and trends begun in the first year can take a turn for the better or worse
in the next. (One of the grown children interviewed by Crowell and
Everett, who had been anxiously attached to his mother at age one,
seemed to have emerged into a secure adult because at age eight, when
he developed diabetes, his mother become for him the loving caretaker
she had not been before.
But 1 think we can also say that the more entrenched a position is in
the first year or two, the harder it will be to change. And many apparent
changes will represent the ability of some children to successfully move
on to other things — sports, school work, friendship — that veil their
wounds. It will only be later, perhaps in adolescence, when these chih
dren again long for closeness, that the early damage will be apparent.
While attachment critics have been justified in pointing out the fallacy
of interpreting followmp studies in such a way as to assume that every
pattern seen in older children flows ipso facto from quality of attachment
to the mother in the first year, I think they have been too sanguine in
dismissing its importance. The reports of child therapists, describing the
enormous struggles they go through to help young children deal with dis-
turbed relationships with their parents, are sobering in this regard: Yes,
change is possible, but it is often difficult, and we should value every
effort to get the parent-child relationship off to a good start.

Because of Ainsworth and her followers, we now have considerable

information about what yields secure attachment in the first year, and as
we shall see in subsequent chapters, that information keeps growing.
This, too, is important. It is valuable in helping parents, and in helping
society to help parents, to relate to their babies in a healthy way. That
parents and society should want to make that effort seems self-evident. It

does not depend on proving that the effects of the first year necessarily
last forever, and it seems unfortunate that the debate should get stuck
there for even a moment.

Like many other attachment critics, Kagan was particularly dubious

regarding the interpretations that were being made about the avoidant
child. He could not accept the idea that these children, some of whom
had little trouble separating from their mother in the Strange Situation,
many of whom settled down eventually once she was gone, and all of
 A New Generation of Critics 259

whom showed little interest in her upon reunion, were revealing some
kind of insecurity. Such kids were either bold, or precociously indepen-
dent, or had been trained by their parents to control their fear.

A child whose mother has been otherwise attentive and lov-

ing IKagan wrote] but has successfully encouraged self-

reliance and control of fear, is less likely to cry when the

mother leaves and, therefore, is less likely to approach her

when she returns. . . . Although some psychologists might
regard these latter mothers as less nurturant, the mothers may
have behaved as they did because they valued control of fear
and sturdy self-reliance in their children.

Kagan saw Bowlby, Ainsworth, and Sroufe as narrowly preoccupied

with security and not attentive enough of the advantages American soci-
ety confers on being able to handle adversity. Thus, a parent rated as

insensitive on Ainsworth’s scales might actually be giving a child supe-

rior training for the modern world, and all those children that attach-
ment researchers had been labeling securely attached may be poorly
adapted to real life.^®

The debate over the avoidant child made the findings of the German
attachment studies a source of much concern. These studies were the

work of a husband-and-wife team, Klaus and Karin Grossmann. A dis-

gruntled Ph.D. student in psychology at the University of Arkansas in
in
the late 1960s (he disliked behaviorist rigidities), Klaus’s interest
ethology led him to a second doctorate at the University ot Fraiburg,
and ultimately to Ains-
combining psychology with behavioral biology,

worth’s longitudinal studies. “1 thought this is exactly what ethologists

would do if they worked with human babies,” he says. He decided to

launch what would be the most comprehensive replication of her re-

search ever undertaken. Karin gave up her studies in mathematics
to

become a psychologist and join him. “We decided that if we did a study

like Ainsworth’s, she says, “we would have to see the children right at
”

children were
birth. Because otherwise how would we know that these

not different from the beginning? But at that time in Germany

it was

practically impossible for a strange male to enter the maternity ward. So

work with a hospital, and was present at all the births ot
we started to 1

the infants.
The first sample of children studied by Klaus and Karin Grossmann in
 .

260 The Fate of Early Attachments

Bielefeld in northern Germany seemed to confirm Kagan’s view. Fully

two thirds of these children were classified insecure and half of the entire
sample were classified avoidant — two and a half times the typical figure
for avoidant children in middle-class American samples. In conversation
with the Grossmanns, Kagan discovered that they attributed the large
number of avoidant classifications to the fact that in this part of
Germany early independence was highly prized. “Should we conclude
from these data,” Kagan asked rhetorically, “that many more German
than American children are ‘insecurely attached’ . .

The Grossmanns ultimately did a second study in Regensburg, in

southern Germany. Here the percentages of the children in each attach-
ment classification group were, much to their relief, similar to the
American breakdown. But the data from the southern study came later
and so were not available when the researchers attempted to understand
the high rate of avoidants in the north and what this might suggest about
German child-rearing practices and German culture in general.
At first the Grossmanns were inclined to take the view they reported
to Kagan and that he, in turn, wrote up in The Nature of the Child: that
an avoidant classification in this part of northern Germany, where peo-
ple tended to “distance themselves a bit from each other,” should not be
a matter of great consequence. The British ethologist Robert Hinde,
Bowlby’s “tutor in ethology” and an important attachment theorist him-
self, was inclined to agree. To be avoidantly attached here should not be
seen as a psychological liability but rather representative of an appropri-
ate adaptation to this culture. So when the Grossmanns published their
data in 1985, aware of the prejudices against the German character as
being cold and authoritarian, and hesitant to be in the position of pro-
viding ammunition for those who might see the huge number of
avoidant children as evidence for this view, they were cautious and pro-
tective in their conclusions: “We simply said that it may be caused by the
fact that these mothers wanted to have their children self-reliant as early
as possible.

The wish to train their babies for early independence would explain
why the Bielefeld mothers looked so different from many of their
American counterparts. They were quite sensitive to their children when
they were newborns, only becoming less responsive in the latter part of
the first year. Indeed, according to Klaus Grossrnann, “Most of these fam-
ilies that had avoidant children are actually very nice families, and they
are very nice to their children in other ways.”^^ On Ainsworth’s mater-
 .

A New Generation of Critics 261

nal acceptance scale, for example, the mothers of these avoidant chil-
dren were average to low average, considerably higher than their
American counterparts, where low sensitivity was always accompanied
by very low acceptance.
“You can still think your child is a nice child,” Karin Grossmann
explains, “and that you’re going to cherish him and support him, but you
just don’t want him clinging to you. We have studies that show that
these mothers are quite sensitive when the child feels good; what they

can’t stand is a crying child coming to you, clinging to you, and wanting
contact. So they push that out.”^^ She notes that it’s quite possible to

“imagine a whole culture functioning on the avoidant principle: You

mind your own business, you don’t show your emotions, you don t go to
anyone for help. In old Prussia I think it would have been a high ideal to

keep a stiff upper lip and never bother anybody else with your prob-

lems.
Nevertheless, the Grossmanns came to believe that the avoidant

attachment could not be so easily discounted as culturally normative. To

have to achieve the required level of physical independence at ten

months — to abruptly cease displays of love, need, emotional hunger, and

distress — is something that the human baby adjusts to at a cost.
The latter part of the first year is a time when many observers, includ-

ing Ainsworth, have noted the onset of separation anxiety. It is a period

when the baby’s perception of his parents as unique individuals is fully

coming into being and his love affair with them is reaching a peak. He
follows like a puppy, wants to be held a great deal, climbs onto his
mother with glee* In a book addressed to mothers, Penelope Leach writes
of the extreme emotions that frequently prevail at this stage:

The baby has learned to know and love you better than any-

one else and now he wants you all the time. . . . His ideal

would be your continual presence and constant attention. He

feels passionately for you physically. He will sit on you, play

with you, stroke and pull you, pop food (and w'orse) in your
mouth, behaving as if your body belonged to him. [He . .
.

demands] to be cuddled and kissed, patted and stroked. He

holds out his arms for more, laughing with glee when he is

tickled, sucking his mother’s nose if it comes within range and

purring like a sensuous kitten when it is time for a bath or a

new diaper. .
 262 The Fate of Early Attachments

Leach warns that even a mother who enjoys and reciprocates this
intense love affair may find the baby’s extravagant demands and inces'
sant shadowing difficult to take at times. According to Bowlby, of course
the baby behaves this way because all of his instinctual attachment
behaviors have by now been unified into one coherent system, and they
are directed toward specific favorite persons, with the primary caregiver

being the apple of his eye. And because of his helplessness, and his
inability to soothe himself with images of the absent mother or to delay
gratification, there is an urgency to his love that is beyond anything we
can fathom in later life. It is just at this moment that many parents in
northern Germany feel it necessary to pull back and shut the door on
this all-consuming baby-love.*
It is interesting in this respect to note what the mothers of the securely
attached children in the Bielefeld sample looked To the Gross- like.

manns, they seemed unlike the average German mother. “They were not
so eager to present themselves as a perfect household,’’ Karin says, “clean
child, clean house, clean mother. The mothers were less dressed up, the
children were less dressed up, they weren’t overeager to serve coffee or
tea right at the beginning. I actually had the feeling that they were more
secure with themselves. There was no indication that these mothers
or their children were poorly adapted to this society. Indeed, quite the
opposite.
To the “shock and disbelief of their German colleagues, follow-up
studies demonstrated the same sorts of continuities in Bielefeld and
Regensburg that Sroufe was finding in Minneapolis, the secure German
children demonstrating numerous psychological advantages. At five the
Regensburg children, for example, played with greater concentration and
quality, theywere better able to manage conflicts with their playmates,
they had a better understanding of social situations. What’s more, they
felt free to go to their parents when they were lonely, sad, or distressed.
The avoidant children, in contrast, were left alone with their negative
emotions.

Most American parents can, think, understand the impulse behind the stop sign erected hy the
1

Bielefeld mothers. Our culture has mixed feelings about closeness and dependency, and besides,
we
are busy people. On the cultural spectrum we are rnuch closer to the citizens of Bielefeld than to
many traditional or primitive societies where babies and mothers are inseparable for years and where
physical and emotional intimacy is the norm. Every culture requires that its children adapt in differ-
ent ways in order to live successfully in that society, and the human being is malleable
enough to
adapt well to an extraordinary variety of conditions. But that does not mean that every
adaptation is
healthy or useful or that culture cannot he organized in harmful ways.
 A New Generation of Critics 263

Unlike Sroufe’s sample, however, there were few behavior problems

among the anxious German children. They were essentially normal kids
from normal families, with none of the stresses of poverty, single parent'

hood, or extreme neglect. Also some of the avoidant children in the

Bielefeld sample may have been different even from avoidant children

in American middle^class samples, a reflection of different parental mO'

tives. Many of the parents of the avoidant Bielefeld children supported
them in school and in sports and were generally accepting of them.
Certain feelings could not be displayed or discussed, but unlike many of

the avoidant American families, there may not have been the same

buildup of rejecting experiences. The avoidant children in the U.S. had

mothers who rebuffed them despite a general cultural attitude that is

accepting of warm responsive care to children under one. It implied a

more deep-seated psychological aversion to having a warm connection

with their child. Because the mothers of avoidant children in north
Germany were behaving according to a cultural norm, their psychologi-
cal qualities may in some cases have been less of an issue and the rejec-

tion less pervasive as a result.

This, of course, raises a question about the Minnesota studies. Do
their follow-ups demonstrate the expectable effects of anxious attach-

ment? Or are they more properly the effects of severe anxious attach-

ment? Clearly, after Bielefeld, it could now be said that if a child qualifies

as anxiously attached in the Strange Situation at year one, this can imply
a large variety of outcomes, some far less grave that what Sroufe’s follow-

ups might suggest. (More on this in Chapter 25.) That many ot the

German parents raised avoidant children because of the dictates of cul-

ture rather than psychology might also help account for the fact that the

Grossmanns found so many more reversals of attachment quality when

assessing these children as adults.
As the Grossmanns continued to follow the Bielefeld sample, they
nevertheless became convinced that these children paid a lasting price

for their avoidance. At age ten, unlike their secure counterparts, the
avoidant Bielefeld children did not form close friendships. They reported
many more problems with peers, like being excluded, taken advantage of,
or ridiculed. They had less confidence, less self-reliance, less resiliency.

Although the avoidant pattern was less pronounced than among the
Minnesota children, “When put under stress, the patterns ot withdrawal

and reduced peer interactions so vividly described by Alan Sroufe,” Klaus

Grossmann recently said, “can also be readily observed in our chil-
264 The Fate of Early Attachments

dren.”^^The Grossmanns, therefore, have come to agree with a point

Bowlby made when accepting an honorary doctorate at the University of
Regensburg in 1989 — that the mere fact that parents are behaving in
accordance with a cultural norm does not necessarily spare the children
any harm. (Bowlby, however, was strikingly out of touch, I think, when
he stated, “No matter where the rejection comes from, it is bound to
have the same outcome.
As a result of such findings, the Grossmanns are urging changes in cer^
tain strongly held German opinions about child rearing. “We fight the
notion,” Klaus says, “that crying doesn’t matter, that it strengthens the
lungs. We fight the notion that independence has to be trained, or that
you must punish a child by withdrawal of availability or love. We’ve
explained over and over again that those children who get the most sen-
sitive responsiveness on the part of the mothers are the ones who will be
least clinging.

In 1975 Ainsworth, then sixty-one, left Johns Hopkins for the Univer-
sity of Virginia at Charlottesville. She had been dissatisfied at Hopkins
(and made nervous by their official retirement age of sixty and the need
to get yearly waivers). As someone who readily turned students into
members of an extended family, Ainsworth found Virginia’s appeal
enhanced by the fact that Robert Marvin was already teaching there and
that another former student, Mark Greenberg, was studying with him.
Leaving Hopkins when she did meant that Ainsworth would no
longer be able to work with a bright South African student named
Michael Lamb (who would soon go on to do the early research on at-
tachment to father). Lamb had been eager to study with Ainsworth, but
was only able to complete one year at Hopkins before she left. Lamb
immediately moved to Yale, and his relative lack of experience with
Ainsworth, including never getting Strange Situation training from her,
would become a matter of heated concern ten years later when his name
appeared as the lead author of the most global critique of Ainsworth’s
Baltimore research that had ever been made —
and the first to stir bad
blood. Lamb, who subsequently became chief of the Section on Social
and Emotional Development at the National Institute of Child Health
and HumanDevelopment, accurately noted that the consequences of
anxious attachment were much better documented than the causes. He
questioned not only Ainsworth’s sample but her methods and numbers,
and questioned whether the Strange Situation is complex enough to
 A New Generation of Critics 265

“capture all significant variance in infant behavior.” Lamb was hardly

alone in having doubts about the Strange Situation. Psychologists of

many different stripes and persuasions had voiced strong doubts about it,

even ridiculed it. The English developmentalist Michael Rutter referred

to it as a curious procedure “involving mother, caretakers and strangers
not only going in and out of rooms every minute for reasons quite
obscure to the child but also not initiating interactions in the way they
might usually do.”"^^ But the Lamb critique of Ainsworth’s work was the
most comprehensive and, some felt, the most insolent. (“He just slan^
dered her,” Sroufe said."^^) That it was made by her former student and
written in a tone that could be read as suggesting there was little value to

her contribution was a personal blow to her. “There was a time,” Inge
Bretherton says, “when it was difficult for her to talk about anything else,

because she was so hurt.”"^"^

Lamb recommended new procedure be devised to replace

that a the

Strange Situation, warned that we really don’t know what much of the

attachment findings really mean, and suggested that, considering its

small sample size and other failings, Ainsworth’s original research be

downgraded to a “pilot study.”"^^ This was perceived as an intolerable
insult. It must have been especially stinging to Ainsworth, since, unbe^
known to people in the field, her Baltimore research had, in fact, been
funded only as a pilot study — she was unable to get any more money for

it — and she had taken secret pride in the fact that she had, with such

limited resources, achieved such a huge scientific triumph.

The leading attachment theorists found Lamb’s tone so debunking
and dismissive toward Ainsworth and claimed that the article was so full

of misstatements of fact that they refused to submit the usual comments

or rebuttals to the journal (Behavioral and Brain Sciences) that published
it. Ainsworth herself maintained a stony public silence. Lamb, despite
his own valuable contributions, instantly became, in Bretherton’s words,
“the black sheep of attachment theory.”"^^
But Lamb had raised legitimate points. Ainsworth’s study was not
unimpeachable. It was difficult, for example, to get perfect reliability

checks on each observer in the home situations (were they definitely

measuring the same thing?), and there were no videotapes to review tor
greater certainty. Also, at the time Lamb was writing, it could justifiably
be argued that Ainsworth’s study had seen virtually nothing in the way ot
replication, which was quite surprising when you think ot the skyscraper

of research and theoretical conclusions balancing on its small base. It

266 The Fate of Early Attachments

must be kept in mind that in every study that began by assessing infants
in a Strange Situation at twelve or eighteen months and that continued
to evaluate those children for years afterward, assumptions were being
made about the style of parenting each child had received, and conclu'
sions were being drawn about the effect that style had had on the child.

But the style of parenting was never assessed the way Ainsworth assessed
it. A number of studies, including Sroufe’s, assessed maternal sensitivity
in the first year, sometimes over the course of a number of hours when
the baby was six months old, and those assessments predicted attach'
ment classifications in the expected way. But there was nothing like

Ainsworth’s seventy'two hours of in'home observation over the course

ot the whole year. In many other cases there was no assessment of the
quality of care at all; it was only inferred from the infant’s Strange
Situation classification. And that inference was possible mainly because
of Ainsworth’s twenty-three Baltimore families. If her study was flawed
and the correlations it demonstrated opened to question, the whole
attachment edifice would begin to wobble.
Ainsworth was not insensitive to this and wanted to see more replica-
tions. She regretted not having had the opportunity herself. But longitu-
dinal studies of that magnitude take time, money, and patience. As
Bretherton says, “If you have thirty children and you’re observing them
once a month for four hours, there’s just an incredible amount of mater-
ial that you have to go through and analyze”"^^ — which may require
more funding, and more grant proposal writing, for several years hence.
Young workers prefer to expend such efforts breaking new ground rather
than tilling the old. Since Lamb’s critique, however, the Grossmanns
have reported their replication Germany with home visits throughout
in

the first year,"^^ and Ainsworth rightly asserts that much has been estab-
lished in partial replication by researchers who performed detailed clini-

cal ratings of maternal sensitivity if only for a few hours. Nevertheless,

legitimate questions have lingered.
 PART IV

Give Parents a Break!

Nature-Nurture Erupts Anew
 20
Born That Wav?
Stella Chess
AND THE Difficult Child

By the 1950s it was becoming increasingly rare for young parents in

urban America to be living with or near their families of origin. And

even if they were nearby, the cultural bonds that had held them together
had loosened to such an extent that, far from being models and accepted
figures of authority, the parents, grandparents, uncles, and aunts of young
adults were often seen as imperious and intrusive. As a result, many new
parents, especially among the nontraditional urban middle class, no
longer had the guidance (or interference) from older generations in car-
ing for their children.
At the same time, the largely new fields of psychology, sociology,

anthropology, and education had all developed huge funds of knowledge

about child rearing and were eager to impart them to the masses. Parents
received instruction on proper baby care from the child guidance clinics,
which had by now spread all over the country and had had a huge effect
on child psychiatry; from nursery school teachers; from social workers;

from books written by various experts; from magazine and newspaper

articles written by authorities, nonauthorities, and ignoramuses alike.

For many years, John B. Watson’s behavioral formula had held sway.
 —
270 Nature-Nurture Erupts Anew

favoring a strict regime and warning parents against coddling. These

ideas had waned after the thirties but still carried weight. By the early

1950s a countertrend developed advocating a much more permissive

child rearing. It had been influenced by new ideas in progressive ediica^

tion; by the continuing popularization of psychoanalytic theory (which

was quoted frequently by people who had only the faintest understanding
of it and often cited aspects of it that had long ago been abandoned as

obsolete); and by cultural anthropologists, some of whom had discovered

that certain primitive tribes whose members were happily adjusted
indeed quite free of the anxieties typical in industrialized countries
raised their children without imposing any of the discipline common in

the West. The permissive crusade advocated leniency in all things to

avoid unnecessary frustration, prolonged breast feeding (just recently
considered a barbaric throwback), feeding on demand, and relaxed bowel
and bladder training. The new regime was considered not just a happier,
less pressured way to raise a baby, but essential to good mental health.^
As a result of all the advice, much of it conflicting, that now bom^
barded parents, as well as a parallel onslaught from advertisers who
swooped in to exploit parental anxieties in order to sell all manner of
products, many new mothers, especially those who were educated and
sought to be well informed, felt overwhelmed and more uncertain of
themselves than ever. The result, as Benjamin Spock noted (in 1953),

was that they had even more problems raising their children. To some
observers at the time, the atmosphere seemed positively totalitarian. It

was as if parents were cringing in their own homes, while some ethereal
outside authority judged their every move.^ Bowlby’s report on maternal
deprivation, once in the hands of more zealous crusaders than he, became
fuel for this fire, contributing to the anxieties parents felt about making a
wrong move. Indeed, at the very moment that he was trying to prove that
mothers are terribly important to their babies and that what they do or
fail to do has lasting impact on their children’s personality and adapta^
tion, a ferocious caricature of his philosophy was already taking shape.
Millicent McIntosh, president of Barnard College, concerned that
young women were becoming frightened at the prospect of becoming
mothers by what they were learning, in child development courses, was
among those who protested: “Even the most innocent-appearing act or a
carelessly spoken word may ‘harm’ a child or ‘damage’ his future happi-
ness. You hurt them by comparing them to other children; you hurt them
by not comparing them and praising them for being special; you hurt
 Born That Way ^ 271

them by being too affectionate to them and by not being affectionate

enough.”^
While some of this avalanche of advice was useful, much of it was
meaningless because it attempted to tell parents what they should feel.

Hilde Bruch, offended by the widespread misuse of psychoanalytic the^

ory, wrote, “It is fallacious to assume that psychological advice can be
carried out in this way. . . . Children respond to the genuine, though
often unconscious emotional attitudes of their parents and are not fooled
hy intellectualized and pretended feelings.”'^

It was in this uniquely American atmosphere that Stella Chess, the

child psychiatrist who had worked with Loretta Bender on the children’s

ward at Bellevue in the thirties, began the study that would define her
career. Chess was disturbed by the assumption common among mental
health workers that if the child had a problem one need only look at the
mother for its source. “Many of the mothers of problem children,” she
reported in 1958, “develop enormous guilt feelings due to the assumption
that they must necessarily be solely responsible for their children’s emo'
tional difficulties. With this guilt comes anxiety, defensiveness, increased

pressures on the children, and even hostility toward them for ‘exposing’

the mother’s inadequacy by their disturbed behavior.”^

In an editorial in the American Journal of Orthopsychiatry, deliciously
titled “Mai de Mere,” Chess cited several cases of childhood disturbance

in which speculations were made “concerning the mother’s relationship

with her own parents, her degree of immaturity, her presumed rejection
of this child and her overcompensations for this rejection.” In each case,
however, further examination revealed that the problem lay elsewhere.
In one case, of school failure, the child turned out to be retarded, a con-
dition that had gone undiagnosed through age eight. In another, that ot a
fourteen^year^old delinquent girl, the pressures of poverty and racial dis-
crimination had been overlooked. The third case follows:

An ICyear'oId boy is argumentative in class, fights the

teacher’s statements of fact, and is greatly upset by marks less

than superlative. When criticized by his parents he becomes

angry, storms into his room only to emerge cheerfully ten
minutes later. He is a “sore loser,” yet is very popular with
friends. In the course of psychological testing he describes his

figure drawing as a boy, age seven. Mother is then asked

whether anything important happened at age seven, and she
272 Nature -Nurture Erupts Anew

recalls that this is the year in which she returned to her suS'
pended magazine editorial work.^

This creates an immediate “Ah'hah!” among those assessing the case

who now quickly attribute the boy’s problems to maternal desertion. It is

assumed that the boy needs psychotherapy and that the mother needs to
resolve “her ambivalence toward the maternal role,” by giving up her
job.^ What is not known, because no one bothered to look into it, is that
the boy has been inclined toward stormy episodes from earliest infancy.
Far from being a reaction to his mother’s return to work, his bursts of
temper at school and with friends appeared to be a manifestation of an
inborn personality trait.

As far as Chess was concerned, the child always brought with him a
great deal of who he was. This obvious fact had long been plain to par^
ents, especially if they had more than one child, as well as to nurses and
pediatricians who worked with newborns and saw the striking variety of
temperamental types that emerged at birth. Chess did not believe that
genes were destiny, but that the child’s inborn qualities interacted with
the type of care he received and other aspects of the environment,
including the cultural milieu, to produce his ultimate constellation of
personality characteristics.
Like Bowlby, Chess had been influenced by the work of David Levy,
but in the opposite direction. A former supervisor of hers. Levy had writ-
ten one of the pioneering works on maternal deprivation. He was also
one of the first psychiatric researchers to take note of inborn tempera-
ment. In his studies of “maternal overprotection,” Levy found that chil-
dren whose mothers tended to fret over them needlessly often had
significant personality problems. But they were not all of the same type.
Some of the children became tyrants, dictating their bedtimes, their

menus, when they would come in from play. The mothers catered to
their wishes, meekly offering a new food if the first one was rejected. The
other children, equally coddled, seemed like little robots. They passively
accepted their mothers’ excessive ministrations and seemed to have little

initiative of their own. Levy held that the parents were not that different
from one another hut that they were dealing with different types of chil-
«

dren. So that it was not maternal overprotection per se that determined

the child’s personality, hut rather the child’s reaction to it, based on his
innate proclivities. Children who are inherently active react to such
 Born That Way? 273

mothers by becoming domineering; children who are inherently passive

react by becoming automatons.^

Many others had held that personality traits were inherited. Plato
argued (metaphorically) that men were composed of gold, silver, or baser

metals and that the men of gold should rule. Hippocrates, addressing
himself more to temperament than to intelligence or general worthiness,
divided mankind into four inborn personality types: phlegmatic, melan^
cholic, sanguine, and choleric. Francis Gabon, the great nineteenth'Cen^
tury English scientist and the father of modern behavioral genetics, did a
pioneering study of hereditary genius, which helped launch the eugenics
movement, whose name he coined.^ Carl Jung proposed that people were
inherently introverts or extroverts. More recently, William H. Sheldon,
an American psychologist and physician, had argued that human beings
fall into one of three constitutional types — endomorph (round and soft),

mesomorph (strong, athletic), or ectomorph — and

(thin, delicate) that

each had certain emotional styles — easygoing and dominant

tolerant;

and aggressive; sensitive, cerebral, and high-strung. But Levy’s empiri-

— and recognition
cal data his of the interplay between heredity and
environment — were unique.
“I saw it in my own children,” Chess recalls, speaking of inherited
styles. “If one of them came yelling and shrieking, I just knew automati-
cally that you couldn’t tell by the amount of yelling and shrieking how
important it was, because that’s the way things had always been with that

particular child. If another one came running into the house with a
puckered and pouting mouth and went directly upstairs, I knew some-
thing dreadful had happened. And I knew that the difference between

the reactions of these two children had nothing to do with the environ-
ment, because they had always been that way.”^^
Another of her children showed marked stranger anxiety. “If he was
eating in his high chair and somebody came to visit. I’d turn around and
there he was, frozen. game of statues. He was frozen in the
It was like the

exact attitude and nothing was going to happen until this stranger went.
I might say. Could you just go out until we finish with the meal? And as
soon as the stranger was gone the movement came back. That’s the way
he was. And then I began to see the same kind of individuality in the

children I was treating and in my friends’ kids whom I’d seen while they
were growing up, and began to talk with my husband about his adult
patients. The power of heredity seemed unmistakable.
 .

274 Nature'Nurture Erupts Anew

* * *

Chess, a thin, attractive woman, and a grandmother many times over,

turned eighty in 1994, the year after we met. She has a delightful,

straightforward, refreshing style. Her speech is loaded with parenthetical

statements, and you can get a whole education in child psychiatry or the
history of the New York psychiatric community before she finishes a sen^
tence. Her husband, Alexander Thomas, also a psychiatrist and former
chief of Bellevue Psychiatric, speaks adoringly of her and reveres her
achievements. Because he is the superior writer, he received first author^
ship on many of their joint works, but he readily credits her as the cre^
ative force.
During the late thirties and early forties when Chess and Thomas
were beginning their careers, the atmosphere in New York psychiatric
circles was so decidedly environmentalist that the interest they were tak^
ing in inborn traits seemed retrograde. “Our friends accused us of going
back to the nineteenth century — constitutional psychopaths, the consti^
tutional inferior, the bad seed. Why are you bothering with that?”^^
In the midTifties Chess and Thomas, later joined by Herbert Birch, a
pediatric researcher, began the now famous New York Longitudinal
Study to determine how infant temperament might contribute to later
clinical problems. The first 136 infants came from eighty educated, mid'
dle'class families; an additional ninety^five came from working-class
Puerto Rican families. Mothers were interviewed extensively about every
aspect of their children’s behavior and responses. The families were fol-
lowed at regular intervals until the children reached adulthood, making
the NYLS one of the seminal studies in child development.
The team found nine variables upon which they could assess the chil-

dren’s temperamental characteristics:

Activity level. “I can’t leave him on the bed or couch because he always
wriggles off’; “He kicks and splashes so in the hath that I always have to
mop up the floor afterward” versus “In the bath he lies quietly and does-
n’t kick”; “In the morning he’s still in the same place he was when he fell

asleep.”

Rhythmicity How regular is the baby’s sleep-wake cycle, his eating

times, elimination, and other biological functions.
Approach or withdrawal. Positive reactions to the new: “He always
smiles at a stranger”; “He loves new toys” versus negative: “When I gave
him his orange juice the first time, he made a face”; “Whenever he sees a
stranger he cries.”
 Born That Way? 275

Adaptability. Not the child’s initial response to the new, but how he
adapts to it over time: “At first he used to hold himself perfectly stiff in

the bath, but now he kicks a little and pats the water with his hand” ver-

sus “Every time he sees the scissors he starts to scream and pull his hand
away, so now 1 cut his nails when he’s sleeping.”

Intensity of reaction. “He cries loud and long whenever the sun shines
in his eyes”; “Whenever she hears music she begins to laugh loudly and
to jump up and down in time to it”; “When he is hungry he starts to cry,

and this builds up to a scream, and we can’t distract him by holding or

playing with him” versus “He squints at the bright light but doesn’t cry”;
“If he does not like a new food he just holds it in his mouth without
swallowing and then lets it drool out”; “When other children take a toy
away from him, he plays with something else; he doesn’t try to get it back
or cry.”
Threshold of responsiveness. How intense does an external stimulus
have to be before the child reacts? “You can shine a bright light in his

eyes and he doesn’t even blink, but if a door closes he startles and looks

up” — high visual threshold, low auditory threshold. “He doesn’t pay any
attention to new people; he doesn’t cry, but he doesn’t respond to them
either” — high threshold for social relations. “He laughs and smiles at a
stranger, and starts to cry if they don’t play with him” — low threshold.
Quality of mood. Positive: “Whenever he sees me begin to warm his
bottle he begins to smile and coo”; “He loves to look out of the window.
He jumps up and down and laughs”; “If he’s not laughing and smiling, I

know he’s getting sick.” Negative: “He cries at almost every stranger, and
those that he doesn’t cry at he hits”; “I’ve tried to teach him not to

knock down little girls and sit on them in the playground, so now he
knocks them down and doesn’t sit on them”; “Every time he sees food he
doesn’t like he starts to fuss and whine until I take it off the table.”

Distrac ability. If he’s about to chew up the electrical cord can he be

thrown off course with a toy? If he’s hungry and crying, will he stop when
picked up?
Attention span and persistence. How long does he tend to stay involved
in an activity once begun? And how determined is he to stay involved
regardless of obstacles? For example, will he continue to struggle with a

new toy even though he can’t get it to work? This category, at a later age,

includes frustration tolerance.

Any parent reading this list will immediately recognize aspects of his
own child and will probably have little trouble determining whether the
276 Nature -Nurture Erupts Anew

child is high, low, or middle on many of the scales. But Chess and her co-
authors went a step further, creating four broad categories into which
they were able to fit each of the babies in the study. “Difficult babies”
(about 10 percent) were those who displayed negative mood, were irreg-
ular in biological functioning, were slow to adapt, tended to withdraw in
new situations, and reacted with high intensity. Another group of babies,
who were also negative in mood, slow to adapt, and inclined to withdraw
from novel situations she labeled “slow to warm up” (about 15 percent).
Unlike the difficult babies, these children were not necessarily high in
activity and reacted with low to moderate rather than high intensity.

Chess defined “easy babies” (about 40 percent of the sample) as those

who were positive in mood, regular in body functions, and quick to adapt.
They also tended to approach the novel rather than withdraw, and to
react with low to moderate intensity. The remaining babies (about 35
percent), displayed various mixes of the temperamental qualities.
Parents, teachers, and researchers associated with the study all used
pejorative terms, including “mother killers,” to characterize the children
who fell into the difficult temperament group. They slept irregularly.
They were hard to train regardless of the techniques used. They typically
reacted to frustration with violent tantrums and were just as likely to cry
loudly over a scratch as a serious wound. They became obsessed with cer-
tain foods to the exclusion of all others, then abruptly changed, would
not touch the favored food, and would have to eat something else all the
time instead. Toilet training was often an ordeal because they lacked bio-
They were slow to adapt, needing repeated exposures
logical regularity.

to anything before accepting it. (And if parents were timid and intro-

duced a new thing only sporadically, the initial negative reaction would
only he prolonged.) Adverse responses to the new meant an endless
stream of shrieks and wails because so much in a baby’s life is new. If,

with long and patient effort, he comes to like the bath he hated at first,

the difficult baby may now howl when taken out. Parents could never
know what the baby might like in the end, because he disliked every-
thing in the beginning. Going to school of any kind was certain to elicit

howls and protests. Each new demand invited a new tantrum. Some of
these children ended up having learning problems as a result of not
adapting to the school environment. Of these, some were wrongly seen
as mentally slow.
Chess found that parents of difficult children were no different on the
whole from other parents, neither in their attitudes about having a child
 Bom That Way? 277

nor philosophy about raising them; although some of them did develop
unfavorable attitudes after having lived with their difficult child for a
time. But no parent is free of emotional hang'ups, and the difficult chib
dren were much more likely to push the buttons of their parents and
cause those hang'ups to get worked into the relationship with the child.
One mother, guilty that she’s not giving or maternal enough, is deter-
mined to respond to the child’s every cry, stay up with him at night, and
virtually become his slave — a course of behavior that might placate the
child in the short run, but is ruinous to both his personality development
and ultimate adaptation. Another mother, competitive and needing con-
trol, sees the child’s negative mood, failure to fall asleep when put down
at night, and food refusal as an attempt to control her or as a conscious
defiance. But her insistence on compliance only heightens his negative

responses. Yet a third is thrown into a rage by the child’s strident alterca-
tions with his younger siblings, and soon brands him a troublemaker.
Whatever the parents’ emotional makeup, they are likely to find a hook
in the difficult child to hang their hang-ups on.

The difficult children in Chess’s study needed patience and consis-

tency and only a few parents seemed able to give it across the board. The
successful parents were either remarkably free of the sorts of intense emo-
tional reactions these children tended to elicit or, more likely, they were
able to control those reactions and remind themselves that many of their
reactions were inappropriate. Resourceful parents would spell each other,
the father staying home with the screaming child while the mother had
time away to recuperate, and they would alternate in taking him on
excursions with their inevitable tantrums. Clearly, it seemed to take an
unusual parent — patient, good-humored, consistent, firm — to handle

these children well. (And since the problems seemed to be inherited, the

parent might have the same temperamental difficulties as the child.) In

an exceptional book, written for parents, the first of its kind to integrate
the nature and nurture perspectives. Chess, Thomas, and Birch offered

reassuring examples of how difficult some children can be and how best

they can be handled:

Jane’s mother seemed inexhaustibly patient and consistent

with her. When Jane was one, two, and even three years old, a
denial in the supermarket would turn her into a screaming,
kicking little fury. But her mother almost never blew up. She
would patiently pick Jane off the floor, take the purchases to
 .

278 Nature 'Nurture Erupts Anew

the check-out counter, and go home without screaming or

fussing hack at Jane. These tactics were markedly successful,

and neighbors were frequently amazed to see the youngster
playing contentedly a few minutes later. When, time after . . .

time, her violent demands brought firm, consistent, and quiet

removal, her tantrums diminished. In time they began to look
more and more like token attempts at self-assertion. For-

tunately, neighbors and relatives took their cues from jane’s

parents. They let her scream, hut refused to let her inconve-
nience others. They were pleasant and ungrudging when the

child made her lightning switch to positive behavior.

jane’s mother alerted her nursery school and kindergarten teachers to

her temperament, and they too learned to react to her with firmness and
calm, jane continued to display intense reactions at times, hut on the
whole she became a cheerful and cooperative girl.

Tommy, another child who fit into the difficult group, had very differ-

ent parents:

They felt harassed from the beginning. When the child’s

pattern persisted as he passed his first, then his second, and
finally his third birthday, they grew increasingly upset. Noth-
ing seemed to satisfy Tommy, even though they were con-
stantly trying to make him happy. They gave in to his
demands, reasonable or unreasonable. If he wanted a toy in

the supermarket, his mother bought it quickly to avoid trou-

ble. Rut no matter how much his parents tried to satisfy him,
every excursion, every visit, indeed every play period was
marked by some commotion. . .

ITommy’s motherl could take much and then would

just so

explode, screaming: “You always make trouble. Nobody can

.satisfy you.’’ She would also make endless threats hut would
not carry them out. Tommy’s father kept akxT from this fran-
tic interplay between mother and child. One fuss from his son
and Daddy '
left him to his own •
devices. As time went on there
was less and less contact between the two. B(.)th parents were
convinced that Tommy was impossible to please or satisfy.

Indeed, this had become true, not only at home, hut at play
and in school.*^
 Bom That Way? 279

Chess was particularly critical of the impact psychoanalysis and re-

lated theories were having on the beleaguered parents of such children:

In these theories, a loving and accepting mother should have

a happy and contented child, from which it follows that an
unconscious maternal attitude of rejection could be the only
explanation for a difficult screaming child. As a result of
reliance on these theories it was not unusual for the mother of
a difficult infant who screamed frequently and who made all

routines a crisis, to develop self-doubts and feelings of guilt,

anxiety, and helplessness unless she was unusually confident

in her adequacy as a woman and a mother.^®

Just as Bowlby had said, in effect, that behavioral theory was fine and
useful as long as it didn’t overreach its boundaries and apply itself to

areas where its explanatory power was nil (as in attachment or speech
acquisition), so Chess was saying that psychodynamic theories were fine

as long as they didn’t overstep their boundaries and attempt to explain

temperament in terms of maternal influence. It is only later, when we see

how the child with a given temperament has fared, that we can begin to

see what impact the parents have had.

The slow-to-warm-up child also presented a challenge to many parents.

The initial negative response of these children is typically quiet. They
cling, they withdraw, they turn away, they fuss mildly. “Their first few
weeks in nursery school,” Chess wrote, “are typically spent on the side-

lines, quietly watching the activity of the group. In elementary school, a

new academic subject may evoke the announcement, ‘I don’t like it,’ or,

more typically, initial silent nonparticipation.”^^

She recalls that it was common when she first started work for nursery

school teachers to tell parents that a child who was shy and slow to warm
up was anxious and needed treatment. Such parents often brought their
children to see Chess for an evaluation. Invariably she would ask the par-
ents of these shy children to bring a toy from home so the child would
not have to use a toy that was not familiar. “I’d say, ‘Oh, I see you
brought a toy from home, is this the way you use it?’ and I’d do it all

wrong. The kid would forget he was supposed to be shy and say” — Chess
uses a reprimanding falsetto
— ‘“That’s not the way you’re supposed to do
it! This is the way you’re supposed to do it!’ ‘You mean like this?’ And
 ^

280 Nature 'Nurture Erupts Anew

we’d be interacting.” Further inquiry of the parents might reveal that

something had happened which had compromised the child’s adaptation.

He may, for instance, have been out sick a lot and never had a chance to
feel familiar at school. “So the school only saw this child who was stand'
ing at the periphery. Well, I also saw a child who was standing at the

periphery, but he wasn’t scared of me. Otherwise how could he come out
and tell me was
I doing it all wrong
One slow'tO'Warm'up child was four'year^old Stanley who, with his
two older sisters, was taken on a weekend trip to a seaside motel.

The three children [Chess wrotel were put to bed in the same
room, and when the parents opened the adjoining door a half
hour later to check on them, the two older children were
asleep in their respective beds while Stanley was sitting in the
middle of the floor, solemn but not crying. Their attempts to
induce him to get back into bed were fruitless. He simply
repeated “I don’t like that bed.” The parents, prepared for this
response by similar incidents in other new
him places, picked

up, tucked him into bed, made soothing noises, sat with him
for a bit, and then left. As his mother was on her way out of
the room, she heard the child climbing out of the bed again.
When she asked him his intentions, he said, “Sit on the
floor,” and was apparently prepared to sit out the night. When
he was brought into the parents’ bedroom and provided with a
bed more closely resembling his crib, he promptly fell asleep.^

The patient acceptance of such parents paid off handsomely as chib

dren like Stanley grew older. By the time he was five, Ralph, another
slow'tO'warm'up boy,

was a well-adjusted member of his kindergarten group. He

looked forward to going to school, greeted his playmates
pleasantly, and visited back and forth with his friends after

school. He continued to be placid rather than exuberant,

even with good friends, but ^he was clearly contented and
happy. When he visited in a new home, went to a new place,
or met new children, he still took a long time to function eas-
ily. But both he and his family had learned that his shyness
would wear off in time, and everyone was willing to wait.
 Bom That Way? 281

Ralph’s parents had come to understand very early that his

hesitancy about accepting the unfamiliar needed to be bon-
ored.^^

Different parents, of course, react differently to the slow^to^warm-up

child, based on their own temperament, philosophy, and psychology.
Some mothers are impatient and fast^moving, can’t understand why
their child is not the same, and unwilling to make what seems an exces-
sive accommodation to his needs. Other parents are disturbed that their
child is hesitant and unsure of himself in new situations and take it as an
intolerable reflection on themselves. Still others give up too easily so
that the child never has a chance to get used to new things, thereby sur-

rendering him to a narrow range of foods, friends, and activities — and

hence a narrow life and diminished selhconcept. Teachers are likely to

see the slow-to^warm-up child as insecure or maladjusted, which could

create additional problems for him.
Jerry’s parents, like Ralph’s and Stanley’s, were also nicely attuned to
his temperament, but his mother’s attitude changed when she began tak-
ing him to the playground and was embarrassed by his clinging, anxious

reaction to new situations.

The mother was dismayed, since to her, as to many other

American mothers today, this moment was a big test of social
maturity and healthy personality development. She felt that
jerry had failed this test and was sure that the other mothers
in the playground were blaming her for having such an “anx'
iously” timid baby. They probably were.
. . . Instead of holding the child in her lap or giving him
a familiar toy to play with near her until he warmed up to the
new setting, she began to push him insistently to play “like

the other little boys.” The more she pressured, the more he
clung. The more he clung, the more pressure she applied.

Finally she gave up taking him to the playground at all.

In kindergarten Jerry did not participate with the other children. He

dreaded going, cried, and clung to his mother. “He had in fact, become
the anxious, fearful child his mother dreaded.
282 Nature-Nurture Erupts Anew

If difficult children could make otherwise reasonably normal parents into

tyrants or saps, easy children seemed to be able to make almost anyone
into a paragon. They fussed little, smiled a lot, fell into regular patterns

of eating and sleeping, including daytime naps, enjoyed new experiences,

smiled and babbled at strangers, and rarely cried unless hungry or sleepy.
As they got older, they were more able than other children to see the

good in an otherwise disappointing situation and were most likely to

comfort other children in distress.

Nevertheless, some easy children did develop problems. They were

not immune to traumatic events, such as death of a family member or
lengthy separations. But also their very adaptability made them suscepti'
ble to the neurotic demands of parents and to replicating the dysfunc-
tional qualities they see in them, “sometimes,” as Chess notes, “to the
point of caricature. It would seem that this type of child would be par-
ticularly vulnerable to the demands of an insecure mother who wants to
be taken care of herself, leading to a false caretaking self (and perhaps in
later life into a caretaking profession). Some of these children developed
a mode of functioning at home that met with disapproval or even deri-

sion from peers or teachers. (Chess cites the case of a mother who had
turned her easy, adaptable three-year-old into a perfectly mannered little

aristocrat.) Easy children were not immune to disappointing their par-

ents either — if, for example, they turned out to be not as bright as the

parents hoped or if they learned a behavioral style from the parents that
the parents hated in themselves.
Easy children can also run into problems with well-meaning parents
who fail to give them what they need. Chess cited the case of Pammy, a
winning hahy who was raised by parents who felt their own childhoods
had been spoiled by perfectionistic parents for whom nothing was ever
good enough. Wanting to spare Pammy the pain they’d suffered, they
“Whenever they tried tentatively
failed to set limits for her. to do so,

Pammy’s charm would always divert them from their purpose.

Pammy not only failed to learn limits and responsibility, she began
relying on an ever more grating cuteness to avoid every sort of request.
She did not learn the rules of games, and so, when she went to school,
did not play with the other children.
«
Although her IQ was normal,
she began to experience herself as stupid. By age seven, she was a virtual
misfit.

In studying these cases, where parents’ reactions to their children

turned out to he contrary to their temperamental needs. Chess came
 Born That Way? 283

across a number of cases where the temperament of the parents and the
temperament of the child seemed to represent a poor fit. She cites the

case of the child who is very high in activity. Some parents are delighted

by such a child. He may be always getting into things he shouldn’t be,

always disappearing to run off in some new direction, always on the verge
of some physical disaster, but they are charmed by his energy and patient
in teaching him rules. Other parents find the highly active child intoler-
able. They may have less stamina, need more control, or be unwilling to

put their fine breakable objects in storage. They find themselves con-
stantly correcting the child, criticizing him, admonishing him to “Sit

still,” “Don’t touch,” “Put that down,” “Stop running.”^^ As a result their
child not only becomes resentful of all instruction, he fails to learn the

difference between important and unimportant commands. He develops

a chip on his shoulder, which carries over into his relations with teachers
and peers.

Another example of a poor fit occurs between lively, exuberant par-

ents and a child who is low in positive affect and high in persistence.

Susie, for instance, would work for long periods on a task that interested

her, but never seemed, as her parents said, “really enthusiastic.” They
wanted her to lighten up, to be less fussy, to show some feeling. By the
time Susie was four, her mother had become so fed up with her lack of
demonstrativeness she would exclaim, “Aren’t you ever satisfied? I’m not
going to take you out any more because you don’t really like anything I

do To make matters worse for Susie, she had two younger broth-
for you.”

ers who were much more like her parents in temperament and to whom

they gave more and more of their positive attention. Susie began to
believe that she was stupid and couldn’t do anything right. “Quiet persis-
tence,” Chess wrote, “was her strength, but it earned her little positive

recognition from her father and mother.”^^

As the study progressed. Chess found that a number of the children dis-

played emotional or behavioral problems that required attention.

Children born with difficult temperament predominated. Indeed, 70 per-

cent of them ended up with behavior problems — phobias, learning diffi-

culties, stealing, lying, defiance, excessive concern with pleasing others.

Not, Chess insisted, because this is the inevitable fate of children with

this temperament, but rather because they were more likely than other
children to encounter an unsatisfactory environmental response.
Although Chess and her colleagues found that temperament made a
284 Nature 'Nurture Erupts Anew

powerful contribution to personality, and that difficult or aggressive tem-

perament may predict adjustment problems later, the environment
always played a role. Their close scrutiny of the children and their par-
ents over the years revealed what enormously different psychological

outcomes were possible for children with very similar temperamental

profiles.

Chess also found that family influences could either be attenuated or

compounded by outside forces, like peers and teachers. “A parent may be
fostering negativistic tendencies in a slowly adaptive child by insisting

on quick adaptations to various new situations that are either very diffi-
cult or impossible for that youngster to make. If the nursery school
teacher makes similar demands, the influences encouraging negativistic
trends will be intensified. If she allows [him] to adapt at his own pace,”
however, the negative parental influence will be alleviated.
Culture had to be considered a factor as well: “It must be emphasized,”
she wrote, “that the temperamental constellation we label as difficult is
difficult in those cultures such as ours in which demands are made on the
young child for the early establishment of regular sleep and feeding
schedules, for quick adaptation to new situations and new people, and for
cheerful acceptance of the rules and routines of the family.”^^ To illus-
trate this point, she referred to a study of families in rural Kenya where
researchers had identified a group of children who were biologically com-
parable to the children considered difficult in her sample:

However, for the Kenyan mothers, these babies were not diffi-

cult to manage. Irregularity in sleep or feeding rhythms was no

problem, since the infant was always with the mother or an
older sib, slept beside the mother and was nursed whenever he
or she awoke during the night. Fussing and slow adaptability
were not issues of concern, because these parents, unlike so
many American parents, did not view such behavior as alarm-
ing signals of a beginning behavior problem. Rather, their
common statement was that “one cannot know about person-
ality until the child is old enough, about 6 or 7 years, to take
responsibility in household economic chores.”

To Chess, interaction was all. Although she had impressive statistics

on the power of inhorn temperament, she did not want to fall into the
trap that she’d seen psychiatry fall into so often before, of launching a
 Born That Way? 285

general theory based on a fragment. Rather, she saw temperament and

environment as continually interacting and modifying each other.

Children have an impact on their parents, causing them to be certain

things they might not be otherwise, at least not so strongly. They have
traits that may push their parents’ buttons, eliciting reactions another

child might not get. How easy it is to care for a child, how well his tem^

perament dovetails with the parents’, how well his behavior fits with
their prior expectations will all influence their behavior and their atti-

tude toward the child. By the same token, children can change. Parents
and others can help a child to learn how to deal with himself. They can
help him to adapt to situations that are hard for him. A tendency toward
negative mood may become more positive if caretaking has been benign.
Similarly, a highly adaptable child, repeatedly confronted with impossi'

hie demands, may become less and less adaptable over time.^^
Chess’s work on temperament and its interaction with the environ^

ment has continued for over three decades. Reports on the New York
Longitudinal Study and the new conclusions drawn from it have
appeared in the 1960s, 1970s, and 1980s, as the children from the sample
have reached adulthood and formed their own families. The work she’s

done with Thomas and Birch has moved child care away from the one-
styleTits-all recommendations formerly given to parents by pediatricians

and other advisors to a recognition that different children have different

needs. Some children do well being fed on demand, others do not. Some
can easily be toilet^trained at an early age, some do better later. Some
easily adapt to new learning situations in school, some need encourage^
ment, persistent pressure, or repetition of the new activity before they
will get fully and effectively involved. Sensitivity to such differences in

children is apparent now in books such as those addressed to parents by

pediatrician Berry Brazelton, in which he illustrates childhood develop-
ment by following several children with different temperamental styles.^^

In an age where deviations from the cultural ideal (uptightness with

the new, anxiety about change, the lack of assertive self-confidence, a
high level of aggressiveness) are quickly seen as character flaws. Chess
and her colleagues have also been a force for tolerance. She has helped
make parents and child care workers more aware of the fact that a child
can have a disposition that the culture does not value without that dis-

position having been created in some way by pernicious parenting. Her

work has helped ease the pressure on parents. With knowledge of tem-
peramental differences, they could perhaps see a tantrumy baby without
286 Nature-Nurture Erupts Anew

feeling ashamed or guilty or enraged, but just know that this is biologi'

cally natural and they face a difficult task.

Chess’s work could also be seen as a reconfirmation of certain aspects

of the Bowlby^Ainsworth thesis, for it reinforced the importance of sen^

sitive care. Indeed, it took the issue to a higher level of sophistication,
because she not only recognized that sensitive care means different
things to different babies (as Ainsworth did), but documented in detail

what those differences were for specific temperamental types.

But it was not seen this way hy either camp. For one, Bowlby and
Ainsworth were viewed by Chess and her allies as radical environmen-
talists and parent hlamers. It seemed to Chess and to those who followed
her in the examination of hereditary traits that behavior in the Strange
Situation was as much a factor of temperament as anything else. In the
data that Sroufe reported, children in the behavior problems group could
easily be seen as children who were difficult from birth, while the secure

group looked an awful lot like the easy child, right down to the capacity
for empathy and the good peer relations. Even the percentages were sim-
ilar. Chess also believed that attachment theory was overly focused on
the first year and ignorant of how radically things could be altered, for
better or worse, thereafter; and that it was overly concerned with the
mother to the exclusion of other family, community, and social influ-
ences. In such views, she was hardly alone.
For their part, attachment theorists saw problems with Chess’s meth-
odology: not collecting data on temperament until the child was three
months old, hy which time mothers would have left a considerable
imprint; not appreciating the immense impact of the mother in the earli-
est days, especially in enabling the baby to achieve stable biological
rhythms; relying too much on parental reports, which are notoriously
biased (Ainsworth felt particularly strongly about this from her own
experience); and mistaking the effects of environmental influence
(schoolyard aggression, for example) for inherited traits. Others, includ-
ing fellow genetic researchers, doubted whether the nine temperamental
criteria reflected actual biological traits or simply convenient groupings
of characteristics. The strongest objections from environmentalists were
reserved for the “poorness of fit” concept. Attachment theorists and like-
minded clinicians were hoping to educate parents about their babies’
needs and how best to respond to them. They grudgingly acknowledged
that responding sensitively might he harder for certain parents with cer-
tain babies, hut what good could come of a concept that seemed to
 Bom That Wayl 287

acquiesce to those hardships, as if there was nothing a poor-fitting parent

could do? Part of good parental attunement meant adjusting to achieve a
good fit with the baby. What’s more, it appeared at times that Chess only
saw parent-child conflict in poorness-of-fit terms and was insensitive to
the fact that many parents are anxious, self-centered, depressed, angry,
imperious, manipulative, detached — and that, without outside help,
they cause their children to become disturbed.* In Levy’s work, after all,

when babies were faced with an anxious, intrusive, overprotective

mother, they became disturbed. The baby’s temperament only affected

how, not whether.^^

More radical environmentalists saw the poorness-of-fit idea as a per-

nicious effort to place baby and parent on the same plane, with equal
responsibility for adjusting to each other and making the relationship
work. Presumably, now, parents could throw up their hands and say of
their child, “We just weren’t meant for each other,” rather than face
the fact that they are the ones, regardless of their own temperaments,
attitudes, or countervailing emotions, who must struggle to give of

themselves in the way the baby needs them. Chess, had not, of course,
meant the fit doctrine to be taken this way, but it was as if the concept

was born on a slant and could barely be probed without rolling in that

direction.
The debate over goodness or poorness of fit, which soon engulfed
developmental psychology, spilled over into the more relationship-
oriented wing of American psychoanalysis, which was slowly influenced
by the wave of infant research taking place in the 1970s. Fashions in
analysis had swung widely on the nature-nurture question, from Freud’s

early environmentalist view (the seduction theory) to Klein’s extreme

genetic view (which assumed that the baby arrived with images of people
and fantasies of relationships already in place). In an influential 1983
work on object relations theory. New
York analysts Jay Greenberg and
Stephen Mitchell warned against another overarching swing of the pen-
dulum:

Sullivan and Fairbaim both write as if the only crucial vari-

able is the caretaker, anxious or not, emotionally available or

not. They do not emphasize the extent to which the care-

*When Stella Chess read this in manuscript form, she wrote in the margin, “I do hlame parents at

times — really!” (July 12, 1993).

288 Nature'Nurture Erupts Anew

taker’s responsiveness or lack of responsiveness is keyed to the

particular baby’s style and rhythm.

The authors argued that, according to the latest infant research, the
mother’s success or failure in meeting the infant’s needs must be under'
stood in terms of goodness of fit:

Each baby brings to encounters with caretakers his own par'

ticular rhythm of engagement, level of activity, distinct affec'
tive and behavioral displays. Each caretaker brings to his
encounter with the baby his own style and intensity of respon'
siveness, attention span, level of interest, anxieties, and so
on.^^

Such an even'handed view of of the parent'child relationship made

another New York analyst, Gerald Schoenewolf, seethe: “To attribute
failure to a bad fit is to absolve mothers of responsibility and blame. In

doing this Greenberg and Mitchell are bowing to a feminist trend of the
1970s and 1980s, a trend that in my opinion has been quite destructive
to child rearing.”^^

The table was set for war.

 Renaissance of
Biological Determinism:
The Temperament Debate

Attachment thinkers had made various efforts over the years, mostly pro

forma, to account for inborn temperament. Bowlby spoke of it in his ear-

best papers. In 1940 he wrote, “It should not be assumed that I ignore

inherited difficulties, which are almost certainly significant in a large

number of cases.” (Although in most cases, he added, neurosis was still

unlikely unless the environment had exacerbated whatever problems the

baby started out with.^) Later he described one of his juvenile thieves as
having been “difficult from birth,” while another “was said to have been
quite an easy baby.”^
Similarly when Bowlby first presented his theories of attachment, he
displayed an awareness of the fact that certain babies smiled more than
others, which, he said, caused them to get more love and attention. This,
too, reflected an acceptance of inborn temperamental difference in —
this case regarding the display of positive feeling —
and of the tact that
children influenced the kind of care they received. In 1973, in Sep-
aration, he acknowledged that genetic factors may play a role in suscepti'
hility to fear, which, he said, had been “welhdocumented in the case of

other mammals, e.g. dogs.”^

 290 Nature 'Nurture Erupts Anew

Ainsworth, too, was aware that babies were different from birth. In
evaluating maternal behavior in her Baltimore home study, she wrote
that “babies differ in the kinds of signals they give” and that “it may take
some time before a mother can learn to read the signals of her baby.” She
believed that if mother and baby were to be optimally attuned, a “mutual
adaption” must take place.
But neither Bowlby nor Ainsworth believed that inborn traits

accounted for all that much in the average child. They were thus unpre-
pared for the surge in genetic research among psychologists in the 1970s
and 1980s and the degree to which inherited qualities would be taken as
fundamental in adult personalities. The evidence accumulated in these
studies seemed to offer a stunning refutation of many of their most funda-
mental assumptions.

Male twins, thirty years old, were separated at birth, put up for adoption,
and ultimately raised in different countries. They had no contact with
each other. Yet when psychoanalyst and personality researcher Peter
Neubauer caught up with them at age thirty, they were so similar as to
make the whole concept of environmental influence seem superfluous.
Both demonstrated conspicuous obsessive-compulsive traits. They were
highly preened, religiously punctual, and so anxious about cleanliness
they both regularly scrubbed their hands just short of bleeding.
Neubauer asked each twin independently to account for his unusual
fastidiousness. The first blamed his mother, noting that throughout the
years of his childhood she had been fanatically well ordered, insisting
that everything he returned to its place and that all the clocks (“We had
dozens of clocks”) he “set to the same noonday chime.” The second twin
was equally certain of the origins of his personality style. “The reason is

quite simple,” he said. “I’m reacting to my mother, who was an absolute

sloh.”^

Since the first reports from the New York Longitudinal Study, the idea
of temperament as a quality that the child brought with it from birth had
begun to catch on in developmental circles. Psychologist Richard Bell
had taken up Chess’s idea that the influence between parent and child is

not just one-way, that the infant and its qualities have a strong impact
on the parent and affect the quality of care it receives, and his work on
this subject was widely influential.^ All this had contributed to a rising
concern that mothers were not being treated fairly, neither by psycholo-
 Renaissance of Biological Determinism 291

gists nor by the general public, as evidenced by the assumptions of the

hand'washing twins.
Developmental psychology, meanwhile, had spawned a new disci-

pline. Called behavior genetics, it occasioned a revival of interest in the

idea of inborn personality not seen since the last century. The new
researchers did longitudinal studies of twins reared together, compared
adopted and nonadopted siblings, and, perhaps most dramatically, exam-
ined adult twins who had been raised apart.^
Finding identical twins who were separated at birth is a difficult task,

but it offers a rare and dramatic opportunity to sort out the environmen-

tal from the hereditary. Because identical twins have identical genetic
material, it’s almost like raising the same person in two different environ-
ments and seeing how powerfully the genes shine through. Finnish twins
Daphne Goodship and Barbara Herbert, for example, were separated at
birth and did not meet again for thirty-nine years. They arrived at their
first meeting wearing the same outfit (beige dress and brown velvet
jacket). They discovered that they both suffered miscarriages with their

first pregnancy, then had two boys and a girl (in that order). Both had
weak ankles from falling down stairs at fifteen, and both met their hus-

bands Their mannerisms seemed identical, as did their shapes

at sixteen.

and weight. Identical twins, of course, are as prone to coincidence as

anyone else, and much of such informal data may be little more than
that. But there was more; Their personalities also had prominent echoes,
like their habit of giggling, the fact that they are both
penny-pinchers

(despite being raised in diametrically different socioeconomic

circum-
opinions.^
stances), and their marked guardedness about revealing their
of Minnesota psychologist Thomas Bouchard and his co-
University
identi-
workers, who, since 1979, have examined over 120 pairs of adult
birth,^ believe that
cal twins who were separated within a few months of
tolerance,
individual qualities like imagination, leadership, curiosity,
alienation, and vulnerability to stress are all strongly herita-
sociability,

ble. “When we set up the study,” Bouchard says, “we just assumed that

social attitudes wouldn’t show a genetic influence.” But he noted with

also
excitement that he found otherwise: Religiosity and traditionalism
seemed to have genetic links.

Bouchard began studying identical twins reared apart after reading a

newspaper story about Jim Springer and Jim Lewis. When these “Jim
twins” first met, also at age thirty-nine, they found that each
had con-
 292 Nature-Nurture Erupts Anew

structed a similar circular white bench around a tree in his yard. They
were each affectionate and sentimental and had the same habit of leav-
ing love notes around the house for their wives to find. They had similar
wood-working shops in their basements. They both drove Chevrolets,
drank Miller Lite, and chain-smoked Salem cigarettes. Both had worked
as sheriff’s deputies. Both had had vasectomies. The first wives of both
men were named Linda, and their second wives were both Bettys. One
had named James Allan, the other James Alan. They both had
a son
had dogs named Toy.^^ “I thought for sure the story in the newspaper
was exaggerated,” Bouchard told me. But he discovered it was not. The
finding of such uncanny similarities in twins reared apart (who until
then had been mainly compared in terms of IQ) altered the direction
of Bouchard s career. He would soon find more uncanny material to
match it.^^

Bouchard put all the twins through a huge battery of standard assess-
ments, such as the Vocational Interest Test, the major personality inven-
tories, a life stress interview, and measurements of interest, values, and
expressive style, all of which yielded impressive correlations.^^ But it was
the anecdotal material — much of it reported in the press — that was
often most captivating: Twin men, living in New Jersey, meet for the first
time in adulthood. They find that they are both volunteer firemen, both
like hunting, both married women with the same first name, both drink
only Budweiser and grasp their beer with the pinky lodged underneath.
“We had the same moustache,” said one, “the same sideburns, even the
same glasses.”^"^ Another pair of male twins, reported by Neubauer (who,
unlike Bouchard, followed his subjects from infancy), are introduced to
each other for the first time at nineteen. Both had won boxing champi-
onships, both had artistic interests; even the cavities in their teeth were
in the same location. A
third pair of twins, introduced later in life, share
an extreme chameleon-like capacity for adaptability. Although quite dif-
ferent from each other, they are eerily similar to the people in their sur-

roundings.^^ A fourth pair find they like to play the same oddball
practical joke. A fifth that they share a phobia about water and that they

have an identical means of coping with it when they want to go for a

swim — wading in backward.
It is difficult to overestimate the impact the twin studies have had on
the field. They seem to have established that a great deal of what we
think of as personality traits have, at the very
compo- least, a genetic
nent. Other evidence, including the wealth of material coming out
of
 —
Renaissance of Biological Determinism 293

the Colorado Adoption Project, has been equally compelling. Judy Dunn
and Robert Plomin have reported that biological siblings are quite differ^
ent despite having been treated the same way at the same ages by their
mothers. “What little resemblance there is among siblings,” they con-
cluded, “is due to hereditary similarity, not to the experience of growing
up in the same family.”^^ Recently, more physiologically based studies
of variability in brain chemistry, heartbeat, and hormonal levels in

infants — have added a new level of confirming data. University of Wis-

consin psychologist Richard Davidson, for instance, has found evidence
that people with greater activity in the right frontal cortex of the brain
are prone to be gloomy, while those with more left frontal activity are

prone to be cheerful. Although newborns cannot be tested to deter-

mine if this pattern is present at birth, some researchers feel strongly that
it an inherited quality that may make some people more at risk for
is

depression. As a result of work like this, it is now frequently assumed that

not only certain types of severe depression but other mental disorders
hysterical character, hyperactivity in children, manic-depressive illness,
vulnerability to addiction, schizophrenia, the ohsessive-compulsiveness
disorder of the two hand-washing twins — are genetically based.

Temperament research is still in its infancy and there is much confu-

sion and disagreement among researchers. What is a trait? What is per-

sonality? What is temperament? How can it be measured? Perhaps most

important: Are behavioral or biochemical variations observed or mea-
sured at birth stable traits; are they susceptible to environmental influ-

ence; do they change over time?^^ No one knows for certain what the
building blocks of temperament are. Chess still prefers the nine tempera-

mental variables she and her husband worked out with Herbert Birch.
Others favor the more parsimonious categories emotionality, activity, —
and sociability —
worked out by Plomin and Arnold Buss.^*^ Still other
researchers favor different categorizations.
Most genetic hehaviorists give at least a passing salute to the environ-
ment, acknowledging, to cite an obvious example, that not everyone
whose heredity makes them more vulnerable to addiction becomes a
drug abuser, that something else, something experiential, is needed as

well. They describe temperamental traits as “heritable,” which means

having a genetic component (as opposed to “inherited” — applicable to

traits like eye color — which means fully determined by genes). But the
clear trend hasbeen toward more and more emphatic demonstrations ot
the power of heredity —
and a dismissal of work, such as that being done
 294 Nature 'Nurture Erupts Anew

on attachment, which places such a large emphasis on the child’s experi^

ence.
In this debate, too, Jerome Kagan has figured prominently. Early in his
career Kagan had been an environmentalist and a strong proponent of
maternal influence but had come to believe that the primary con^
stituents of personality are genetic and constitutional (present at birth

but not necessarily genetic) and that many important personality vari'
ables are determined by the chemical broth of hormones, neurotransmit-
ters, and brain peptides in which the brain sits.^^

In 1978 Kagan began studying a sample of children, representing some

15 to 20 percent of the population, who were similar to Chess’s slow-to-
warm-up group. According to Kagan, such children all display a cluster
of behaviors and emotional states that can be summarized as a tendency
toward fretfulness, anxiety, and inhibition. As early as two months of age
these infants are easily hyperaroused, cry, or become fretful in the pres-
ence of novel stimulation, and are often difficult to soothe. Like fearful
strains of laboratory rats, long known to researchers (and quite easy to
breed), they have a distinct, excitable brain chemistry.
In his longitudinal study, begun in the toddler period, Kagan found
that young children with this shy, timid, fearful profile are reluctant to

interact with others and tend to play apart from other children in play
groups. When an unfamiliar event occurs, they’re more anxiously reac-
tive than other children, clinging to their mothers, stopping play, and
cutting short their own chatter, humming, or other vocalizations. When
observed again at age seven and a half, 75 percent retain the pattern,
revealing prominent anxieties, like fear of going to camp, of watching
violent TV shows, of sleeping over at a friend’s house. At ten they still

prefer to play alone rather than in groups. At thirteen, the anxious pro-
file remained. “And, as you might expect,” Kagan says, extrapolating
from other evidence, “extremely shy, phobic children, are likely as adults
to select bureaucratic jobs with minimal risks.

Where Ainsworth sees ambivalent attachment, Kagan sees the inher-

ited susceptibility to stress typical of his fretful children. Where Sroufe
sees children displaying the emotional consequences of secure or anxious
attachment, Kagan sees the unfolding of inborn traits. His work has
led him to the conclusion that people, like dogs, come in a variety of
breeds— some cocker some Dobermans, some
spaniels, terriers, some
Labradors — and no matter what happens we don’t change to us, breeds.
(“Every dog owner knows,” he has said, “that it’s hard to make a terrier
 Renaissance of Biological Determinism 295

behave like a police dog.”^^) The implication of such tenacious patterns

has at times been upsetting to him. “I’ve struggled with this a lot,” he
told me. “My training, my beliefs, my upbringing all came together when
1 was a younger person to lead me to minimize the role of biology. I

believed that even if you started with a little bit of biology that your will

and experience could overcome it. I’ve now seen many children for

whom that’s difficult.

Although he often sounds as if he believes that temperament is des-

tiny, Kagan denies some attachment people charge, a radi-

that he is, as

cal genetic determinist. In particular, he has begun in recent years to

acknowledge the power of family influence, noting, for example, that

inhibited children, if they are lucky enough to grow up in families that
value education, often go far. “Some,” he says, will “select rewarding

careers that require solitary intellectual activity — science, computer pro-

gramming, history, music, and art. T. S. Eliot and John Maynard Keynes,
who were very inhibited children, both won Nobel prizes. In his thir-

teen-year assessment, Kagan found one inhibited child from an intelli-

gent, well-educated family who said he wanted to be an astrophysicist.
Why? Because he likes to work alone. Another boy, from a disadvan-
taged family, was clearly less well adjusted, and seemed to be on his way
to a conduct disorder. While acknowledging such important environ-
mental influences, Kagan nevertheless maintains that the underlying,
inherited style remains. Thus, all of the children in the anxious, inhib-

ited group, “whether they are headed for a conduct disorder or president
of the class,” are low in the expression of positive feeling and display very
few spontaneous smiles.^^
Behavior geneticists like Kagan see their work as providing a source of

optimism. If the fundamental message of attachment research is that

children need to be cared for in a consistent and sensitive way, that they
love their parents powerfully and need to have that love returned and
sustained, then the fundamental message of temperament research is

that people are inherently different, that those differences need to be

tolerated and respected, and that much of what we once saw as

parentally induced is actually part of the natural range ot human differ-

ences. “1 think if we recognize that individuals differ from each other in

these fundamental ways,” Bouchard says, “we’re going to have a lot more
respect for one another. We know that we’re physically different. We
respect and understand that a kid who’s only four and a halt teet tall is

not going to compete with a kid who’s six feet tall. Well, the same may
296 Nature-Nurture Erupts Anew

be true tor many psychological traits and characteristics.”^^

Kagan agrees. “There are some people with a very short fuse. They
blow up easily; it’s hard to get along with them. Many people assume that
it’s a function of their past and they should be able to control it. So then
you get angry at these people. But if you believe that this is partly tern--

peramental, and that their biology prepares them for this, then you
become a little more forgiving.”^^ Kagan also expects temperament
research to take the pressure off mothers who, he believes, have gotten a
raw deal as a result of decades of behaviorist and psychoanalytic influ-
ence. He quotes with pleasure the words of a famous scientist who suffers
from terrible stage fright whenever he makes a speech. After hearing
Kagan present his data, he said, “I’ve been blaming my mother for fifty
years, but after hearing this. I’m going to stop.”^^

The sudden rise of behavior genetics and its direct attacks on attachment
theory caused distress in the attachment ranks. No one was more roiled
by the new trend than Sroufe, who, with Ainsworth’s gradual retirement,
was becoming attachment’s chief spokesman in the United States. His
distress over the issue was apparent the first time we spoke in 1988 :

“There’s a tremendous wave of biological determinism,” he said. “Every-

thing’s genes these days. There have been dozens of papers saying that all
this stuff is just temperament. Hell, we’ll find a gene for security of
attachment any day here!”^^
The behavior-genetic challenge to attachment theory was essentially
two-pronged. One prong indicted the attachment idea for blaming
mothers and ignoring the fact that infants could be difficult or that there
could be a poor fit. This charge still causes feathers to fly: No one will
admit to blaming mothers, and Bowlby himself was at pains to point out
the sympathy with which he, as a clinician, viewed both the mothers and
the children in anxious pairs and the awareness he always retained of the
mother’s own hardships and psychological burdens. What’s more, most
attachment workers readily concede that the father or another person
can be the primary caregiver, so that mothers are not really the issue
except for the fact that in most families they do in fact play that role. Be
that as it may, attachment theorists have not been so eager to let moth-
ers— or caregivers in general— off the hook: They want it understood
that sensitive, consistent parenting is vital, and they see proclaiming
that as part of their mission.
The blame issue is similar to the poorness-of-fit concept in that it
 —
Renaissance of Biological Determinism 297

has been highly charged politically, and the antagonists are often more
concerned about the impact certain types of statements will have
than whether or not they are true. Many developmentalists recognize
that, of course, parents are sometimes to blame for their children’s suh
fering, but they believe that making an issue of it will only tend to gen^
erate guilt — and a guilty parent is more likely to do a poor job than
one who had been reassured and encouraged. Because they believe
and I think for the most part rightly so —
that an atmosphere of guilt

is so destructive, they lean toward never saying anything, even in prO'

fessional contexts, that might suggest that mothers ever behave badly.

PoornesS'ohfit and other temperament'based explanations are more

reassuring.
This prong of the opposition’s thinking has made some inroads into
the attachment canon. Alicia Lieberman, firmly in the attachment
camp, has narrated a training video. Flexible, Fearful, or Feisty ?, to help

acquaint child care workers with temperament. She readily acknowh

edges that some babies make it easier for parents than others. “There are
children,” she says, “who are more resilient and who forgive more easily
and for whom the momentum toward mental health is stronger. They
make a mother into the kind of mother who provides security of attach'
ment because of who they are, because of how easy they are to read,
because of how predictable they are, how rewarding they are, how easily

they respond to even awkward soothing. So that the mother has her own
wounds healed by the baby. And I think there are children who teiTi'

peramentally are much more fragile, for whom smaller deviations in sen'

sitivity or predictability have a bigger cost.”^^ Bowlby summarized the

attachment position this way:

An easy newborn may assist an uncertain mother to develop a

favorable pattern of care. Conversely, a difficult unpredictable

newborn may tip the balance the other way. Yet all the evi'

dence shows that a potentially easy baby is still likely to

develop unfavorably if given unfavorable care and also, more

fortunately, that with only few exceptions a potentially diffi'

cult baby can develop favorably if given sensitive care. The

capacity of a sensitive mother to adapt to even a difficult
unpredictable baby and thereby enable him to develop favor'

ably is perhaps the most heartening of all recent findings in

this field.
298 Nature'Nurture Erupts Anew

Although Sroufe has staunchly resisted the notion that a child can be
born inherently difficult and cites compelling evidence that parental
input or lack of it — even in the very first days — plays a significant role
in at least some of the difficult infant profiles, he, too, has made an
accommodation:

If Karen is ambivalent about being a mother Ihe wrote with

colleagues in a recent developmental text] and Meryl is

cranky as a newborn, interaction between the two will tend to

intensify these traits. As long as Karen’s situation does not
improve, Meryl is apt to become a truly difficult child, even
though her initial irritability could have been only temporary
in different circumstances.^^

This interactionist view, now given at least lip service by everyone in

the field, albeit with different emphases, has been supported by various
studies among humans and lower primates. In one study, neither the
infant’s susceptibility to stress nor the mother’s desire for control alone
predicted quality of attachment, but taken together they did.^"^ Steven
Suomi, now at the Institute for Child Health and Human Development
in Washington, found that heredity seems to determine whether a rhesus
monkey will he socially forward or retiring and that monkeys with exces'
sive timidity — similar to the fearful profile observed in some human
children — will often tend to have problems later in relationships. But
these are only tendencies, he says. A nurturant mother can often erase
temperamental deficits. The exceptionally nurturant mother, Suomi has
found, takes the time to teach her fretful child coping styles, such as how
to get help from others when he becomes afraid. As a result, Suomi says,
the fretful hahy “later in life becomes a very effective and high-ranking
member of its social group.”
The second prong of the behavior-genetic attack struck more force-
fully to the core of the attachment idea, not necessarily by denying that
something which might reasonably he called “attachment” existed or
that it developed in better or worse ways, hut by dismissing virtually all
of the research done in name, beginning with Ainsworth’s Baltimore
its

study. It asserted that what Ainsworth measured in the Strange Situa-

tion was really temperament — or that whatever she measured was so
confounded by temperament as to make her assessments meaningless.
 —
Renaissance of Biological Determinism 299

“Ainsworth’s Strange Situation could be appropriately rated,” Chess and

Thomas wrote, “under the temperamental categories of approach/
withdrawal, adaptability, quality of mood, and intensity.” Subsequent
descriptions of older children done by Sroufe and his students, they said,

“could also he rated under the same four temperamental categories, plus
•
... ”36
persistence.

The idea that inherited temperament might be determining attach-

ment outcomes was preposterous to Sroufe. If this were so, why is it that
some infants come out with different patterns of attachment with each
parent — that, indeed, only a small correlation has ever been found be-
tween attachment to mother and attachment to father? If this were so,
how could we explain the fact that quality of attachment can change?
And, finally, if this were so, why would depressed and unresponsive
mothers almost always have anxious babies? In one study, Sroufe evalu-
ated the newborns of such mothers. “We could show — really clearly

deterioration in those kids,” he said. “They looked pretty good at three

months, but at six months they didn’t look so good. About half of them
were anxiously attached at twelve months and all of them at eighteen

months. We’re talking a serious downhill slide. What are you going to
?”^^
say? The baby was born a downhill slider
Sroufe does not hold much stock in the whole idea of inherited tem-
perament, at least not in terms of the qualities in children that interest
him, and he does not believe that temperament and quality of attach-
ment are worth discussing in the same breath: “I believe that tempera-
ment is a useful concept, that is, that infants and children differ in terms

of preferred tempo and more meaningful individual dif-

so forth [but]

ferences are best captured by other concepts.”^® And he is just as good at

dismissing temperament research as the behavior geneticists have been
at dismissing his. He (among others) argues that Kagan’s data on the
continuity of temperamental behavior at different ages are vastly inflated
by studying a small group at the temperamental extreme and that for the
majority of the population such continuities are insignificant.^^ He is
particularly scornful of the huge amount of temperament work based on

parental reports, noting that a parent’s concept of what a baby is like has

been shown to have little in common with the appraisals of neutral

observers. In one study, expectant parents were asked — before the fetus

had begun moving in the womb — what they thought their baby would

be like. The parents’ imaginary descriptions of their unborn infants

300 Nacure-Nunure Erupts Anew

tended to be similar to how they described their actual baby many

months later, suggesting, at the very least, that their expectations played
a role in their perceptions."^^

The temperament^attachment debates soon focused on the avoidant

babies. Kagan, among others, insisted that many children classified as

avoidant appeared to be indifferent to their mother’s comings and goings

in the Strange Situation not because they’ve given up getting anything
from her but because they are truly indifferent. Unlike their counterparts
who have been labeled secure or ambivalent (with equal injustice), the
sO'Called avoidant children are simply constitutionally less fearful.
Sroufe countered such arguments by pointing out that the children
who were classified avoidant only looked calm and independent in the
Strange Situation; at home they were as clingy and tantrumy as those
who were classified ambivalent. And, of course, he had his poignant
observations in preschool and later that indicated ongoing dependent
behaviors, of an often pathetic character, among these supposedly fear^
less and independent kids. Even in the Strange Situation, there was evi'
dence that avoidant calm was only superficial and that avoidant children
were masking more powerful underlying feelings. One Minnesota study,
confirmed later by the Grossmann team, showed that the heart rates of
avoidant one^year-olds go way up when the mother leaves the room and
way up again when she returns, even as their behavior remains calm,
data that seem to suggest that the avoidant child is indeed in a state of
angry emotional cutoff from the parent."^^ Kagan would have none of
this, arguing that heart rate acceleration in such situations may also be a
function of temperament and that his own data suggest just that."^^

Sroufe countered that if the elevated heart rates of the avoidant child
illustrate a temperamental style, it is just the opposite of the bold, fearless
child Kagan perceives him to be. And indeed, it would seem that Kagan
is trying to have it both ways in this case, for the type of child to whom
Kagan is referring here, whose heart rate tends to be elevated in such cir-

cumstances, is the shy, inhibited child, not one that could be called bold
or precociously independent.
The ambivalent child has been, if possible, an even greater source of
contention because it seems to display the fussy and irritable style of the

one temperamental group that Kagan has spent most of his time studying
and that has, on the whole, been more firmly documented than any
other. Behavior geneticists took note when attachment workers in Japan
found that the majority of Japanese bahies were classified ambivalent —
 Renaissance of Biological Determinism 301

an extraordinary finding when one considers that ambivalence turns up

in only one tenth of the middle'class American babies. Kagan immedi'
ately charged — as he had with the north German sample — that it would
be absurd to assume that the child-rearing practices of an entire nation
were at fault. Rather he took this as another sign of temperament at

work, in this case picking up a racially based constitutional inclination

to be more easily distressed. In an unusually pointed response, which sug-
gests the degree to which Kagan’s arguments and the way he expresses
them can get under the skins of his opponents. Jay Belsky, a rising young
attachment star at the Pennsylvania State University, wrote: “What
Kagan has not acknowledged, although making this inference that
sounds sensible, is that research on constitutional differences between
Japanese-Oriental and North American-Caucasian infants would” lead
to just the opposite conclusion, since “Japanese newborns are actually

less susceptible to distress.

It did not take long before attachment workers were setting up their
own studies to try to settle this question directly. In one, Belsky found

that certain aspects of the mother’s personality, assessed before her baby
was born, were a more reliable predictor of the child’s later attachment
status than the baby’s temperament. Mothers who showed greater empa-
thy and concern for others, mothers who were more mature and stable,

were more likely to have securely attached children. Indeed Belsky found
that none of the direct assessments of infant temperament, including
fussiness and alertness, measured at various times during the first year,

had any predictive value in terms of Strange Situation classification at

twelve months. But mother’s appraisals of their infants’ temperaments

were predictive. Although it is impossible to know for sure what the
mothers’ appraisals were really all about — the mother, the infant, the
relationship between the two — what was particularly interesting in
Belsky ’s study is that the mothers’ appraisals often changed over the
course of the year. The mothers of children who would later be classified
as secure did not see their babies as any less challenging or more attrac-
tive than the mothers of children who would later be classified as anx-
ious — indeed, they rated them as somewhat more challenging and less

*Many attachment researchers now assume that the early Japanese studies of attachment were
flawed because Japanese children, who were not accustomed to being left alone tor even a minute,
were — contrary to Ainsworth’s instructions — allowed to get overstressed and cry too long in the

separation episodes of the Strange Situation. Thus, they were inaccurately classified ambivalent. In
fact, Japanese babies tend to be raised with an extraordinarily high level of attunement and atten-
tiveness on the part of their mothers.
302 Nature'Nurture Erupts Anew

attractive. But over the course of the year their assessments changed, and
they saw their babies becoming more adaptable and less difficult. Just the

opposite was true of mothers whose children were destined to be classi-

fied as anxiously attached. While many of them rated their babies higher

in the beginning, they tended to see their babies becoming more difficult

as the year wore on.'^'^

Belsky further found that although laboratory assessments of the

infant’s temperament could not predict whether the child would be
securely or anxiously attached, they did seem to predict the style of

secure or anxious attachment. Fussy and irritable infants who later

became anxiously attached were more likely to be ambivalent than

avoidant. If securely attached, the fussy child was more likely to fall into

the more clingy and excitable subgroup. Belsky felt he had squared this

circle with this study, creating a rapprochement between behavior

geneticism and environmentalism, without giving up any of attachment’s
tenets. His finding — essentially, that temperament affects the way secure
or anxious attachment is expressed — has become widely accepted in

attachment circles. (Strange when you consider how directly it seems to

contradict a fundamental attachment tenet — that babies become avoid-
ant or ambivalent based on the style of parental care.)
In Germany Klaus and Karin Grossmann have also attempted to
unravel this problem, assessing the children at birth by using state-of-the-
art temperament measures. “We did establish that some children make it

easier for mothers than others,” Karin says, with the result that positive
correlations did exist between certain temperament styles and secure or
anxious attachment. They found, however, that the mother’s sensitivity
as a caregiver was the more powerful predictor. The Grossmanns also
assessed the levels of cortisol, a stress-related hormone, in children going
through the Strange Situation. Their findings further supported the idea
that avoidant children v/ere distressed but not showing it."^^

Other studies, including one by former Sroufe student Brian Vaughn,

discounted any association between inborn temperament and attach-
ment classification. Vaughn found that early assessments of temperament
— whether determined by a widely used temperament questionnaire or
by measurements of hormonal secretions — could not predict whether
the baby would be securely or insecurely attached.

The overlay of inherited temperament and parental psychology, not to

mention family circumstances and cultural differences, all came into play
 Renaissance of Biological Determinism 303

with the report of a huge study by Dymphna van den Boom at the

University of Leiden in the Netherlands. For five years, van den Boom
had been a teacher of children with learning and behavior problems and
had been impressed with the fact that children who ended up in special

schools had usually exhibited problems when they were quite small and
that their mothers frequently complained of their having been difficult.
She became interested in parent^child interaction, began reading widely
in attachment theory, and became concerned that attachment theory
seemed indifferent to the effects of the child’s inborn characteristics, a

problem she was determined to remedy in her own doctoral thesis.

Van den Boom waded through 588 neonatal assessments in order to
find and study one hundred babies who were all highly irritable at birth,
displaying a low threshold for the expression of negative feeling, as
assessed on Berry Brazelton’s Neonatal Behavioral Assessment Scale.
Since critics of attachment theory had repeatedly alluded to studies
showing that children assessed as irritable at birth were more likely to be
classified anxious at one year, this was an important target population.
Van den Boom’s children were particularly at risk. Not only were they
inherently more trying than more placid and smiley babies, they were

bom into low'income families, where uneducated and pressured parents

might not have the time, the knowledge, or the emotional reserves
needed to guide their babies through to a solid adaptation.

Van den Boom now asked all the key questions: Is this select group of
irritable children destined to be classified anxiously attached at one year
regardless of how they are raised? If so, will they, in keeping with Kagan’s

thinking, belong to the ambivalent group, a manifestation of their fussy,

fretful style? Will the quality of attachment be at all affected by how sen-

sitively they are responded to? If so, by how much?

In order to answer these questions she divided the hundred pairs of

children and mothers into two groups of fifty each. One group of mothers
received three counseling sessions of about two hours each between the
child’s and ninth months, the purpose being to enhance the
sixth
mother’s sensitivity and effectiveness. The other mothers were left to
their own devices.

Because the irritable babies were less responsive to their mothers,

smiled and made pleasing sounds less frequently, van den Boom found
that the mothers tended to become discouraged and give up on them in

various ways. One mother, van den Boom recalls, “turned on the radio to
help her shut out the crying and fussing” as soon as the baby became
304 Nature-Nurture Erupts Anew

upset. Van den Boom educated the mother in the importance of soothing
a crying baby, and then worked with her to determine what soothing
strategy would work best. Like most mothers in the study, this mother
found her success at soothing her baby rewarding and that encouraged
her to keep at it. Another mother, apparently rattled by previous failures,

responded to her baby’s cries by trying to soothe it every which way she
could, using one strategy after another in quick succession. In short
order, both baby and mother were at wits’ end. Van den Boom helped the
mother determine which soothing strategy was most effective and
encouraged her to keep doing it until the baby calmed down. Other
mothers, exhausted from failed soothing efforts and fearful of disrupting a
quiet infant, had stopped interacting with their babies when they were
not crying. Van den Boom encouraged such mothers to play with the
child. But because the mothers of irritable babies turned out to be very
interfering, often by pushing one toy or another on a resistant baby, she
asked them to begin play without toys, a process that helped promote
attunement."^^
The outcome was stunning. In the families that received these simple
interventions, 68 percent of these fussy children were classified securely
attached at one year. In the control group, only 28 percent were classic

fied secure. There could be no doubt that quality of care was a key vari'

able in quality of attachment, but it was not the only one, and it was
influenced by other factors. Also, that forty^eight of the one hundred
turned out to be securely attached refuted any idea that it was their tern-

peraments that were being measured in the Strange Situation. They all

displayed the fussy, easily stressed, quick^tO'Cry style, but almost half of
them were classified secure. If this study stands and is replicated by oth'
ers, it would seem to have settled some major questions.
But while settling some questions, van den Boom immediately raises

others. Attachment theorists assume that in many cases the mother’s

deep'Seated psychology determines her behavior toward her baby and
thence the baby’s quality of attachment. If that is the case, how can six

hours of intervention be enough to insure security? It leaves one incredu'

lous. Will a secure relationship built on such a foundation be quickly
eroded as unaltered aspects of the mother’s psychology come into play in
later months and years? Or will a solid start help mother and child to
remain securely related thereafter? The fact that many mothers do better
with their second child than their first suggests that, at least for some
parents, just knowing the ropes makes a difference. (Other parents do
 Renaissance of Biological Determinism 305

better with a second child because they enact all their psychological
needs and hang-ups with their first.)

But Van den Boom’s evidence indicated that extreme temperament

also plays a role in anxious attachment. Her study confirmed what Stella

Chess would have expected: that irritable children did indeed seem to
place additional stresses on their parents (who, in turn, no doubt, be-
came more rejecting) thereby making it harder to establish a secure rela-
tionship, a fact suggested by the phenomenally low secure rate of the

nonintervention group.*
Van den Boom also found a surprising breakdown among the anx-
iously attached children: In both the experimental and the control
groups, the great majority (77 percent) of the anxious children were
avoidant, not ambivalent as Kagan, Chess, and others would have
predicted and as Belsky had found in his study. In the Strange Situation,
van den Boom’s irritable children looked nonchalant. Why this should
be so remains a mystery, as does so much in the interplay of biology and
experience. It may be that the mother’s style of impaired caregiving is in

most cases ultimately a more powerful influence than the baby’s tempera-
ment in determining whether the baby will be ambivalent or avoidant
and, further, that mothers of irritable babies tend to respond to them
differently based on the cultural setting. This is what van den Boom
believes.^

Genetic research has been an important corrective, not necessarily

because it has all the answers its proponents claim to have, but because it

forces us to reconsider simplistic environmentalist assumptions. Unfortu-

nately, as of this writing, the corrective seems to be turning into another

huge swing of the pendulum. At this moment behavior geneticists are
excited by their own successes and, with some important exceptions, not

*Ainsworth noted the tendency of some mothers to become more rejecting when dealing with a
also
fussy infant, although it seemed to her that in the cases she observed the fussiness began with the
mother’s frustrating the baby. “In all of these cases there seemed to be a vicious spiral in which the
baby’s fussy demands exasperated the mother, who then overtly or covertly rejected the baby, who in

turn responded to the rejection by anxiety and by increasing his demands. The mother’s and the
baby’s anxiety seemed to be in a spiraling interaction.” Ainsworth, 1967. See also Chapter 11.

^Van den Boom, 1988, 1993. Van den Boom notes that in Japan, where irritable babies usually end
up classified as anxious-ambivalent, mothers tend to encourage a high degree ot emotional depen-
dence in their small children. If a factor like irritability should become involved, the mother is more
likely to become inconsistent than totally cut off. In Germany, on the other hand, especially north-
ern Germany, where irritable babies often turn out to be anxious-avoidant, early independence is
highly valued. Thus the baby is more likely to face a total ignoring of his attachment-related signals.
306 Nature'Nurture Erupts Anew

yet ready to integrate an environmental view. Meanwhile, their work has

struck a responsive chord in the culture.
Biological explanations of who we are have a compelling quality.

They offer hard science rather than ambiguity; they circumvent ques'
tions of inner conflict or anxiety about selBworth; and they hold out the
promise of an efficient way to deal with problem people like alcoholics,

depressives, or children diagnosed with attention deficit disorder.

According to Dr. David Comings, a medical geneticist at a Los Angeles

hospital, interviewed for the PBS documentary Medicine at the Crossroads
in 1993, “One could argue that up to twenty percent of the population
has some type of genetic behavior problem.” He envisions a time when
one could send a behavior^problem kid “to the laboratory and find out if

he has any of these abnormal genes. (Then what?)

Biological explanations have become dominant in psychiatry where it

is commonly assumed that moderate depression, relationship difficulties,

neurotic anxiety, and other common emotional problems are based on
malfunctions of the brain and nervous system, even though little evi'

dence exists to support this view. Biologically based research into emo'
tional problems now gets the majority of government funds, while
research on psychological or sociological factors often suffers.^*^

The press is particularly susceptible to news of genetic breakthroughs

(which make an easily grasped newspaper or TV story), and it often gives
little attention to later withdrawals. As a result, the public is probably
much more aware of claims that a specific gene was found to cause
manic-depressive illness than the later retraction or that excessive drink-
ing is inherited despite studies that indicate otherwise.^® If the press
reports that some human quality, like homosexuality, optimism, crimi-
nality, or anxiousness has a “genetic component,” the public often
assumes that this quality is when in fact, for most people who
inborn,
display it, the hereditary component may he very small.
In developmental psychology, biological determinism reached an
apotheosis of sorts in April 1991, when Sandra Scarr, an old bridge-play-
ing friend and intellectual antagonist of Ainsworth’s from the University
of Virginia, took the podium to deliver her presidential address to the

Society for Research in Child Development. A behavior geneticist who

had become interested in inherited differences as an undergraduate
(when “1 was told there were none”^*), Scarr’s central theme was that
“being reared in one family, rather than another, makes few if any differ-

ences in children’s personality and intellectual development.”

 Renaissance of Biological Determinism 307

Smiling, cheerful infants who evoke positive social interaC'

tions from parents and other adults seem likely to form posi'
tive impressions of the social world and its attractions. Infants

who are fussy, irritable, and who receive negative or neutral

interactions with their caregivers and others would seem less

likely to form the impression that social interactions are a

wonderful source of reinforcement.^^

Citing recent studies, Scarr went on to emphasize that at every stage

of development our genetic makeup more or less determines how people
respond to us. Two toddlers from the same family get different responses

from their parents because of their different inborn personalities. Chih

dren raised in poverty will receive more attention and encouragement
from their teachers if they are spunky and intelligent. “Adults who are

considered physically attractive by others are more likely to be chosen as

dates and mates than others considered less attractive. Attractive adults
are also more likely to be hired for jobs and more likely to be promoted
on those jobs than their less physically attractive peers.” Thus, the envi'

ronment does not form us, we form it.^^

This view does acknowledge that there are powerful influences at

work within the family, but only in one direction: from child to parent.
In the family Scarr portrayed, a parent’s psychological makeup or his

conscious efforts to adjust his caregiving to a particular baby’s needs have

little impact on the child’s emotional development.
Such a major playing down of environmental influence has caused
uneasiness in the field, because of the message it sometimes gives to the
general public. Ever since Arthur Jensen used a genetic argument to
oppose compensatory educational programs for blacks as a waste of pub'
lie funds, people have worried that the contributions of genetic behav'
iorists might be used to suggest that whole groups are poor, downtrodden,
burdened with high crime rates, drug abuse, or family violence, because
they’ve found their rightful genetic niche; that people are what they are
and cannot be changed. Scarr does not endorse this view as applied to

ethnic or racial groups, and her research with black adopted children has
helped refute it. But for individuals within a group, she sees heredity as

the primary cause for individual differences, including social class status.
In one of their papers, Bouchard’s team quotes, with evident triumph,
a pre'twin'Study comment (1981) by psychologist Walter Mischel, a

prominent behaviorist: “Genes and glands are obviously important, but

308 Nature -Nurture Erupts Anew

social learning also has a dramatic role. Imagine the enormous difference
that would be found in the personalities of twins with identical genetic
endowments if they were raised apart by different families. The impli'

cation of quoting Mischel this way was that their work had so obviously
proved him wrong. But, in fact, Mischel was right: there are substantial

differences. On the average, according to Bouchard’s data, 50 percent of

the personality traits displayed by identical twins reared apart are diver-
gent.^^ But because the similarities have been so arresting — and so sup-

portive of the inborn temperament theory — not much has been said
about the differences or how they come about. Even less is said about the
fact that the similarities may have been promoted by similar or parallel

environmental experiences. Less still about evidence that temperament

itself may not be immutable and may be altered by experience. “They
make generalizations about the environment,” Cornell’s Urie Bronfen-
brenner protested at one point, “but don’t analyze the data to determine

what influence the environment had.”^®

For instance, many of the similarities in twins, to the extent that they
do have an hereditary component, may represent the way a person with a
certain genetic makeup develops when he has encountered a certain type

of environment while growing up. What’s more, it’s not always the sur-
face trait but the way it is manifested that is most critical in a person’s

life. What may be more important to the twin Jims and their families
than the fact that they are both highly sociable men who leave romantic
notes around for their wives to find is that they have both managed to
develop a warm and secure connection with their wives. Where does this
capacity come from? Attachment research suggests it evolves from secure
attachment early in life — feeling loved, feeling that one’s love is valued,
expecting good things to come of close relationships, feeling good about
oneself, feeling that one has the power to have a positive effect on
another person. Is it not possible that both twins had such an early envi-
ronment? A child who is endowed with great linguistic potential will
barely learn to speak if not regularly spoken to by a certain age. That
both Jims have a tendency to he demonstrably affectionate and that both
men have the same unusual way of displaying it may owe a great deal to
their temperamental makeups. But the tact that they are able to get to

the point with another person where those traits can manifest them-
selves prohahly reflects similar secure attachment histories.

Also, although the two men may achieve similar scores on certain per-
sonality inventories, we don’t really know very much about how they
 Renaissance of Biological Determinism 309

experience themselves or what they are like to be with in a relationship,

since such in-depth analysis is beyond the capacity of most studies like

this. We do know that both Jims leave affectionate notes for their wives
to read; and yet these notes may have a different effect on the two
women. One may feel that her husband is playful and secure in his love.
The other may feel that he is too much like a puppy, maybe even a
whipped dog at times, constantly wagging for affection. The notes, which
she sees as emblematic of a nagging insecurity, could thus become a secret
source of embarrassment or disgust for her, which she wouldn’t mention
in a superficial interview about her husband’s qualities and habits.

Psychopathology raises similar questions. Some of the earliest of the

twin studies found that when identical twins are raised separately and
one is later found to have become schizophrenic, there is a 50'50 chance
that the other will also be found to have become schizophrenic.^^ That is

an extraordinarily powerful argument for genetic involvement, because

only one person in a hundred normally contracts this disease. Such find'
ings have largely discredited the once popular notion of the “schizo'
phrenogenic mother” who drives her child crazy with interpersonal

torments, such as the famous doubk'bind form of communication.*

Indeed, it has so discredited that theory that many experts in the field
are now scornful of the idea that parental influence can make any contri'

bution to this condition.^^ But the fact that another 50 percent of those
siblings — with an identical genetic makeup — do not become schizO'
phrenic is an equally powerful argument for the impact of the environ'
ment. Finnish psychologist Pekka Tienari attempted to assess the genetic

and environmental components by comparing adopted children whose

biological mothers were schizophrenic with those who came from
healthy families. The first group became disturbed at a much higher rate,
suggesting a genetic component, but severe disturbance developed only
among those children who were living in homes that were evaluated as

emotionally dysfunctional.^^
Consider again the case of the hand'washing twins. Many will take as

the moral of their story that the twins’ blaming of their mothers is wholly
mistaken, that, in reality, they were destined to be compulsive neurotics
from birth. This is clearly what Neubauer believes: “Neither the mother

*For example, the parent invites the child to openly express a courageous opinion and then dispar-
ages it. Or the parent says, “1 love you, darling,” with hate in his eyes. The child feels entrapped hy
thisform of communication, unable to react in any direction. How can you hate a mother who loves

you? How can you love a mother with those eyes?

310 Nature'Nurture Erupts Anew

who was compulsive nor the one who was slovenly made any difference

in the eventual behavior of their sons.”^^ But even behavior geneticists

will generally agree that the environment can play a critical role in

bringing inborn predispositions to flower; so what else can we say about

the origins of the twins’ condition?

The most economical explanation for their similarity is certainly
genetic. Their genes may have dictated both how much stress it would
take to push them over the edge, as well as what symptoms they would
display once they got there. But what was the stress? The stress chat the
first twin met up with may well have included an adoptive mother who

coincidentally had the same symptoms to which he was prone in full

flower. How might an obsessive^compulsive parent, who insists that

every knickknack be kept in its inviolable place, affect a child? In all

likelihood, she cannot tolerate the child’s natural messiness, she reacts

with disgust when his hands or face are dirty, she makes changing a dia-
per an ugly necessity rather than a playful and sensual exchange, with
loving pats, nuzzles, and kisses. The child’s cheerful smiles and other
efforts to engage her get cut off because she is easily distracted by a dirty
sleeve, a sock half off, or sticky hands. Although she may love her baby
very much, her behavior has a rejecting quality to it. And it continues
that way, into toilet training, room cleaning, playtime, and dinner man-
ners. Eventually, the young child feels that he himself is dirty and shame-
ful and must constantly be hiding, disguising, and cleaning away the
evidence of who he is.

Avoidant attachment would seem a plausible result. We’ve seen that

avoidant children often seek something to focus on in order to keep their
attention away from attachment concerns. They are able to stay in prox-

imity to their mother if they are concentrating on a toy or a coloring

book. Extreme avoidance would thus seem to provide a natural path-
way toward obsessive preoccupations. It the child has an obsessive-
compulsive predisposition, the stress of a relationship with an obsessive
mother might push thus him in that direction, causing him to take on his

mother’s concerns with a vengeance.

Now let’s consider his brother, raised in an antithetical environment.
A mother who is an “absolute slob” is not necessarily rejecting. A messy
mother can be loving and consistently responsive, and her messiness
need not get in the way of secure attachment. But something in the sec-

ond twin’s experience caused his genetic weakness to give way. It could
have been that he falls into the small minority of the population who
 Renaissance of Biological Determinism 311

will become psychologically disturbed in almost any environment. It

could have been a cruel and authoritarian father or an excessively rigid

or rejecting nanny. Or the young man could have been right in assuming
that in some way his slovenly mother played a part.

A slob may not only be slobby about material things. She may be as

neglectful and forgetful of the baby as she is of the dishes, the clothes,
and everything else. She may not be able to provide the baby with the
feeling that he is contained in a secure and welhmanaged environment
— Winnicott’s “holding” environment — which gives him a sense of

safety and regularity and which enables him to build his own steady

rhythms. Again, she may love her child very much, but her behavior is
the essence of neglect. Not feeling securely held and contained, the
child’s own impulses, feelings, and bodily functions may feel overwhelm-

ing and out of control, and at a certain point he may find a “solution” to
this problem in various forms of rigidity, especially if he is prewired to
respond to unbearable stress in this fashion.

Messiness may also grow out of an uncontrolled emotionality. In such

a person, every feeling may run to excess in the way it is expressed and
yet little is felt profoundly and even less is understood. She is scattered.

She does not tune in well to those around her, being perpetually preoc-
cupied with and perhaps fascinated by herself. Not being tuned in to or
given a stage for his feelings, the baby’s feeling side may never have a

chance to develop. Indeed, he may rigidly contain his feelings in an

effort not to be overrun by hers; he erects, in effect, an inner wall against
her intrusions. He forms an avoidant attachment to her in part to hold
on to his core integrity, to protect his vulnerable self from her impinge-
ments. The wall he builds takes the form of the rigid obsessiveness
toward which he is genetically inclined.

None of this is to say that the hand-washing twins were not headed
for some form of compulsive behavior regardless of the environment
they came from or that it might not have taken extraordinary parental
efforts to prevent it. But parental input should not he ignored. It one is

genetically vulnerable to phobias, obsessions, or other disorders, parental

input and the attachment style that follows may in many cases either

dampen those proclivities or be the fuel that keeps a pathological distur-

bance burning. On the whole, one’s first relationships would seem to

remain a primary force in all those issues — whether you trust others or

not, whether you anticipate love or rejection, whether you feel good
about yourself as a person — to which attachment theory has drawn
312 Nature-Nurture Erupts Anew

attention. Nothing in genetic research suggests otherwise.

New investigations, meanwhile, indicating that brain biology and

even developing brain structure are sensitive to the earliest environment
tal inputs, may radically alter the terms of this debate. For there is grow^
ing evidence that very early attachment experiences influence some of
those temperamental traits that were previously seen as genetic. We
now know that the brain is rapidly organizing itself in the first months of
life, and the latest research suggests that it does so in intimate concert
with the responses it receives from the caretaking milieu. The baby’s
ability to regulate itself, especially in all those areas related to emotion,
depends on parental attunement and empathy; and if the mother fails to

attune to the baby emotionally, the baby’s brain may exhibit lasting
physiological deficits. These may show up early in the chronic dysreg^

ulation of a “difficult child” and later in various personality deficits. All

of this, obviously, raises major questions about what we have assumed to
he inborn.
This is not to say that genetics are irrelevant. Far from it. And, attach-
ment workers have been as guilty of ignoring (or giving lip service to)

genetic influence as behavior geneticists have of disregarding the envi-

ronment. And they certainly have done little to understand the minority
of children who don’t fall into the expected attachment category given
the quality of care they received, or who don’t show the expected behav-
ior in later years to match the attachment group they belong to. Rather,
there’s been a too convient tendency to dismiss such anomalies as result-

ing from measurement error. (Ainsworth would eventually come to

believe that she had been too rigid in rejecting the temperament view.)
For these reasons, despite the frequent rancor and misunderstanding, the
struggle between the two developmental camps seems to have been ben-
eficial to both. In recent years, some behavior geneticists, like Robert
Plomin, have become more interested in the differences between people
with similar genetic makeups and the role the environment may play in
their creation. Stella Chess is undertaking a major re-analysis of her data
to determine the ways in which temperament itself, not just adaptation,
can change. Meanwhile, the hard-core environmentalism that has often
characterized attachment thinking has softened, as it has become more
apparent that we bring certain qualities with us from birth and that those
qualities influence how we respond to the world and how the world
responds to us. Exactly what those qualities are, though, and how mal-
leable they are, still demands considerable investigation.
 22
A Rage in the Nurserv:
The Infant Day-Care Wars

By the time Sandra Scarr addressed the Society for Research in Child
Development in 1991, the blaming and defending of mothers had
become a central subtext of the naturemurture debate. In the 1970s and
1980s mothers were beginning to feel extraordinary pressures to provide
optimal environments for their children. Like the mothers Stella Chess
was concerned about several decades earlier, these mothers faced persis'

tent and contradictory admonitions about how to raise healthy children.

Only now, the desiderata had multiplied, and a new obsession had arisen
over intellectually stimulating books, gadgets, and exercises for infants
and preschoolers, the absence of which might condemn the child to a

lower IQ and diminished opportunities throughout life.

The mothers of the seventies and eighties were also likely to have
jobs. Vast numbers of them were returning to work within the first

months or weeks of their baby’s birth, in a social transformation ot revo'

lutionary proportions.^ (By 1992, almost 54 percent of mothers with
children under one were working outside the home, as compared to
about 25 percent in 1970. As of 1990, according to one survey, over 40
percent of American babies receive their primary care from someone
314 Nature-Nurture Erupts Anew

other than a parent — 65 percent of them by a relative or neighborhood

mother who takes in kids, 22 percent in commercial day-care centers,
and 13 percent by nannies.) The new working mothers were particularly
sensitive to how their absence might affect their babies, they were suS'

ceptible to guilt, and there was no lack of voices telling them what to do.

Indeed, perhaps no debate in the history of the twentieth century’s

drawn-out conflicts on the proper ways to raise children generated more
heat than that over working mothers and day care.* Once again, attach-

ment theory was in the thick of it.

In her 1984 book on the subject. Mother CarelOther Care, Scarr, who
had herself raised four children while working, attempted to debunk the
idea that a mother was uniquely important to a child and that any dis-

ruption of the bond between them might be ruinous to its future well-
being. She objected to the notion that a baby had a special need for its

biological mother; that the best thing for a baby was an exclusive rela-

tionship with its mother, undiluted by other attachments; that mothers

are somehow better equipped than fathers or other caregivers to tune in
to the infant and its needs; that the ideal mother finds “passionate per-
sonal fulfillment in fulfilling the needs of her child.
Much of this Scarr intended as an attack on Bowlby and Ainsworth,
neither of whom believed any of these things (with the exception that
Bowlby did see women as naturally better than men at child care^). But
such ideas were in the air, and they arose with ever more ferocity as the

debate over day care heated up.

Why should working mothers feel guilty or deprived? [Scarr

wrote.) Working mothers spend tour to six hours a day with
their babies, rather than all eight to twelve of the baby’s wak-
ing hours. Are they missing significant developmental phases?
Should they “suffer” from being away part-time from maternal
responsibilities, if they also have paying jobs that are impor-
tant to them and to the rest of the family? Why the guilt trip?
Working mothers spend as much time as nonemployed moth-
ers in direct interaction with their habies. What are they miss-
ing, except redundancy?"^

*The new preferred term among many experts is “child care.” The two terms are often used inter-
changeably now.
 A Rage in the Nursery 3J5

Developmental research, including attachment research, contributed

to the defensiveness that many mothers felt. It was not just that the
studies, in keeping with population norms, spoke continuously of the
impact of the mother and failed to adequately emphasize that the father
could also he the primary caregiver, but, perhaps more important, they
neglected to speak of the huge impact of the father even if he was not the
primary caregiver. It sometimes seemed as if there was only one parent in

the twO'parent home, only one whose influence counted, and that the
other one, the father, did not really have to concern himself much with
the parenting role or the lasting effect he was having on his children.
This problem was aggravated by a conservative political movement
that wanted women to stay at home with their babies but rarely consid'

ered what it might mean for them to do What do you do if you are a
that.

single mother (as over 22 percent of mothers now are)? What do you do

if, like a woman I saw in therapy, you bought a house just prior to an
unexpected pregnancy, and two salaries are essential to meeting the
mortgage? What do you do if you are a lawyer or a WalhStreet analyst or
a professor on a tenure track? Will you be able to get your old job back?
Can you opt out of all the competitions and let your peers walk past you?
What does a psychotherapist or internist do with all his or her patients?
Ask them to come back in a year?

Feeling burdened by this insensitivity and attacked by those who

believed that their absence from the home undermined the family and

scarred their children, working mothers felt obliged to justify themselves.

Some legitimately cited studies that demonstrated the benefits of day

care or the fact that, for many women, work is an important ingredient
of their selhesteem and a protective factor against depression.^ As Scarr

wrote, “some women at home fulhtime are lonely, depressed, and not
functioning well” and “high-quality day care settings have in tact been
shown to compensate for poor family environment.”^ Other day-care
advocates went further, denouncing housewifery and full-time mother-
hood as indentured servitude. Mothers who stayed home telt humiliated
by such attacks. And they felt left out by a swiftly changing culture, one

in which virtually every article in women’s magazines on family matters

was now addressed exclusively to working women. Some argued that
child care was not a sacrifice hut a pleasure, and, tor all its hassles, a peak
experience they wouldn’t miss. Others joined the attacks on working
mothers with appeals to biology and tradition, sometimes showing insen-
sitivity to those mothers, often single and without extended-family
316 Nature-Nurture Erupts Anew

support, who had no choice but to work. Soon, it seemed, both sides
were debating more and more with the extreme wing of the opposition
and seeing themselves as oppressed and abused.
Throughout the 1970s psychologists attempted to compare children in

day care with children reared entirely at home, but there were so many
confounding variables that it was hard to draw conclusions. The findings
were at times contradictory, with some studies showing that day-care
children were more positive and less tense in social interactions and had
better relationships with peers, while others showed that they were less

cooperative and more aggressive. Jerome Kagan reported no difference in

separation anxiety between children raised at home and those who had
day care. What’s more, the day-care children in his study played more
easily and with less apprehension in a strange environment than those
raised at home (which, opponents argued, may have been simply because
they were more used to doing so). The mothers of day-care children
found them more patient and less shy with strange adults.^ Research by
Mary Blehar, a student of Ainsworth’s, indicated that day-care children
cried more and showed more anxious attachment, but her study had
some methodological shortcomings and was not confirmed by later
research. Other attachment researchers, like Jay Belsky and Mary Main,
had found no problem with day care, and, indeed. Main had found that
whatever problems children first display diminish the longer they are in
day care. There was one important caveat: For children under three
there was some concern about anxious attachment; but this had not
been definitively shown.®
In almost every case it was difficult to make generalizations or be sure
of what a finding really meant. Were the day-care children in some stud'
ies more secure or did they simply look that way because they’d had the
opportunity to develop their social skills? Were personality differences in
the children a reflection of their differences in daytime experience or a
reflection of the fact that mothers who tended to put their babies in day
care were different from those who did not? If children with extensive
day'Care histories displayed a higher rate of anxious attachment in some
studies was it because, as Sroufe suggested,^ repeated early separations
made them more vulnerable to inadequate or inconsistent parental care
at home? Or was it a reflection of the fact the children in day care were

more likely to come from financially stressed and unstable homes?

Then too there were such vast differences in the nature and quality of
the child care that using a single term to describe all the variations
 A Rage in the Nursery 317

seemed pointless. There was part-time and full-time care. There was day
care in a family setting, with just a few other toddlers, and day care in a
large facility. There were situations in which the child had a specific

caregiver assigned to him, which provided him with many of the benefits

of an attachment figure (assuming the person stayed), such as comfort

and security when he was stressed, and those — far more numerous as it

turned out — where he had no special person, where he felt lost and
uncared for, and where he was at the mercy of bigger, more aggressive

children whom he would eventually succumb to or emulate.^® There

were facilities with high staff turnover and facilities where the staff was
stable. The ratio of caretakers to children ranged from a handful, some-
times a large handful, of toddlers to a single adult — who had to spend so

much time managing, controlling, and disciplining them that she had
little time for loving or stimulating interaction — to the caretaker who
handled as few as two or three babies at a time. (There was also the
nanny, sitter, or housekeeper who came into the home where the child

lived, who, in effect, served as a substitute mother, and who was often

found to be more stimulating and responsive and less controlling than

the mother herself. But this option, which Scarr herself had used,^^ was
not normally considered under the heading of day care, which had essen-
tially come to mean “group care.”) And then, of course, there were chil-

dren entering care at different ages. None of the studies attempted to

assess the mother’s emotional state, her feelings about leaving the child
each day, or the quality of her home life, all of which would naturally
have an impact on the child and on her relationship with him.^^
To complicate matters further, numerous studies were flawed. For one
thing, many of them used the Strange Situation, often in modified form,

to look for evidence of insecurity in children two or older. But, as Brian

Vaughn noted, the Strange Situation had been designed for one-year-olds,
the idea being to put them through a series of stressful separations to see

how their attachment system manifested itself. There was no evidence

that the procedure had any validity for assessing quality of attachment in

children over twenty months old, who were presumably much better able

to withstand such short separations.^^ Other researchers, meanwhile, had

modified the Strange Situation in ways that may have destroyed its use-

fulness; some had (incredibly) used it in the child’s home where short sep-

arations from mother are common and thus unlikely to stress the child;

and still others had resorted to the meaningless task of counting behaviors
rather than using Ainsworth’s carefully constructed scales.
318 Nature 'Nurture Erupts Anew

The only two studies that used the Strange Situation as it was
intended did find more anxious attachment in those children who had
begun day care before their first hirthdaysd'^ Indeed, in Vaughn’s 1980
study, 47 percent of these children were anxiously attached, and all of
them were avoidant, the form of insecure attachment associated with an
inaccessible or unavailable mother.
Nevertheless, when the two major surveys of the day-care literature
were written — in 1978 by Jay Belsky and Larry Steinberg, and in 1981
by Michael Rutter — the authors seemed justified in concluding that
high-quality day care did not compromise the child’s bonds with the par-
ents or appear to pose any serious risks. Thenceforth, Belsky in the
United States and Rutter in Great Britain were widely cited as having
given day care the green light. Both authors had raised important cav-
eats, however, which were frequently ignored. Rutter, for instance, wrote
that while the risks of day care had been “greatly exaggerated in the
past,” it did raise some questions. There was some concern, at least for a

minority of children, that day care for the very young child could lead to
emotional disturbance. There was also some indication that day care
caused behavioral change, sometimes for the better, hut not always. Both
he and Belsky were careful to point out that whatever we knew about
day care was the result of relatively crude research day care versus —
home care — that paid no attention to the many variables in quality and
quantity of care. Finally, and perhaps most important, virtually all of the
studies had been done in high-quality, university-based centers that had
little relationship to the day care that was available to the average per-
son, which was almost universally of a much lower, even dismal, quality,
with lower staff ratios, high staff turnover, poor training, and more
cramped conditions. The research had, in effect, been conducted in an
ivory tower. It was unrelated to the choices facing most parents.
While the evidence was by no means certain, opinions were. Many
were opposed to day care in principle and assumed it had lasting deleteri-
ous effects. Bowlby was especially wary of separating mother and baby for
long or repeated stretches before the age of three, and wished to see a
campaign, equivalent to the surgeon general’s crusade against smoking,
to convince parents that day care was had for their babies. His concerns
were exaggerated as they were picked up and spread by others of a more
dogmatic frame of mind.
Meanwhile partisans on each side were carefully selecting quotes and
partial conclusions from the research to holster their own arguments.
 A

A Rage in the Nursery 319

such that academic work was being picked over by cultural advocates
and journalists with an angle for bits and pieces to use as ammunition.
When I published a lengthy article in The Atlantic on attachment theory
in 1990 which assessed some of the debates on day care without forming
,

any firm conclusions, the founding editor of the magazine Working

Mother wrote a letter to the editor of The Atlantic denouncing Bowlby,
Ainsworth, and the magazine for claiming that mothers shouldn’t work

and used an out'ohcontext quote from Rutter as proof that day care had

been given a clean bill of health. similar distortion was made by the
other side, who, in Belsky’s words, take any evidence that certain kinds
of child care may be deleterious for certain children, and proclaim. Aha,
this demonstrates that only mothers can care for babies, that day care is

bad and we shouldn’t have working parents.”^^

In Britain, the great sinner from the proAnfant'day^care, feminist per-
spective was, of course, John Bowlby, who remained obdurate on the

issue and seemed unconcerned about whom he offended. “This whole

business of mothers going to work,” he said, “it’s so bitterly controversial,

but I do not think it’s a good idea. I mean women go out to work and

make some fiddly little bit of gadgetry which has no particular social

and children are looked after in indifferent day nurseries. It’s very
value,
difficult to get people to look after other people’s children. Looking after
your own children is hard work. But you get some rewards in that.
Looking after other people’s children is very hard work, and you don’t get

many rewards for it. I think that the role of parents has been grossly
undervalued, crassly undervalued. All the emphasis has been put on so-
called economic prosperity.”

Bowlby believed that “an affectionate, stable person who will act as a

substitute mother over a period of years” — the nanny — is an appropriate

solution for women who preferred to work or who could not afford to

interrupt their careers. But finding someone to fill that role isno longer
she expected to stay until the child is old enough
so easy. Especially if is

to leave home, which Bowlby saw as important — because otherwise the

child would be struck with a traumatic loss. “You get a nice girl who stays

a year or two and then moves on; this is very, very had.”

Even many of his own followers would part company with Bowlby
primary
over the contention that the full-time nanny becomes the
while
attachment figure, supplanting the parents in the child’s affections;
the idea that she must stay till the child is grown and leaves home strikes

others as preposterous. As Alicia Lieberman says. Normal parents cher-

 320 Nature'Nurture Erupts Anew

ish their babies more than anybody else can cherish them, and this pas-
sionate commitment is recognized and responded to by the baby. [It] is

not lost just because they may spend the day apart.”* Virtually every
working parent knows the truth of this. But that the nanny is still an
important figure to the child, in many cases an attachment figure, and
that her early loss or abrupt replacement can sometimes be traumatic
would find little argument.
“You see. I’m all in favor of mothers’ having assistance,” Bowlby told
me, “the more assistance the mother gets, the better. My wife was the
chief carer for our four children, and we always had a girl, a university
student, and we didn’t expect that girl to stay. That worked out well,
because they were just assistants, they were never playing a major part in
a child’s life.”

If a child was not going to have a long-term nanny and if the mother
was going to remain the main caregiver, Bowlby believed she should not
work at all during the first few years. For mothers of older children,
Bowlby was more flexible than many people realized. “A child of three
can make use of a nursery class, nine-thirty to three-thirty even — that’s
maximum.”^ ^ But then he felt it was essential for the mother to be home
at 3:30 when the toddler returned. This, of course, meant part-time work,
at best, during the preschool years and perhaps for several years thereafter.
Such views made Bowlby a household name in Britain all over again,
but more of a demon now than a saint (“the notorious Dr. Bowlby,” as
one female journalist wrote In the United States day-care advocates
saw the culprit as psychoanalysis in general (for placing so much empha-
sison the mother-child bond) and attachment theory in particular (vir-
tually unheard of before). Then, in 1986, the demon’s horns were passed

to Belsky.

Jay Belsky is a dynamic, witty, bearded man with a warm smile and sub-
stantial presence, some of which comes from his command of the data,
some from his power with language, which barrels out of him in whole
paragraphs of tanklike intensity. Belsky had studied developmental psy-
chology in the mid-seventies at Cornell, where he first read Ainsworth,

*”There is a j^reat deal of research regarding children in the kibbutz,”

Lieherman says, “showing that
these children are preferentially attached to the parents rather than to the metapelet
(or caregiver).
My own clinical experience and observations of normal children in child care arrangements indicate
that whenever the parents are adequately emotionally invested in the child, they
become the pre-
ferred attachment figure even if the child spends eight to ten hours a day with
someone else.”
(Personal communication, September 28, 1993.)
 A Rage in the Nursery 321

whose writing he found refreshingly clear and direct. Despite the mild

resistance of his mentor, Henry Ricciuti, who saw attachment as a spent

force hy that time, he was drawn to her ideas. He considered himself a

hard'Core empiricist, however, and, believed, with Michael Lamb and

others, that there were “serious limitations” in Ainsworth’s

Baltimore

study. Hedetermined that when he had the chance he would do larger^

scale replications of her work in order to test her fundamental
assump-

tions about the origins of attachment differences.

As a young professor at Pennsylvania State University, Belsky proved
that he was not only energetic, fast-working, and prolific, but that
he was

an excellent proposal writer who could pull in large grants. He quickly

had several huge longitudinal studies under way. Having studied with
Urie Bronfenbrenner at Cornell, he brought to this research some of

Bronfenbrenner’s interests in the broader family and social context in

which the mother and child live. Thus, Belsky was the first to evaluate
the impact of the marital relationship on the baby’s attachments.

Although Belsky had not been mentored by any of the major attach-
ment figures or by their students, in a remarkably short time he became
an attachment reckoned with himself. This was a notable
figure to be

accomplishment, especially when one considers that in attachment cir-

cles, as in other academic coteries, genealogy —
who studied with whom
and for how long — can sometimes play an excessive role in the accor-

dance of both respect and cooperation. But Belsky was not to be put
off.

“He runs a big shop, with a lot of graduate students, and he cranks out a
lot of data,” says Jude Cassidy, also at Penn State,
although across campus

in another department. “You can’t write an attachment paper today

whatever
without quoting him, because he has a lot of the basic data for
case you’re trying to build.
In his 1978 day-care review, written at twenty-six when
he was barely
Situation,
out of graduate school, Belsky had been critical of the Strange
but reversed himself after Sroufe alerted him to the
Minnesota studies,

just then coming out, which suggested strong

predictive validity for the
capitulated, and
Strange Situation classifications. Belsky immediately
“It was a rare
Sroufe was disarmed. In his next letter to Belsky he wrote:
experience for me an open response to a critical message. You are
to get
ability to honestly
in a very small group of academics who have the

change their position (that without pretending that they haven’t). 1

is,

think this bodes well for your career.”^^ This comment

would prove to be

perversely prophetic.
 322 Nature^Nurture Erupts Anew

A leading commentator on day care, Belsky wrote and spoke fre^

quently on the issue, his outer^borough New York accent and spirited,
rapid'fire speech becoming known to a variety of audiences, including
the United States Congress in 1984. As a result of his expert status, he
was constantly having to evaluate new evidence, some of it suggesting
that group child care might be more of a problem, especially in certain
situations, than he had believed before. The new data, all of it related to
children who had begun extensive care — more than twenty hours a
week — in the first year, described preschoolers with higher rates of
avoidant attachment; first-graders who were more difficult, argumenta-
tive, and aggressive — hitting, kicking, swearing, shoving; and nine- and
ten-year-olds who were seen as troubled by their peers — in other words
many of the sequelae of anxious attachment found earlier by Sroufe.^^
Although at least one of the new studies was clearly done with children
who had inferior day-care experiences, another had studied children
with high-quality care, while others were mixed.
“There was a slow, steady trickle of evidence,” Belsky told me. “1

would acknowledge the disconcerting evidence and, like everyone else, 1

would explain it away. In fact, 1 had this lovely correspondence with a
psychologist named Judith Rubenstein. In 1981 she and Carolee Howes
found that kids with infant day-care experience threw more temper
tantrums, had more fears and that kind of thing. And they argued in
their work that these kids were having difficulty negotiating the develop-
mental task of autonomy. 1 wrote about it, hut 1 wasn’t willing to
embrace it, because 1 felt at that point there wasn’t enough data to sub-
stantiate it. 1 was saying, Judith, 1 think the data are interesting, but, in
the face of other data, these aren’t enough to turn me around.
In 1985 Belsky was preparing to give a talk at the University of
Minnesota, outlining his thinking on day care and about the new data
that were continuing to trickle in and cause him to scratch his head. “All
of a sudden 1 realized, what am 1 spending all this time explaining away
every piece of disconcerting evidence? A lot of energy was going into
explaining away data, and every year there’s more explaining away to he
done. ^ At that talk and at the next one, before the American
Academy of Pediatrics the same year, he voiced a concern. The data
«

were suggesting to him that more than twenty hours per week of day care
in the first year placed a child at a somewhat increased risk for anxious
attachment and future behavior problems, with the child tending to dis-
play aggressive and noncompliant behaviors between the ages of three
 A Rage in the Nursery 323

findings were
and eight.* There were caveats, however. Some of the
quality. The data
based on community day care, which was mainly low
could be reflecting the fact that families who tend to put their children

day care in the first year are under more stress than families
in extensive

who don’t, and those than the day-care experience itself

stresses rather

could have been causing the anxious attachment.

There could be other
for concern.
confounding variables as well. Nevertheless, there was cause
The distinguished Yale developmentalist Edward Zigler, who was in the

audience, told Belsky he had come to a similar view, and asked him
to contribute a chapter to a new book he was editing on the parental
leave crisis.
Three, the huh
Also in the audience was the associate editor of Zero to
letin of an organization by the same name
that reports on childhood

issues to policymakers. The editor asked Belsky if

he would write up his
them. He was reluctant, partly because he considered
new thoughts for
of the promn
Zero Three to be a marginal publication and was unaware
to
both in the
nence the organization and its newsletter had achieved
a piece under review
professions and in government. “At the time, 1 had
the American
The American Psychologist''— the flagship journal of
at

Psychological Association
—
“and I begged off. But she was very persua-
sive, she said. Oh, come on, and 1 finally
agreed to do it.

The article that Belsky wrote for Zero to Three was an excruciating

exercise in walking He began by talking about the huge variety

on eggs.
to draw conclusions about
of day-care options and how difficult it was
emotionally charged the
the institution as a whole. He discussed how
negative.
issue isand the warnings he’d received against saying anything
were possible, that
He carefully stated that no irrefutable conclusions
differently,
someone looking the same evidence might justifiably see it
at

relatively persuasive circumstantial case can be

made that early
but that “a
associated with increased avoidance of mother,
possi-
infant care may be
greater insecurity in the attachment relationship,
and
bly to the point of
may also be associated with diminished
compliance and
that such care

was the way the available studies were struc-

* Belsky chose the figure of twertty hours because that
twenty hours” per week). But clearly there is a vast
differ-
tured (i.e., “under twenty hours” vs. “over
poor group care, four hours per weekday and a ba y s
ence between a baby’s being in group care, even
eight, nine, or ten hours per weekday,
being in group care, even good group care,
tew have cc in-
this overly encompassing hourly
range has not figured prominently in the debate;
should not be umped in
plained that day-care experiences of twenty or twenty-five hours a week
to depend on.
a stable and available substitute figure
with whole days away from the mother without
usually concerned about the needs
1 suspect this is so
because supporters of early infant day care are
usually require the longer hours.
of full-time working mothers who
 324 Nature'Nurture Erupts Anew

cooperation with adults, increased aggressiveness, and possibly even

greater social maladjustment in the preschool and early schoohage
years.” (Italics are all Belsky’s.) He even went so far as to allow “that my
gender and the more or less traditional nature of my family structure
could bias my reading of the evidence.
At this time it was not at all difficult to find pediatricians and deveh
opmental psychologists who were openly critical of infant day care, and
who did not couch their message as carefully as Belsky had. Urie
Bronfenbrenner, who was opposed to most day-care options as they cur-
rently existed, had said, “You can’t pay for an irrational commitment,
and yet a child needs that. He needs somebody who will not just be there
certain hours and then say, ‘I’m off now, 1 work nine to five.’”^^ But nei-
ther Bronfenbrenner nor any other day-care critics had endured the
onslaught that Belsky was about to suffer.

As far as was concerned,” Belsky recalls, “Zero to Three was nothing,

1

who was going to see it? But it ended up on every congressman’s desk
because ZERO TO THREE knows how to play the Washington game. And
then it ended up in the Wall Street Journal as a big story.” According to
Eleanor Szanton, the executive director of ZERO TO THREE, “This was
definitelymore controversial than anything we’d ever published.”^^
Emily Eenichel, the editor who had invited Belsky’s contribution and
found it quite moderate before publication, remembers thinking after-
ward, “Oh, my God, what have 1 done?”^®
Belsky s article was political dynamite, for it represented an important
reversal by an iconic The person who had reassured the public that
figure.

day care was okay now warned of risks. The risks were only for extensive
day care begun in the first year, but the warning was perceived as a crack
in the wall of the day-care defenses. Later that year
Belsky reported his
own day-care study that showed higher avoidant attach-ment among
early day-care babies, thereby confirming his belief in that
waming.^^
In the political climate that existed at the time, and still exists, such a
warning is not merely the statement of a scientist evaluating new evi-
dence. It was taken as a political move. In a subsequent issue of Zero to
Three four prominent developmentalists teamed up to rebut
Belsky,
accusing him of the selective use and misinterpretation of data.^® In
an
additional rejoinder Stella Chess reflected the underlying political
anxi-
eties when she wrote: “An unsupported dictum that such
day care is a
‘risk factor’ can only cause unnecessary among working mothers and
guilt
provide ammunition for the many elements in our society who are hos-
 A Rage in the Nursery 325

tile to the idea of spending public funds sufficient to

provide good day

care facilities for all young children. Belsky, a supporter of better day-

care options for all ages and a believer in the value of day care for many
a scientist
children over age one, was stunned by such attacks, as if

should not report an uncomfortable truth. What good are we doing

mothers, he asked, if we tell them full-time infant day care is okay if it

turns out that it isn’t?

appeared in March 1987. In it the noted

The Wall Street Journal article

developmentalist Tiffany Field characterized Belsky ’s views as

“bunkum.”

Sandra Scarr dismissed them as part of a backlash against the

womens
movement.” She said: “The advice for women has always been to get out

of the work force. This is just another way of saying the same thing.
The intensity of feeling surrounding the issue was such that ZERO TO
Three organized “summit meeting,” chaired by Edward Zigler, to try to
a
at least hammer out
get the opposing sides to speak to each other and
emerged
some minimal basis for agreement. The research guidelines that
of Child
from that meeting were adopted by the National Institute
collaborative
Health and Human Development for use in a major
research project on infant day care that continues to this day. But the

quarrels and polemics would not die down.

that year, Scarr,
Ata conference at the University of Virginia later
development
who was quoted as saying that infant social and intellectual
ridiculed the idea
isrendered nearly invulnerable by biological design,^^
If parents had to
that early day care was worse than later day
care:

choose a time to take off a few months to be with their babies, for the
few but the end of the first year
baby’s benefit, it would not be the first

and any time in the second year.” Before that, she told the New York

Times, children’s “brains are ]ell-0 and their

memories akin to those of

decorticate rodents.”^"^
Belsky’s opponents pointed out that many parents, due to financial
no choice but to put their babies in
necessity or lack of a spouse, have
Better to work to improve it. Belsky
child care. So why denigrate it?

in favor of improving but meanwhile what-

responded that he was all it

may be should be known, especially to parents who do

ever risks there
to decide whether
have choices. Such parents, he said, may be “trying
might be wiser to
two incomes are absolutely necessary and whether it

the mother) to the labor

defer the return of the second parent (typically
especially given the quality of day
force until after the infant’s first year,”
care currently available.
 326 Nature 'Nurture Erupts Anew

As the controversy grew, the gray fell out of the picture. Belsky was no
longer issuing a limited warning, he was blackballing all of day care. That
spring Kagan was quoted as saying, “Ten years ago, I wrote a big book on
day care that said essentially that ‘Day care is safe.’ When I read ten years
later that suddenly day care is bad, it makes me symbolically cry. . . . No
one can say day care is blatantly bad.” Yale psychologist Deborah Phillips
asked rhetorically, “Have we raised a bunch of thugs so far in day care?”^^
In the furor that followed, one began to hear new things about Jay
Belsky in psychology departments, in private conversations, and at con-
ferences where developmentalists gathered. It was said that Belsky pur-
posely sought out Zero to 1 hree so that his anti-day-care message would
have the maximum possible impact. In a caricature of his own earnest
search for possible bias, it was said that Belsky, who was married at the
time and whose wife stayed home to care for their two sons, then four and
six, thought everyone should raise kids the way he did. Also, that he was
opposed to women working and that he was antiwoman in general. Brian
Vaughn’s contention that the child experiences being left off at day care
each day as a rejection was now wrongly attributed to Belsky as well.
constantly heard these comments,” Jude Cassidy told me. “Jay
“I

Belsky says that if a kid is in day care he’ll be scarred for life. Jay Belsky
says that day care really screws up kids. Or Jay Belsky hates day care and
says that day care is like death. As I say to my students, what Jay Belsky
actually says is that the evidence at this point seems to suggest that if

children in the first year of life are in day care more than twenty hours a
week, it’s an increased risk for insecurity. While at the same time noting
that most babies in day care are securely attached, over 50 percent. It’s

just that the percentage for home-reared babies is better. Jay says: ‘I

believe that most day-care babies are secure.’ But it gets translated as.
Day care just stinks and if you put your kid in day care it’s horrible.”
Belsky: “The interesting thing is, in the 1978 piece, when we essen-
tially gave day care a green light, we made it very clear that we were
talking about research on day which was done on high-quality pro-
care
grams in universities and that therefore we didn’t know anything about
community day care. The fact that that work got publicized in the media
as ‘No risks for day care,’ without the appropriate qualifications and
caveats, nobody had any problems with that. That was a politically
correct thing to say, so it didn’t need caveats. When it came to politically
incorrect things, even if you gave the caveats, it was as if you hadn’t
given them.”^®
 A Rage in the Nursery 327

It was not long before Belsky felt a shift in the atmosphere around

him. ‘d was at SRCD [Society for Research in Child Development], the

year after, 1987, in Kansas City. This was the first big meeting after this

thing had broken. I have a fear that I m going to be attacked, that I am

could feel and hear the hissing. And I don’t cow. I make
the enemy. I

I’m not against day care, this is a scien^

my arguments and basically say,

tific issue to me. I see Urie Bronfenbrenner, and I go over to say some-

thing. It turns out he’s sitting with my biggest nemesis at that point,

Sandra Scarr.

“I really want to avoid her, but Urie tries to make peace. So we start

and Sandy starts to bait me on day care. 1 finally say to her, ‘You
talking
know what I don’t understand, Sandra? You’re so certain that what I m
saying is absolutely wrong.’ I said, ‘To tell you the truth, I m plagued with
my argument circumstantial, I called it inferential. I’m
doubts. I called

not convinced of this. But you have no doubts. She says. Don t play
Hamlet with me. Jay!’ So I figure. Hey, this is not the type of encounter 1
wanted, it’s time to disengage. I say— and I’m just trying to he collegial

in the worst sense of the word — I say. Well, Sandy, you know this has
these, its impor-
been an interesting discussion, we should have more of
but at that point 1
tant to talk, you know.’ I hate that kind of bullshit,
just wanted to get away. And Sandy says,
in an angry voice, ‘We don’t

have to talk anymore, you just have to figure out why you’re so angry.

Why are you so angry. Jay? Why are you so angry about all of this?

does not recall this encounter and has been surprised

to
Scarr, who
notes that others dis-
learn that Belsky found her such an antagonist (she
like him more), did and still does —
believe that he had an axe to —
differently, emphasizing
grind. If not, he would have reported the data

the urgency of improving infant day care rather

than “how can we make
mothers feel guilty and home.”^^ But whatever happened at
stay at
however
the SRCD and at similar meetings during this period, and
much Belsky may have invited it by dint of his own style— perceived by
even obtuse at times there can be
many as brash, cocksure, abrasive,

doubt that he stepped on a raw feminist nerve and

took a lot ot
little

heat for it.

“People say to me. How do you take it? They say, why d you get

And the naive truth is, 1 had no idea. 1 had no idea that
involved in this?
what 1 was going to he that controversial.”'’'
had to say

Even in the professional journals, where the debates were

much more
felt he had become a target.
polite and well reasoned, Belsky sometimes
 328 Nature^Nurture Erupts Anew

One of the most important articles replying to him, by Alison Clarke^

Stewart in The American Psychologist, bore the title “Infant Day Care:
Maligned or Malignant?,’’ implying, Belsky felt, that he hadn’t merely
raised a concern about early day care but had smeared it. But Clarke'
Stewart, who had dedicated her book on the subject to her son who
“spent his first year in day care so that this book could be written,” went
on to offer a series of reasoned objections. She argued that critics of
infant day care tended to overstate the degree of risk — her reading of the
various studies placed the actual increased risk of anxious attachment at

8 percent (Belsky placed it at 10 to 15 percent, while some evidence sug'

gests could be as high as 22 percent); that the aggression and noncom'
it

pliance seen in some children who had spent their infancies in day care
might actually be a reflection of the increased independence that day
care promoted; that insecurity with the mother did not necessarily mean
a child who had many other relationships was altogether insecure; that,
for children who were getting inadequate care at home or who had diffi'
cult relationships with their mothers, a daytime spent with a pleasant,
competent caregiver could improve their chances for healthy develop-
ment; and, perhaps most important, that the Strange Situation might
not be the appropriate tool for assessing security of attachment where
day-care children were concerned.
The debate placed the Strange Situation hack at the center of contro-
versy. How could this brief, artificial procedure, some asked, measure the
quality of an entire relationship? Belsky, who had himself initially
doubted the validity of the Strange Situation, now found himself its most
embattled defender. ‘This business about its being artificial is the dumb-
est scientific critique have ever heard. Does anybody protest the cardio-
I

vascular stress test because its artificial and short? Does anybody
say that
a thermometer can t work because it’s a silly piece of
glass with some
mercury in it and you only keep it in your mouth for a couple of minutes?
The issue here is not what it looks like, not its face validity, the issue is

its empirical utility.”"^^

In a recent paper written with Jude Cassidy, Belsky was

able to show
how consistent the Strange Situation has been in assessing the effects
of
various risk factors. If you take the data on the Strange
Situation, in
terms of rate of insecurity first for kids who were abused, then
for kids
who have depressed mothers, then for kids who have early and extensive
day care and then for just regular kids, you get this nice linear function.”
(Eighty-five percent insecure among maltreated children; 50 percent
 A Rage in the Nursery 329

insecure among children with clinically depressed mothers; 40 percent

insecure, on average, among children with over twenty hours per week

of infant day care; 25 to 30 percent insecure among children in “low-

risk” samples — that is, children from stable, middle^class homes with less

than twenty hours of outside care.^^) “And 1 make the argument, 1 say.
You people who are so ready to call this an invalid assessment, you show
me another developmental assessment of twelve^to^eighteen-month-olds
that
that is this sensitive not only to variation in care, but to conditions
one would think would be greater or less risky. It’s gotten to the point
where the Strange Situation is like an IQ test; it’s a procedure everybody
loves to hate. And I think they hate it because it works and it tells
them
things they don’t want to believe. What if we found that kidswho went
to day care were more likely to be securely attached? Would all these
people be going nuts over the Strange Situation? Of course not.”
Alison Clarke^Stewart, Jerome Kagan, and others, however, argued
separations,
that because children in day care were accustomed to daily
Strange
they were less stressed by their mother’s departure during the
Situation and so they cried less, played more, and were relatively uncon^

cerned about their mother’s return. To label such children avoidant

was

therefore a terrible misnomer. Belsky dismisses this argument

much as
avoidant
Sroufe had dismissed earlier arguments that children labeled
were actually more independent. “If you go to the 1983 Handbook of
tables
Child Development^ Belsky says, “Alison Clarke^Stewart actually
care and
the data on this question by looking at how much kids in day
not in day care cry in Strange Situation^like procedures. Her
analysis

clearly shows that there is no difference in the amount of overt distress

shown by kids with and without day-care experience in these separation

paradigms. That’s her data that she’s amassed. And yet, she has the in-

tellectual audacity again and again to make the argument that these
avoidant kids in day rare are falsely appraised.”

It wasn’t long before Belsky had his own study up and running to test

this proposition in an even more thorough way. And we found no sup-

port for it. The avoidant kids with day-care history were not spending

more time playing and were not less stressed than avoidant kids without
was just the reverse. So that argument cob
day'Care history. In fact, it

doesn’t stop people from using it, because, see, it’s a political
lapses. It

debate. The idea is to keep throwing darts, undermining jay Belsky ’s

arguments with these darts. It’s a political dialogue, it’s not a scientific

dialogue.
 330 Nature -Nurture Erupts Anew

Some believe that Belsky thrives on controversy and wades in where

others loathe to tread. (“I think he did ride this horse for quite a ways,” an
erstwhile supporter says.) It is impossible to miss the enjoyment he takes
in presenting his data, being in the spotlight, smiting the opposition.

When I asked University of Alabama psychologist Brian Vaughn why

other people like himself did not get into trouble for their reports on the
problems of babies in day care, he said, “How many of those people went
on Good Morning, America!''^^^

Belsky ’s verve in speaking out has caused him to rush headlong into
an area of acute female sensitivity. Many mothers find it agonizing to
leave their children, especially if they are very young; they constantly
second'guess themselves about their choices; and they feel jealous of the
people who replace them during the day. Mothers like this do not want
or need to be told that they may he harming their babies, especially by a
man, especially one more inclined to hammer out conclusions than to
tread gingerly or to soothe.
The contrast Belsky creates with the rest of his profession is stark.
Other psychologists approach the day-care issue with a delicacy that hot'
ders on dread. I m a little leery these days of being put into the pigeon
hole of being anti'-the'advancement'ohwomen,” a famous develop^
mentalist says when I point out the care with which he chooses his words
and the disclaimers that adorn his speech. “You’re going to trap me,”
laughs another when I question her about her feelings regarding the
quality of the debate that has taken place on this issue. I asked one
researcher what he would do, given what he now knows, if he were a
young man just starting a family. How would he feel about his and his
wife’s both working fulhtime through the child’s first year? There was an
extended silence. Then: I can’t answer that. I can’t imagine myself as a
young man.”
Well, what if young couple should ask you your opinion: Should we
a

try to work it out so that one of us can stay home during much of the first
year? Would you say it really isn’t that important.^”

Another protracted silence. “No, I wouldn’t have any opinion.”

“You wouldn’t have any opinion or you wouldn’t have one for publican
tion.^”

“I wouldn’t have one for publication.”

1 asked psychiatrist Peter Barglow if he thought that people censored
themselves when talking about day care, especially if they have anything
 A Rage in the Nursery 331

cautionary to say. think they do” he said. The reason is one gets

quickly bashed by people who are politically correct.” When Barglow

published findings about the risks of anxious attachment with nanny

care, he himself was criticized “by feminist researchers” as well as by

“some personal friends and colleagues.” His wife, who was assisting him
like that.
at the time, was warned by friends, “You can’t say things

For his part, Belsky recognizes that he does not behave like the typi'
cal academic. He points out that his parents were not college educated.
They owned a luncheonette in Manhattan’s garment district where he
worked for many years when he was growing up, and his family life was
riddled with open conflicts more typical of a milliner’s shop than a uni-

versity department. “The last thing I ever envisioned myself being was a

scientist or an academician. In many ways,” he laughs, “the personality

doesn’t fit the job.” But rather than brash or abrasive he prefers to see
himself as a rugged individualist, someone “who’s not willing to say
all

the right things,” who doesn’t play the “smile in your face stab
you in the

back” academic game, who doesn’t see “achieving high levels of cohe^

sion” as “the ultimate goal of intellectual discourse.”

Meanwhile, he
did not help
believes that while growing up in a “high'Conflict family
him develop the correct personality for academia, it has enabled him to

know how to fight back, he says, and while it has

weather its storms. “I

been stressful, it’s not obliterating, as it apparently would be to a lot of

people.”^®
he them, have been costly. His
And yet “the day-care wars,” as calls

reputation for the most part remains intact. He is still a professor with
pub'
many graduate students, getting grants for big studies, prolifically
a member of the Multi-Site Day Care Consortium,
the huge
lished, and
study sponsored by the National Institute of Child
Health and Human

Development that will hopefully get to the bottom of the day'Care queS'
tion in future research. But there has been a subtle
change in his status as

viewed by many of his peers. He is a known quantity now, a

man with a
flag, the wrong flag.

whole thing has been an effort to marginalize me, and

“What this is

By 1985, went to a child development

it’s been remarkably effective. if I

meeting, I had lots of people coming up to me, wanting to talk to me,

wanting to talk to me about my work, wanting to talk about their work. 1

in a hallway and not being able to finish a con'
can remember standing
versation because of all the interruptions. I mean, don t get me wrong, 1
332 Nature'Nurture Erupts Anew

enjoyed it. Now I walk through a meeting like Hamlet’s ghost. Nobody
sees me, nobody comes up to me. I’m exaggerating — nobody 1 don’t
know, nobody who isn’t a good friend of mine. I’m a pariah. violated I

the Eleventh Commandment of Developmental Psychology Thou —

Shalt Not Speak Any 111 of Day Care, whatsoever, ever.
“And think the second good example of this is that my friend Larry
1

Steinberg and 1 —
he was the guy who co-authored the first day-care
paper with me in 1978 —
we published a textbook two years ago. And
you know at the beginning before the textbook came out, said to 1

the publishers, we should think about leaving my name off this book.
They said. No, don’t be ridiculous. Well, lo and behold, sales people
walk in wanting to sell this book. And whose offices do they go into?
Assistant professors, associate professors. And who are they? They’re
working females. ‘1 don’t want a hook by Belsky.’ They show them
what it has to say about infant day care, its reasons. ‘1 don’t care, 1 don’t
want it.’ In fact, it’s gotten to the point where the sales manager has told
them, ‘Don’t even bother pushing that book, you’re just going to lose
credibility.’” Belsky’s next text, also a collaboration, did not have his
name on it.

“Urie Bronfenbrenner said to me: ‘You know. Jay, what you have to do
is always make it clear, up front, that day care is here to stay and that
you’re not against day care.’ 1 said, ‘Urie, c’mon.’ 1 was in the Soviet
Union with him in 1983. was amazed that everybody who got up and
1

gave a talk on the Russian side had to start off with something about
Marx and Lenin. said, ‘I’m not against day care, and if raising concerns
1

is going to be taken as being against day care, let it be. We can’t have a
climate in which everybody kowtows.’
“The whole thing has opened my eyes to the nature of academia, and
how low-brow it really is. How people can pontificate and just be so full

of shit. 1 expected better out of academics, and I was probably naive in

that regard. They’re no different, 1 guess, from anybody else. But it really
burst my bubble about what the academy was all about. When honest
disagreements are not treated as things worth having, but as a battle
between good versus evil — with the evil claiming to be good, as far as
I’m concerned — that’s been disappointing. You know. Wait a minute,
we’re supposed to be having an intellectual discourse here, we’re sup-
posed to he considering.
 —
A Rage in the Nursery 333

Although few have come forward to say, Count me with Jay, Belsky’s

views are not considered controversial by most attachment workers, and

even outside the attachment camp they are recognized by many as quite
“1
moderate. Meanwhile, he has won some admirers for voicing them.
think jay did the field and women an enormous service,” says Arietta

Slade, a clinical psychologist and attachment researcher at the City

University of New York. “Because in subtle ways the women’s movement
has encouraged women to minimize attachment experiences during the
child’s early years. And that’s very destructive. Mothers who leave their

children for long hours each day often convince themselves that what'
ever arrangements they came up with for their kids are just fine. Slade
believes the 1980s were a damaging time for child rearing, because many
working mothers overrode their best instincts in an effort to simplify an
inherently difficult situation.
“It’s hard for women to have a career and be a parent. Very hard. Its

horrible that it costs money to have good day care, it’s horrible that

women don’t have the same freedom that men have, it’s horrible that

you can’t have it all and raise kids who still feel confident that there’s

going to be somebody there for them. But it’s better to say it’s hard and

struggle with it than to say. Well, it feels hard but it s really simple and
I’ll just do what I need to do.”^^
Eleanor Szanton of ZERO TO THREE also believes that Belsky did the

right thing, and despite the fact that he has not made many friends in

psychology, others have been affected by his views. “It’s interesting to see

how minds have changed,” Szanton says. “When I came into the
people’s
organization in 1979, Selma Fraiherg, who was one of our founders, had
just published Every Child's Birthright
"
—
which urged mothers to stay
home with their babies, opposing even part-time group care until the
fourth year
—
“and she was persona non grata after that with the
women’s movement. The book was seen as a putdown of women and an
effort to push them back in the kitchen. 1 think many people now
because of a combination of Belsky, and the tremendous upset with the
quality of child care, and more women realizing that they can’t do it all
— are much more willing to listen.”

Speaking of professional child care advocates like herselt, many ot

whom were distressed and enraged by Belsky’s original article, Szanton

says, “I think everybody feels now that Belsky did a service. It has made
policymakers at least begin to consider whether parental leave” — the
334 Nature 'Nurture Erupts Anew

bill finally passed into law in the opening months of the Clinton admin-
istration
—
more appropriate than they had previously thought. It’s
“is

forced them to consider whether there should be more funds for better
child care.”
Also, partly because of the controversy Belsky stirred and, perhaps
even more because of the tireless educative efforts of Berry Brazelton and
others, there is a greater appreciation of the importance and delicacy of
the infant period. remember meeting Senator David Durenberger
“1

[Republican-Minnesota] in the early eighties,” Szanton says. “He was

always a friend of child care. And 1 made a remark to him that infants
and toddlers, of course, needed more expensive care, because the ratios”

— of children to staff
— “needed to be lower. And he looked at me and
said, ‘Why would that be?’ He thought they just needed baby-sitting at

that age. 1 think that mentality has begun to change.”

Lieberman elaborates: “The caregiver must have the time, the talent,

the experience, and the training to learn to know a baby.” That includes
“treasuring this baby’s most charming, endearing attributes,” and helping
the baby cope with potentially problematic aspects of its temperament.
The fussy baby, she says, needs to learn to cope better with its fussiness,

the shy baby needs to be encouraged to explore, the reckless baby needs
someone attentive enough to build in safeguards.
The real world of infant day care rarely provides this, at least not in
the United States. As of 1992, day-care workers averaged $5.35 an hour,
less than parking attendants and kennel workers, which would help
account for the huge turnover rate (41 percent per year in 1988). Only
three states set staff ratios that met ZERO TO Three’s hardly utopian
standards (one to three for infants, one to four for children under three).
Nine states still allow one caretaker per six infants; nineteen one per
eight toddlers. Meanwhile, few knowledgeable parents who could
afford better would consider even the approved ratio of one to three for

their own babies when they are first learning about loving relationships
and whether people can he trusted. (Any mother who’s cared for triplets

can tell you why.) In recent years, study after study has found the quality
of care in this country — in centers, with relatives, with the kindly
neighborhood woman — to be wretched, humdrum, or barely adequate.
In a major 1994 study of family day care by Ellen Galinsky of the
Families and Work Institute, only 9 percent of the homes she rated were
good, with 56 percent rated adequate/custodial, and 35 percent develop-
mentally harmful.
 A Rage in the Nursery 335

What Scarr said in 1986 still appears to be true: “Most families find it

far easier to purchase quality cars and refrigerators than to buy good day
”55
care.

Meanwhile, in academia, the day-care hostilities drag on. Belsky con^

tinues to be the man when they want a negative view,

the media contact
and he continues tracking the data. The trickle of evidence that diS'
turbed him in the mid'eighties, he says, has now become a steady stream,
adding ever more confirmation to his warnings. In a huge study of subur^
ban Dallas families, psychologist Deborah Vandell, now at the University
of Wisconsin, found that third'grade children who had had 30 or more
hours of day care begun some time in infancy scored the lowest in emo-
tional welhbeing, work habits, peer relations, and compliance (as aS'

sessed by parents, by teachers, by a child’s selhreport quiz, by peer

evaluations of popularity, and other measures). Texas, it turns out, has

some of the worst standards in the country for day care, so many of these

children were probably in poor care. But Vandell found that the day'Care
effects held regardless of the income level, marital status, and race of the

families involved. (Vandell had an arduous time getting this study pub'
lished, more so than any of her other work, which surprised her, “because
I’d never had such strong findings.
In a 1994 study. Jack Bates at the University of Indiana reported that
the more day care a child has received (regardless of when it occurred)
the more likely he is to display problematic levels of aggression at age

five. Bates, too, studied children from all social classes and told me that

there is nothing in his data “to suggest middle class families can be san-

guine on this subject.”^^

Critics continue to point out that quality is the real issue. And they

too can cite studies to holster this point, many of them from Western
Europe where the quality of care is typically beyond almost anything
known in the United States, and where early day care was found in one
study to enhance development. Belsky himself concedes now that if

nonparental care for babies is good enough, there is no need for parents

or society to be concerned, and he too favors a federal

program that
parental
offers quality day care to all Americans, as well as subsidized

leave.Where he differs from his critics is in his insistence in pointing

out how far we are from having such care and how damaging it is
tor

children when they don’t get it. “Everyone says that kids need
good

quality care,” Belsky argues. “But no one is willing to take the next

step and point out that poor quality care is had tor kids — and that
336 Nature'Nurture Erupts Anew

most care in this country is poor quality, so a lot of kids are possibly

being harmed.
Belsky’s overall assessment of the data on day care remains complex
and carefully qualified. Regarding timing of entry: “As I look at the most
recent studies, the problem seems to be with extensive care begun in the
first year and continued thereafter. So, it’s really an entire infancy and
early childhood of fulhtime or near full-time nonparental care. Twenty,

thirty or more hours. If it’s just the first year, the effect can go away.”
Regarding quality of care: “What kids need are stable, enduring, sensitive
relationships. That’s really what’s critical here. And in American day
care, especially in infancy, especially in the opening years, that’s more
the exception than the rule.”^^ It children receive early and extensive
high-quality care from a nonparental figure, Belsky sees no great cause
for alarm — but the parents need to recognize that the alternate attach-
ment figure may now count more prominently in the child’s develop-

ment. Belsky, meanwhile, speaks of day care versus “parental care” rather
than “maternal care,” in order to emphasize his belief — and the evi-
dence of the research — that fathers can also be primary caregivers and
that in terms of security of attachment children do just as well when
fathers play that role.

“The real risk of early and extensive [day] care as we know it in this
country,” Belsky says, “is when risks accumulate. We get unstable care,
turnover in care givers, poorly paid care givers who are unresponsive and
impatient. We get parents who are exhausted at the end of the day who
have no wherewithal to care sensibly for their kids. We have employers
who have no tolerance for the fact that their employees are parents,
too.”^® As these risk factors accumulate, the chance of anxious attach-
ment rises.

To Belsky the idea that people should soft-pedal the facts about full-
time infant day care because they may disturb mothers smacks of “yuppy
narcissism.” He was particularly upset by journalist Anna Quindlen’s col-
umn in the New York Times after that paper reported the Galinsky study
on the poor quality of family day care. Quindlen argued that mothers do
not necessarily provide better care themselves, so perhaps we need not
get all worked up about the quality of nonmaternal care. “My response,”
Belsky says, “is wait a minute! She’s so busy taking care of the emotional
needs of the mothers, she’s forgetting about the kids. It’s a good case of
what I consider disseminating intellectual Prozac. We’re going to regu-
 A Rage in the Nursery 337

late your anxieties here, not give you anything to worry ahout!”^^
Belsky recalls the scathing 1982 federal report on the state of the
nation’s schools. “Nobody worried about making teachers and principals

feel guilty. Nobody pulled punches when they found the schools weren’t
very good and that kids did badly in bad schools. And here we have a

perfect parallel, and people won’t acknowledge that care is bad. And
even when they do, they don’t tie it together, so that you come to a con^
elusion, like with schools, a nation at risk. If we were looking at any other

issue, teenage pregnancy or cocaine use, the same statistics would be a

cause for tremendous alarm. But when it comes to day care everything is

explained away.”^^
Belsky recognizes that all of the data are based on correlations, and so
it has to be remembered that while the findings may apply to some chib

dren, they do not apply to all. Indeed, the difference in rate of anxious

attachment between day-care and home-care infants is small to moder-

ate. Why some day-care children seem to be more at risk for anxious

attachment no one can currently say, although it seems reasonable to

assume that they face some other stress as well, such as a relationship

with their parents that is otherwise only borderline-secure or a tempera-

ment that requires a higher level of nurturant attention. Whether there

are subtler risks as well, affecting children who come out looking securely

attached despite full-time day care begun in the first year (Do they move
into a less secure subcategory? Do they develop a negative trait, style, or

preoccupation that is not picked up in the Strange Situation?) is still

unknown. The question has not yet been asked.

Bowlby approached the problem of working mothers and day care at the

social level. He recognized that “career girls are placed in a difficult situ-

ation” if they miss too much time and wished to see the workplace

restructured so that the choices facing them would not be so stark. “At

the Tavistock, when 1 was the chairman, we had a very able woman psy-

chiatrist who, when she had two boys, gave up work almost entirely for
two or three and then we were able to arrange for her to do two
years,

half-days a week. She kept her hand in, she remained part of the depart-
ment, and then, gradually, as the children got older, she came back.

Unfortunately, this is not the way most employers operate. And 1 think

it’s a thousand pities.

Bowlby did not defend himself against personal attacks, but Freda
338 Nature-Nurture Erupts Anew

Martin, a psychiatrist who worked with him, insists that, despite his rep-

utation, “he was very supportive in the most practical ways” when faced
with the problems of working mothers. “I can think of no other institu-

tion in the 1970s,” she wrote in tribute to him, “where because the
incoming Chairperson of a Department (myself) had young children, the
time of the Clinic Professional Committee Meeting, by long tradition
held at 4:00 in the afternoon, was altered without a murmur to 11:00 in

the morning.”^"^
That such options and adjustments are not available to most women
may be reflected in a disturbing figure: 97 percent of male executives
have children, compared to only 35 percent of female executives. These
statistics seem to suggest that for some women succeeding in the corpo-

rate world means giving up having children. And, of course, those who
do have children will have to give up their rearing to surrogates. The
business world, and our culture as a whole, does not yet take the process
of parenting seriously enough to give mothers — or fathers — the flexibil-

ity they need to care for their babies without insufferable losses in their
careers.

Bowlby was not insensitive to the argument that a mother who comes
home happy and fulfilled is better for the child than one who stays home
bored and resentful. But, despite the rage that swirled around him, it did
not sway him much, because as far as he was concerned the children still

suffered. Better to alter social thinking and expectations, so that women

would naturally want and expect to stay home in the early years and
have less cause for envy or resentment. When 1 told him that in the
United States it had become virtually unthinkable for many women not
to — which under current circumstances does indeed
have a career pull

them away from the home — he “Well, times change. mean people
said, I

don’t smoke; they used to. I think women will realize that if they have
children, then they have responsibilities, and if they want to have happy
relationships with their children in later life, then it’s important to do a
lot of work with them when they’re small. But 1 give it another twenty
”66
years.

From his strong words and firm position, one could easily assume that
he felt anything less than what he recommended would he harmful to
every child — and perhaps at some gut level he did. But he acknowledged
that studies showed that only a few children were likely to he harmed by
extensive infant day care. According to Scarr, “Eight percent, if you
believe that Alison Clarke-Stewart figure, is a very small difference. Is
 A Rage in the Nursery 339

that the stuff of which public policy is made?”^^ To Bowlby it was.

Indeed, what separated him from many of his opponents was how seri'

ously he took the idea of five, eight, or ten additional children in a hun^
dred becoming anxiously attached. “The proportion of people left with a
crippling illness after polio is minute,” he said. “But because we know
that that is a risk factor we take tremendous precautions. Well, now we
have ample evidence that certain types of experience in childhood are

risk factors. Plainly, there is every reason to abolish those risk factors if

you can.”^^
Ainsworth, a “career girl” herself, is more flexible about early day care,

more open to the possibility that supplemental mothering can be ar^

ranged without harming the baby. “From the point of view of the child’s
general welfare, the mother should be pretty consistently available. That
doesn’t mean she has to be there every moment, can never go out,
never

have anybody else look after the child, or anything like that. But fairly

consistently available. It’s very hard to become a sensitively responsive

away from your child ten hours a day. It really is.

mother if you’re

Women’sTib people have been finding it comfortable to assume that it

doesn’t matter what you do and that a woman owes it to herself to work
and do what fulfills her. People who focus primarily on the welfare of

children tend to ignore what suits the mother. But it’s really a matter
of

how do we adjust these two things. Had I myself had the children 1
longed for, 1 like to believe I could have arrived at some satisfactory com-
bination of mothering and a career, but I do not believe that there is any

universal, easy, ready-made solution.

In recent years many attachment workers, Sroufe included, have

noted the positive things day care has to offer, and, although wary of day
for toddlers and
care in the first year, they have come to believe that
older children, even full-time day care is not only not
harmful, but may,

at least in the case of good-quality care — with high staff ratios, well-

and each child assigned to a particular

qualified workers, low turnover,
caregiver — provide children with a range of stimulation
In other words,
and experience

that promotes their cognitive and social development.^^

if done in the right way at the right
time, day care can be a boon to the

family. In this they are largely in agreement with Scarr, Kagan,

whole
Clarke-Stewart, and other day-care advocates.
The need for a balance between the unconditional love and support
that group
parents can give and the exposure to the outside world
care can provide is captured in an observation by Urie Brontenbrenner:
 340 Nature'Nurture Erupts Anew

“It’s good for a child to he in the company of people who are crazy
about him for a substantial number of hours every day. I’m sure of that.
But it is also good to be with people who are not crazy about him. He
needs both kinds of experience. He needs some mothering, some father^
ing, some day care, even some coolness toward him. . . . All of these
needs must be met.’’^^

There is much to consider for both sides in this debate. For those who
are wary of infant day care it should be kept in mind that although the
statistics may be on their side — the risks are higher — statistics do not
determine the fate of any individual baby. More babies may thrive in
home care, but some do fine in alternate care, while still others do better
in day care than they would have done at home. Not every baby has the
option of a willing, nurturant mom. To be home with a depressed mother
who would rather be elsewhere is not a great thing, and studies have
shown that a mother’s satisfaction with what she is doing is a key vari'
able in determining how well children do.^^ If her baby is taken off her
hands for 30 or 40 hours a week, it may free her to be a better mother
when they are together.
Parents need to know that it is possible for both of them to work, that
they can put their baby in some form of high-quality nonparental care
and still have an emotionally sound child. There does not seem to be a
hard and fast rule, written into our genetic code, requiring that a baby be
cared for full-time by its mother or father until it is a year old. Many par-
ents, especially if they are at ease with themselves and their decisions,
and many babies, especially if they have an easy temperament and an
alternate caregiver who loves them and makes them feel secure, can
maintain a solid relationship despite beginning extensive nonparental
care at some point during the first year. Even in the studies of run-of-
the-mill out-of-home care, 50 percent of the children seem to be doing
all right.

But those who prefer not to worry about infant day care need to face
some facts as well. Most infants, in order to feel that their love is recipro-
cated, that they are valued and accepted, and that they are secure
enough to happily explore the world, seem to need a lot of unhurried
time with at least one person who is steadily there for them — preferably,
as Bronfenbrenner says, with someone who is crazy about them. And for
many bahies that need may not diminish as quickly as we might hope.
Because so much gets established in that early period, most developmen-
talists, regardless of their public stance on day care, recommend that one
 A Rage in the Nursery 341

parent spend some months at home after birth, and the number of
months they recommend is often far less than what they would want for

their own child.

Even under what might be considered normal conditions, it can take a

lot of relaxed, deep'time familiarity to understand a baby’s quirks and
preferences; to know, for instance, that he’s only rejecting his bottle
because he wants his blanket first, or that he’s willing to nap only if held
in a tight, rocking embrace until he stops his struggling and cranky cry-

ing and drops off. If the child has a difficult temperament, or the mother,
for whatever reason, has a hard time adjusting to the baby and reading its

signals, or if she is simply slow to develop her confidence, the two will
need long periods of time to get in synch with each other and become a
secure pair.
It has been argued that anxious attachment to a mother or father
might not be the worst thing for a young child, especially if he forms
secure attachments elsewhere. But it is certainly not a happy develop^
ment for the parent-child relationship itself. And because anxiously
attached kids have a way of pushing their parents’ buttons and resisting
overtures toward change, a vicious cycle may set in that keeps the rela-
tionship from getting righted. What’s more, in our primitive state of
knowledge, we cannot be at all sure that anxious attachment is the only

outcome to be feared.

Given all this, no parent can afford to listen to generalizations or

make a decision based on ideological preference.

Since 1980 the major increase in working mothers with small infants has
come among women. For many couples who can afford
college-educated
it — including most psychologists I know hiring a nanny is a common —
solution. This option is rarely discussed because it is open to so few, but it

is believed to be the least risky.

Nevertheless, a study of 110 families by Peter Barglow found a shock-

ing correlation. When full-time nanny care began in the first eight

months, it was associated with a much higher incidence of insecure

attachment to the mother at the end of the first year — 46 percent inse-

cure, compared to 29 percent insecure among home-reared children.^^

This figure emerged, according to Barglow, under optimal conditions,

with all the parents being “well-educated” and “deeply involved in par-
enting.”^^ And yet the results are about same as what is typically found
among babies in routine day care.
 —
342 Nature 'Nurture Erupts Anew

A six'year follownip, reported in 1994, gave Barglow more cause for

optimism. He still found “traces of adversity””^^ in the way nanny^reared

kids related to their peers at age seven (unfortunately, the only variable
he studied), but the percentage differences were small. This improve^
ment could be understood in several ways. One, that nanny care in the

first year has some built-in protective factor — such as secure attachment
to the nanny — that was not assessed in the original study. Another, that
sensitive and involved parents are often able to undo many of the ill-

effects of early insecurity in the child-parent relationship. More than

likely, it is a combination of the two.
My son, Raphael, was born during the writing of this book. My wife,

Thaleia, a partner in a small architectural firm, stayed at home full time

during the first year, with part-time care from a succession of baby sitters

— who took him off her hands for several hours each day — and frequent
help from me. By six months, we had a nanny coming in 30 hours a
week. By the end of one year, Thaleia was more than eager to resume
work, although she was prepared to adjust her hours, depending on the
baby’s reaction. When she started working about twenty-five hours a
week at the beginning of the second year, we increased the hours of the
nanny, who had become a valued person to the baby, not to mention us.

Rafie thrived with this arrangement. At this age we still did not feel
comfortable putting him in group care for the better part of each day
although, happily, he had plenty of children to play with in the building
where we live and in the playground nearby. At age three he started
preschool in the mornings and still had his nanny in the afternoons and
all day in the summers. All in all, hardly an ideal Bowlbian solution
which Thaleia would have found unendurable — but closer to his model
than some of the alternatives.
The political climate in academia has not always been helpful to par-
ents struggling with such concerns. Not much is being addressed, for
instance, to the parent who wants to hire help (what to look for in a sit-

ter or a nanny, what to avoid) or be sure that a baby cared tor at home
has adequate peer interaction, because so much attention is focused on
maternal care versus day care, with complex issues repeatedly reduced to
slogans. What’s more, it is difficult at this time for anyone who has nega-
tive things to say about early infant day care — and therefore working

mothers, and therefore women’s rights — to present the data in a

straightforward fashion. They either soft-pedal the facts or leap on them

with obvious ideological gusto.
 A Rage in the Nursery 343

In this atmosphere it is acceptable to favor a parent’s taking several

months off work after childbirth, as Berry Brazelton does, but to suggest a

parent take a full year or more (as many in the field privately do) seems
pointless, not only because one will be seen as crossing the line from sup'

porting women to oppressing them, but because economic realities do

not allow for that in the case of many families, and one must not seem
elitist.

National politics also plays a role. It has been hard enough getting a
family leave bill passed that allows mothers and fathers in large compa^
nies to take off twelveweeks (without pay) without losing their jobs; to
favor the year or more of family leave that other countries have insti-
tuted has, until now, been seen as jeopardizing one’s chances of getting
anything.
Meanwhile, the position of the poor has become more precarious.
Without going into what causes people to seek welfare, how many of

them are cheating or unhealthily dependent, or what proportion are

capable of being good parents if able to stay at home, the fact is that poor

children, like all others, need the love of a dependable caregiver, and
when their mothers work, they need high-quality alternative care. When
poor single mothers are required to work without adequate provision for
their children, as they are under recent welfare legislation, many chil-

dren will suffer. There are no easy solutions to the problems of poverty
and welfare, but whatever benefits are achieved by forcing single mothers
off the welfare rolls are probably offset by the costs that society will incur

as their children get older. Unless and until universal high-quality day
care is available to American families, it will be in the best interest of
society and probably more cost efficient, as well, to keep welfare mothers
on the rolls and give those that need it yet additional support — educa-
tional, psychotherapeutic, medical, even financial — to help assure the

secure attachment and healthy emotional development ot their kids. (Of

this, more in Chapter 27.) But there is not a politician in the nation who
would say this or a government official who would entertain it.

Nor is this view adequately reported in the press. Journalists, too, are

sensitive to the prevailing winds, and certain truths do not get spoken if

they seem to violate correct thinking. When Public Television aired its

documentary Childhood in 1991, a brief excerpt from Rene Spitz’s land-

mark film on the ravages of maternal deprivation, which helped change

the way a generation of child care workers looked at the needs of infants,

was used as a passing illustration of the point that early researchers had
344 Nature -Nurture Erupts Anew

gone overboard in their concern for the continuity of maternal care7^

That maternal deprivation and early separation had proved a legitimate

cause for alarm was not mentioned, presumably because the producers
did not want to complicate their message that mothers should not be
anxious about working. Because the topic has become so politicized, no
parent can allow this decision to be made by experts. Mothers’ needs are

different, every baby is different, and there is a lot at stake.

 23
Astonishing Attunements:
The Unseen Emotional Life of Babies

The loving mother teaches her child to walk alone. She is far

enough from him so that she cannot actually support him, but she

holds out her arms to him. She imitates his movements, and if he
totters, she swiftly bends as if to seize him, so that the child might
believe that he is not walking alone. . . . And yet, she does more.
Her face beckons like a reward, an encouragement. Thus, the child
walks alone with his eyes fixed on his mother’s face, not on the diffi-

culties in his way. He supports himself by the arms that do not hold
him and constantly strives towards the refuge in his mother’s em-
brace, little supposing that in the very same moment that he is emphasiZ'

ing his need for her, he is proving that he can do without her, because he

is walking alone.
— S0ren Kierkegaard, 1846^

When Bowlby first began arguing in the mid-fifties that the infant had
innate social capacities, that it was a primary actor in the creation of the

relationship with the parent, his ideas met with little resonance among
his peers. But in the early seventies, a new interest began to emerge for

observing the infant and for analyzing the infant-mother bond. Over the
course of the next decade, developmental researchers discovered that
what had been widely considered a suckling blob was in fact a surpris-

ingly social creature. They found that newborns by the eighth day are

able to distinguish the smell of their mother’s milk, and, if offered two
breast pads taken from nursing mothers, would consistently turn to the
one that came from their own.^ In an ingenious experiment, newborn
infants were given the opportunity to choose what to listen to by their

rate of sucking on an electronically rigged pacifier. They invariably

 —
346 Nature 'Nurture Erupts Anew

sucked at the rate that caused them to hear a recording of the human
voice — and, if given a choice, selected their mother’s voice above oth^
ers.^ It was similarly found that they prefer to look at thehuman face

over other visual patterns. That they are able to distinguish among cer-

tain emotions on the face. That they are also able to distinguish between
faces, and that by five to seven months they can recall a week later the
picture of a face they were shown for less than a minute. This was far

from the long^accepted “blooming, buzzing confusion” proposed by

William James.
The presence of complex social capacities in the infant brought new
interest in what Ainsworth had called maternal attunement. For it made
sense that a sensitive and responsive presence was needed for the socially
prepared newborn to develop into a truly social being. Winnicott had
perhaps been most eloquent on this score, holding that the mother’s abih
ity to tune in to her baby was central to every aspect of its psychological
development.
In timing with her baby, Winnicott believed, the mother (or primary

caregiver) struggles to bring the world to him in understandable form.

She him a sense of achieving what he is incapable of achieving on

gives

his own. By knowing what he wants and responding to it, she gives

meaning to his cries and other signals, so that eventually he knows what
he wants and learns to signal with intent. She puts words to his feelings.

As the American analyst Fieinz Kohut argued, the baby experiences her
as an extension of his own self,"^ her assistance contributing to his sense
of competence and keeping his still fragile self from being overwhelmed
by the stresses and tensions that constantly impinge upon him. Her
devotion, her ability to be intricately in tune with his wants, enables the
infant to experience himself as the master of his world, ^ to feel quite con^
fident and wonderful despite his helplessness, and to evolve a healthy

sense of self. The sensitive mother figure also helps the infant to stretch

his capacities by communicating with him at a level that is a little bit

hut never too far — beyond his current abilities. She reaches down to his

level while at the same time presenting him with opportunities to grow
— responding to his nonsense syllables as if they had real meaning or
encouraging him to walk for just a few steps without holding on.^
«
A long
period of intense and harmonious involvement with a sensitively at'
tuned mother is, thus, essential to the baby’s emotional health. She
could he considered the lattice on which much of his psychology and
mental organization develops. But aside from Ainsworth’s Baltimore
 —

Astonishing Attunements 347

study and the attachment classifications that grew out of it, there was lit'

tie empirical evidence to support this view.

Then in 1971, the year Winnicott died, Daniel Stern, a psychoanalyst
and infant researcher who had been studying infant'mother interactions
at the Cornell University Medical Center in New York, decided to slow

down his film of infant^mother pairs and examine it frame by frame. He

soon discovered finely synchronized behavioral patterns that no one in
the field had quite attended to before. He saw, for example, one mother^
baby pair after another play their own variation of an intricate dance of

threat and delight he dubbed the suspense game. “I’m gonna get ya!” says

the mother, over and over again, with rising intensity, her fingers perhaps
marching up the baby’s chest matching the rhythm of her words. “I’m
gonna get ya!” she says with more threat, her eyes gleaming as she looks

into his, her head shaking slightly from side to side. “Here I come. I’m
gonna get you, yes lam!” as her hands form a threatening semicircle over
his chest. And throughout all this, the baby coos and giggles with ever

more glee.^

The microanalysis of film and video of infants and mothers engaged in

such ordinary, everyday interactions was soon picked up by others
most notably Psychologist Edward Tronick, now chief of the Child
Development Unit at Children’s Hospital in Boston; Hanus and Mech-

thild Papousek at the Max Planck Institute for Psychiatry in Munich;

Berry Brazelton at the Children s Hospital in Boston; and Beatrice

Beebe, at Yeshiva University in New York. The world of intricate com'

munication they discovered in what had once been dismissed as rather
charming but meaningless babble turned out to be as rich and unex'
peered as the universe of “cavorting beasties” that Anton van Leeuwen'
hoek observed when he put an ordinary drop of water under his newly
invented microscope.
According to these researchers, infants and mothers continually en'
gage in lengthy reciprocal exchanges, with the mother matching the
infant’s intensity and tempo, adding variations to elaborate the commu'
nication, and attending to the infant’s cues, such as the desire to initiate

an exchange with a coo or a gaze or the wish to terminate it when he

feels he’s had enough, often hy turning away.
me interacting with a six'week'old infant,”
“I have videotape of
Tronick says. “When I purposefully get too close, the baby narrows his

eyes, looks off to the side and sort of purses his mouth. He won’t look at

me. I get even closer and the bahy turns his head away. When I move
 —
348 Nature 'Nurture Erupts Anew

hack, he starts to peek at me. I keep talking to him and move back a lit'

tie farther. He makes eye^tO'eye contact and, since he can’t yet fully
smile, he gives me a sort of half smile. Then, when come closer again, I

he looks off to the side. The infant is not only responding, he is actively

behaving in a way to modify what I do. When am too intrusive and

I fail

to respond to his signals, he turns away. When am sensitive to

I his

needs, I am rewarded by his looking at me, smiling, and vocalizing

behavior that would encourage me.”®
According to psychoanalyst Louis Sander, like Ainsworth a pioneer in

the naturalistic observation of mothers and infants, one of the crucial

results of such nicely synchronized exchanges is that the baby learns to
regulate himself — biologically at first, in the form of regular patterns of
sleeping, eating, and eliminating, but gradually on a psychological and
social level as well. He comes to know when he needs to reduce or
increase his arousal and how to achieve it.^ It has since been discovered
that mothers who are responsive when their babies signal “Enough!,”
have children who are able to sustain longer periods of attention, an
important capacity for cognitive development.^^
Stern found that, as the infant gets older, the mother introduces into
their dialogue new complications. She continues to match the baby’s
feeling level, but now adds color and depth by doing it through a differ-

ent mode of expression:

• A nine-month-old girl becomes very excited about a toy

and reaches for it. As she grabs it, she lets out an exuberant
“aaah!” and looks at her mother. Her mother looks back,
scrunches up her shoulders and performs a shimmy
terrific

with her upper body, like a go-go dancer. The shimmy lasts
only about as long as her daughter’s “aaah!” but is equally
excited, joyful, and intense.
• A nine-month-old boy bangs his hand on a soft toy, at first

in some anger but gradually with pleasure, exuberance, and hu-

mor. He sets up a steady rhythm. Mother falls into his rhythm
and says, “kaaaaa-bam, kaaaaa-bam,” the “bam” falling on the
stroke and the “kaaaaa” riding with the preparatory upswing
and the suspenseful holding of his arm aloft before it falls.

Through such creative echoes the mother not only reinforces the
hahy’s behavior, but she communes with him, which helps define for him
 —

Astonishing Attunements 349

what he is all about, builds their relationship, and shows him that his

experience can be understood and shared with another. Perhaps, most

important, such communion gives the infant a sense of being appreci-

ated, validated, and approved, which seems to be such a vital need at this

time of life. Stern found attunements like these embedded so subtly in

maternal behavior that they often go unnoticed. But he believed that
they account for the impression we form of how well a mother and her
baby are relating.

The burst of scientific knowledge about the mother-baby relationship

coming out of the universities in the 1970s established Winnicott’s pre-

science in point after point and further validated Ainsworth’s emphasis

on maternal attunement and “mutual delight.’’ It was now found that the
baby not only looks to the mother to participate in his play, to echo his
feelings, and thus to have his experience affirmed and understood, he

also depends on her to take him to levels of emotional experience that

he could not achieve on his own. The shared experience helps usher the
baby into social exchanges and the richness of relating to another per-
son. Even knowing what to feel in many ambiguous situations is devel-
oped through this primary relationship. If a baby is confronted with a

strange object, like a walking, flashing, beeping robot — is it dangerous?

is it fun? — he looks to the mother for guidance. If she smiles, the infant

will smile; if she looks frightened, he will become more apprehensive.

Ainsworth had studied attunement across a number of variables
feeding, face-to-face encounters, bodily contact, response to infant cry-
ing. With the microanalysis of video it now became possible to perceive

that such attunements take place in different dimensions as well, includ-

ing intensity, rhythm, and duration. Most of the time mothers seem
mainly interested in participating in a shared experience with the baby.
They explained their behavior to Stern with phrases such as “I just
wanted to join in.”^^ At other times they are trying to slip into the

infant’s experience in order to adjust his emotional state. But how they

do it — the specific techniques revealed in the video analyses — is as

invisible to them as the numerous muscular contractions and coordina-

tions that enable them to drive a car or brush their teeth.

The power and importance of this shared experience is illustrated by

the famous “still-face” experiment. Tronick asked mothers to make

their

babies. By
faces expressionless in the middle of an interaction with their

three months of age, the infants typically reacted to their mothers blank

expressions with distress.

350 Nature 'Nurture Erupts Anew

Babies notice milder forms of misattunements as well Stern asked

mothers to purposely misattune to their babies by responding to them
with an intensity that over- or undershot the intensity the baby was dis-
playing in its behavior. In one example a mother is watching her nine-
month-old play excitedly with a new toy. She jiggles his bottom,
matching with her jiggle the level of his excitement. She does this again
and again, and each time the baby responds to her jiggles with no
response at all. He just goes on with what he was doing. But when asked
by the researchers to jiggle somewhat more slowly than would match
his emotional state, the baby stops playing and turns to give her a per-
turbed look. Every time the procedure is repeated, with the mother over-
shooting or undershooting his intensity, the baby responds this way.
When she then gives him a nicely matched jiggle, he goes on as before,

doing what he was doing. Babies also notice verbal misattunements,

and they seem to be equally perturbing to them. “What is at stake here,”

Stern wrote, “is nothing less than the shape of and extent of the share-
able inner universe.”

The revolution in empirical knowledge about the infant came down

heavily in support of Bowlby and Ainsworth. Researchers, like Jay Belsky,
applying the new techniques, soon found that well-synchronized mother-
infant interactions predict secure attachment. This work enriched the
attachment idea and gave further credence both to the relevance of
the Strange Situation and to Bowlby ’s and Ainsworth’s emphasis on the
quality of care in the first year.

The findings added a powerful new dimension to the day-care debate.

For given the complex world of mutual influence in which the infant’s
psychology evolves, it would seem critical, to use Bronfenbrenner’s
words, for the baby to have plenty of time each day with “somebody
who’s crazy about him.” And, of course, the new findings fortified the
whole environmentalist position, providing data that were as dramatic in
their way about parental influence as the data the behavior geneticists
had presented regarding the influence of genes. The evidence clearly sug-
gested that a mother who fails to attune across a hroad emotional spec-
trum will not only adversely aff(^ct the baby’s emotional and social
development, but its initial biological regulation as well, creating in

some children qualities similar to what Chess saw in the difficult child,

qualities that might be mistaken for inherited temperament.

 Astonishing Attunements 351

Winnicott had long held that it was important for the mother to give

the baby the space to be with himself when he is not making demands on

her. Out of this quiet state, which Winnicott called “going on being,”

emerges not only the child’s capacity to be alone but authentic and spon^

taneous expressions of his self. “It is only when alone . .

.” Winnicott

wrote, meaning, usually, alone in mother’s presence, that the infant can

discover his own personal life.”^^ Mothers who intrude on the baby
because they are more attuned to their own needs than his cause him to

he more attuned to her needs, too, and thus more inclined to develop a
false, compliant self. Stern’s research appeared to validate
this view.

Stern cites the case of Stevie, whose mother continually forced face'
tO'face interaction with him no matter how desperately he turned
away.

Sometimes she chased him from side to side with her face so insistently
that he ended up crying. Whatever the cause of her behavior she
may have been hostile, she may have had a neurotic need for control,
she may have been deficient in her sensitivity to the feelings of others, or

she may have been hypersensitive to rejection, such that each time the

baby turned away she took it personally — Stevie’s developing sense of

relatedness was impaired. ‘‘From observation of many infants such as

Stevie,” Stern wrote, “it is clear that they generalize their experience, so

that they are relatively overavoidant with new persons.”^^ He believed

that children like Stevie gradually discover that they are able to
be most

fully themselves when they are alone, leading to a welbfunctioning but

relatively isolated life, what might be called an avoidant life.*

Molly’s mother was controlling in a different way. She constantly told

Molly how to play with toys (“Shake it up and down — don’t roll it on

the floor”), and, in effect, rode roughshod over Molly’s natural

rhythms
the baby was
of interest and excitement. Her exertion of power over
knot of
such that Stern and his colleagues often experienced a tightening
solution was
rage in their stomachs as they watched the tapes. Molly’s
compliance: “Instead of actively avoiding or opposing these intrusions.
space. She
Stern wrote, “she became one of those enigmatic gazers into
and her
could stare through you, her eyes focused somewhere at infinity
facial expressions opaque enough to be just
uninterpretable, and at the

same time ... by and large do what she was invited or told to do.

based on impingement and intru-

*This suggests an alternate avenue to avoidant attachment, one
sion rather than insensitivity to attachment needs;
although it seems plausible that the latter would
accompany the former. (See Chapter 25.)
352 Nature 'Nurture Erupts Anew

Watching her over the months was like watching her selhregulation of
excitement slip away.”^^
When the researchers saw Molly again at age three, she struck them as

emotionally flat; and Stern offered a troubling but hopeful observation:

“At some point in her development,” he predicted, “anger, oppositionah
ism, hostility, and the like will be sorely needed to rescue her.”^^
The special predicament of a child with a depressed mother is ih
lustrated by a baby named Susie whose solution to dealing with her
mother’s lack of responsiveness was to lean all the more heavily on her
own natural spunk and persistence. Susie would exert herself strenuously
to make her mother come alive, and because she occasionally succeeded,
she continued in this effort year after year. To her, being with an inti'

mate partner is not a matter of dread, as it is for Stevie, or a cause for

tuning out, as it is for Molly, but rather an impetus to hard work, in order
to bring the other person to life. By three years of age. Stern wrote, “She
is already a ‘Miss Sparkle Plenty’ and precociously charming.
As the child acquires language, his early sense of self becomes solidi'
fied by words. In cases like those of Stevie, Molly, and Susie, language
helps solidify a false self. Phrases like

“Aren’t you being gentle with the teddy bear! Sally’s always so
gentle.” Or “Isn’t this exciting! We’re having such a wonder^
ful time.” Or “that thing is not so interesting, is it? But look at
this one”^^

tell the child what to feel, what to be interested in, what inner modes are
acceptable, reinforcing previous manipulative attunements. They build a
verbal foundation for the false self, away of thinking about oneself that
becomes locked in and taken for granted. Other ideas such as “I love to
explore visual patterns,” or “1 love speed and action,” which may better
express who the baby, and later the adult, truly is, remain inchoate and
therefore more difficult to access.

By the time the child is six months old, the mother’s attunement with
the baby has begun to educate him about the emotional states that are
acceptable to share with other people. The mother’s fears, desires, ideas
about propriety, feelings of shame, and personal fantasies all come into
play in determining which aspects of her baby’s being will be reinforced
 Astonishing Attunements 353

and which will, in a sense, The mother who only tunes in

be banished.
to a certain range of the child’s inner life —
who only values his arousal
and engagement but not his passivity or who relates to him fully only
when he’s down, not when he’s up — tells the child what part of himself

to bring forward in interpersonal experience.

Consider the subtlety of such influence: When ten-month^old Sam
became excited and flapped his arms at his mother, she responded with,

“Yes, honey.” The intensity of her response consistently fell just short of

Sam’s own intensity, which surprised Stern’s team because the mother
was otherwise a dynamic and vivacious person. Asked about her level of
response, she confessed that she saw Sam as being too passive and
was
as she
afraid he’d turn out like his father. She hoped that by behaving

did, Sam would rely on his own fire, not hers. Ironically, she was proba-
bly having just the opposite effect. Sam would likely come to feel that his

dynamic side, never adequately responded to, was not something he

could share with others, and it would probably atrophy as a result.

Such findings echoed elements of attachment research. As Bowlby

wrote of avoidant children, who snubbed their mothers upon reunion in
the Strange Situation no matter how pitifully they d searched for her
when she was gone: “Already by the age of 12 months, . .
.
[they] no long-

er express to their mothers one of their deepest emotions or the equally

deep-seated desire for comfort and reassurance that accompanies it.”^^
The dynamics that Stern was able to capture in his video analyses
brought to empirical life issues that psychoanalysts had speculated about
for years. Can we trace certain obdurate, difficult, and strangely self-

defeating behavioral styles to a time beyond memory when we first

learned how to relate? In the past such questions could only be answered

with shrewd guesses about the early experiences of adult patients.

Now
there was more tangible evidence:

The baby takes a doll [Stern wrote] and starts to chew on its

shoes with gusto. The mother makes a number of attunements

to his expressions of pieasure, enough so that she is seen as a

mutualiy ratified member of the ongoing experience. This

membership gives her entree to take the doll from the baby.

Once she has the doll she hugs it, in a way that breaks the

previously established chewing experience. The baby is left

hanging.
354 Nature'Nurture Erupts Anew

The mother is essentially saying, “No, don’t chew on the doll’s shoes!”

and, also, “Dolls should be hugged.” But instead of openly prohibiting or

teaching. Stern wrote, “she slips inside the infant’s experience by way of
attunement and then steals the affective experience away from the
child.”^^ Such manipulative misattunements take many forms and are.
Stern argued, the likely origin of later lying, evasions, and secrets. The
child, and later the adult, comes to feel that if people are allowed access

to his true inner experience, they will be able to manipulate it, distort it,

undo it. Only by freezing them out can he keep his inner experience

unspoiled.
Stern points out, however, that the mere fact that mothers will fre^

quently be out of tune with their babies, even for selfish, foolish, or neu'
rotic reasons, does not necessarily bode ill for the child. No parent is

perfect, and as long as the poor attunement is not dominant, it will be

useful in teaching the child the facts of intimate life: that attunement
unlocks the doors that separate people, that it can enrich one’s life, but
that it has certain builtdn hazards against which one will, at times, have
to guard.
What, meanwhile, are we to conclude about the relationship
between imperfect maternal attunements and anxious attachment? If

Sam, for example, becomes passive like his father, and if his mother is

disappointed in him for this, does that necessarily mean that they will
be insecurely attached? Obviously not. Such limitations and such
wounds seem too common, perhaps even bordering on the universal, to
serve as a marker for disturbed attachment. My guess is that the hurts
have to reach a certain critical mass or be concentrated in attachment'
sensitive areas (such as the parental response to the child’s caretaking
needs) before the child experiences a fundamental doubt having crept
into his relationship with his mother, such that he is no longer certain
of her love and availability.
Stern’s clinical examples reveal the hidden tragedies of misattune'
ment, and by doing so, they suggest the subtle processes by which anx'
ious attachment may come about. They also suggest avenues for future

research in this realm. In a pilot study that was influenced in part by his
work, Wendy Haft and Arietta Slade found “a pattern of selective misat-
tunements” among mothers with anxious babies. Those whose babies
tended to he ambivalently attached to them misattuned to a broad speC'
trum of the baby’s emotional expression, hut were extremely attentive
 Astonishing Attunements 355

training
and well attuned when the baby was afraid. They were, in effect,
to
him to see fear as a primary means of achieving a sense of relatedness
attached
another person. Mothers whose babies tended to be avoidantly
to them, on the other hand, were most likely to
misattune to the infant

when he was expressing a negative feeling, especially toward them,

or

when he was seeking comfort or reassurance. Such mothers excelled,

however, in attuning to their baby’s exuberance, especially
when he was
mastering some new toy or game.^° This is consistent
with Karin GrosS'
of many
mann’s findings, first referred to in Chapter 18, on the behavior
of the mothers she observed with avoidant children:

reports],
Ifthe [avoidantly attached] infant played well [she
the parent attended joyfully and tried to join in. Many
times,

though, the parent’s play interfered with, dominated,

or frus-

trated the infant. If, subsequently, the infant

showed signs of
distress, frustration, or boredom, the parent retreated, oh'

served from a distance, and waited until the child

had pulled

himself together again and had overcome his negative

emo'

tions by himself.
Thus, the avoidantly attached infants experienced friendly
themselves,
attention mostly when they were busy playing by
and when they showed positive emotions; but they were left
alone when they displayed negative emotions. The securely
and
attached infants, in contrast, experienced acceptance
consolation when they displayed negative emotions, but
when
parent did not
they did not signal helplessness during play, the
interfere and accepted his interest in the world
around him.

Findings like these support the idea that what we have been calling

“parental rejection” may often be a subtle process, one that the parent
also suggest that
himself may easily bar from awareness. Such findings
role in the development ot what
attachment themes play an important
Stern called the child s shareable inner universe.

Winnicott saw the early infant period as a time

when the mother is
largely selfless in her relationship with
the baby, that, from her last

she becomes increasingly absorbed with the

baby
trimester of pregnancy,
born she utterly pre-
needs, so that by the time the baby is is
and its
356 Nature-'Nurture Erupts Anew

occupied with him (“primary maternal preoccupation,”^^ he called it;

“enslaved” was Bowlby’s word). But gradually, he said, the mother awak'
ens from her insane and exhausting devotion, and seeks to reclaim her
life. In tandem with the baby’s growing capacities, she stops shaping the
world in perfect accordance with his wishes. The child loses his sense of
omnipotence, learns that he is not the center of the world, and begins to
see others as having an independent existence. This painful lesson is

softened somewhat by the child’s own striving toward separation and the
excitement he finds in his new autonomy.
To ease the transition, the child typically becomes attached to what
Winnicott called a “transitional object.” This is often a blanket or a bear,
which he now carries with him everywhere, talks to, and loves. Winni'
cott saw great psychological importance in the transitional object and
believed that parents must not only treat it with care but collude with
the baby’s emotional investment in it and his fantasies about it.^^

With his term “good^enough mother,”^"^ Winnicott expressed the idea

that no mother can or need be perfect and, indeed, mustn’t be perfect if
the child is to abandon his grandiosity, not be a lifelong nuisance, and
become his own person. (How much imperfection can be tolerated and
at what point will, of course, vary from one mother-child pair to the
next.) The phrase is now famous and has been adopted by people who
play down the importance of maternal attunement Sandra Scarr uses it —
to support the idea of early day care. Yet the functions Winnicott’s good-
enough mother serves are plainly immense. Winnicott makes it clear
that built into mothering is a strong tendency to suppress her own needs
in favor of the baby’s. The new infant research of Stern and others sug-
gested that mothers do seem to be naturally drawn to provide just the
sort of care he described. When mothers fail severely enough, the result
is often anxious attachment.

Quality of attachment, as important as it is to the child’s self-esteem,

access to his feelings, and future relationships, is of course not the only
aspect of the parent-child bond. Parents are also playmates and teachers,
and these aspects of their caregiving can go well at times even if the
attachment is insecure. But it is the attachment side of the relationship
that is most emotionally charged for parents and the most difficult to
control. Many mothers cannot master their anger, for instance, although
they try. They cannot keep from acting out their anxiety. They cannot
 Astonishing Attunements 357

stop themselves from being controlling or hostile, even though they love
their children dearly. One of the unspoken trials of parenthood is
that it

brings out elements of our psychology that seem to take over despite our

best intentions, often compelling us to behave in ways with our

children

that replicatewhat we experienced with our parents. It remained for

another branch of attachment studies to explore this phenomenon.

 J
I

•w.

I
 PART V

The Legacy of Attachment

IN Adult Life
 \

1
 24
The Residue oe Our Parents:
Passing on Insecure Attachment

Bowlby believed that when a baby is bom, it evokes feelings in the pat'

ents that are as profound as those of a young child for its mother or as the

passions of new lovers. But, like other analysts, he recognized that,

inevitably, those feelings are not all positive. Along with overpowering
feelings of love, come resentments, hatreds, and anxieties as well, which
are usually associated in some way with feelings first felt in the distant

past. Thus, wrote Bowlby, a mother who is obsessed with the fear that

may unconsciously harbor an impulse to kill it, and.

her baby will die

adopting the same solution she adopted in childhood perhaps

in regard to her death wishes against her own mother,
strug'

gles endlessly and fruitlessly to stave off dangers from else^

w^here — accidents, illnesses, the carelessness of neighbors.

The father who resents the baby’s monopoly of his wife and

insists that her attentions are bad for it is unaware that he is

motivated by the same kind of jealousy that he experienced in

childhood when a younger sibling was born. The same is true
of the mother impelled to possess her child s love who, by her
362 The Legacy of Attachment in Adult Life

endless self-sacrifice, tries to ensure that her child is given no

excuse for any feelings other than those of love and gratitude.
This mother, who at first sight appears so loving, inevitably

creates great resentment in her child by her demands for his

love, and equally great guilt in him through her claims to be

so good a mother that no sentiment hut gratitude is justified.^

Like other analysts, Bowlby did not believe that the negative feelings
parents have toward their children are necessarily harmful. They exist in
some measure in virtually every parent. When my son was three months
old, I noticed one afternoon that he seemed to be more in love with his
mother than with me. I found myself becoming jealous when, while
holding him in my lap, he constantly looked off to the side to gaze and
smile at her. It rankled me that he did this, and it rankled me that it ran-
kled me — for at some level I felt that my jealous response was infantile
and that it might spoil the love between us. But the important question
from Bowlby s point of view was. Could I tolerate such feelings in myself
or were they so threatening that 1 would have to banish them? For it’s in
the banishment that the feelings become treacherous.
A father might find himself so powerfully jealous of his son that he
needs to disavow it, for any number of reasons. Most likely, he felt

deprived of love himself as a child and has created a life for himself where
deprivation remains a constant, if unseen, issue. In some respect he has
continued to feel like a baby, hungry for attention, and in this way comes
into direct, if unconscious, competition with his son. But these feelings
seem shameful and threaten his sense of competent adulthood. If he has
never attended to this part of his psychology and it remains a threatening
stranger, the situation with his wife and son puts him in an intolerable
emotional turmoil, requiring the mobilization of his defenses. But when
feelings like jealousy or resentment are disavowed and defended against
(perhaps, in this case, by the father’s being scrupulously supportive —
even a cheerleader — whenever the rotten kid turns to beam at his mom),
they do, in the end, emerge in some way. The father might become subtly
unavailable, unconsciously rejecting the little hoy and wanting to be pur-
sued. Or he might become impatient with some real or imagined aspect
of his son’s personality or innate capacities; so that now, instead of feeling
jealous, he finds himself angered by signs of slowness, dependency, or lack
of assertiveness in his son. However his resentment emerges it will very
likely have the flavor ot his own childhood home.
 The Residue of Our Parents 363

The father may have dreamt of a fresh start when he became a parent

and determined never to fall into the traps that snared his dad. But as

Selma Fraiberg wrote, “In every nursery there are ghosts, such that a

parent and his child may moment or a

find themselves re-enacting a
^ As criticism
scene from another time with another set of characters.
and judgmentalness leak or pour out of him, the resentful father may feel
as if he has become trapped inside his own father’s skin.

Various efforts have been made over the years to evaluate the extent to

which a parent’s attachment style is transmitted to the child. That is, if

the father was insecurely attached to his parents, will his son have an
insecure attachment to him? In one study, mothers from Sroufes sample
were interviewed about their attachment histories. “Scales were con-
structed,” Sroufe says, “to assess the amounts of chaos and uncertainty
in

the child care, the amount of nurturance that was available, and whether
this adult portrayedan image of her own mother as confident and com-
petent in the maternal role. If she had that view of her, she turned out to
be sensitive with her infant and her infant was securely attached.”^
Indeed, a clinically trained judge was able, in thirty-one of thirty-six
cases, to correctly predict whether the mother s child was
securely or
In
anxiously attached based on the information from these interviews.
another study mothers were assessed to determine how accepted or
rejected they felt by their own parents."^ The acceptance-rejection factor

just about perfectly divided the mothers of secure and anxious babies.
But the most influential, in-depth, and complex study of intergenera-
tional factors in attachment was done in Berkeley under the direction of
Mary Main.
TheBerkeley Adult Attachment Interview was the final element of
the effort to assess the internal working models of parents
and their six-

year-olds discussed in Chapter 16. Devised by Carol George,

Nancy
Kaplan, and Mary Main and analyzed by Main and Ruth
Goldwyn, it
early
posed a series of ever more probing questions about each parent’s
relationships. It asked for a detailed account of his family situation,
descriptions of mother and father, what he did as a child when he was
upset or ill, experiences of separation or loss, feelings of rejection, and

feelings about the parent’s relationship with his own six-year-old. The
interview, which lasted between thirty and ninety minutes, sought not
only to discover what the adult’s early attachment experiences were like,
more important, how he thought and felt about intimate
but, perhaps
364 The Legacy of Attachment in Adult Life

attachments now — that is, how he represented them in his mind, what
his internal model was like, and whether he could give himself the free^

dom to access painful memories in this realm and open them up for
inspection or whether the whole business was too distressing and needed
to be warded off. To that end, transcripts of each adult responding to
questions about early and current attachments and what they mean to
him were carefully analyzed for such issues as coherency, quality of mem-
ory, and anger; and these variables — not what happened or didn’t hap-
pen in the past — were what Main ultimately used to determine adult
attachment status. More than an interview, it was a psychological assess-
ment, almost like a Strange Situation for adults.
Main was able to identify three major patterns — or, as she puts it,

states of mind with respect to attachment — patterns of adult attachment

that directly paralleled Ainsworth’s childhood attachment categories.
Adults categorized by Main as “secure-autonomous” presented a believ-
able picture of their parents, often using what she called “fresh and lively
language.” Usually, but not always, they presented at least one of their
parents as having provided them with a secure base in childhood, but the
childhoods they portrayed were not trouble-free. These adults were able
to be objective about their parents’ positive and negative qualities and
could offer convincing examples to illustrate their points. Their commu-
nication style was clear and direct, showing that they understood what
the interviewer was asking, and they gave an appropriate amount of
information to provide an adequate answer. They seemed to be comfort-
able discussing attachment themes and able to reflect on themselves and
their relationships.

If the secure adults had unhappy attachment histories, they seemed to

have understood and worked them through, at least to the extent that
they could speak about them without getting into a stew, often demon-
strating insight into the effects their negative experiences had had on
them as well as some forgiveness or understanding of the parent’s behav-
ior. One mother who had been rejected by her parents as a child laughed
when asked to elaborate and said, “How many hours do you have?” She
then continued, “Okay, well, to start with, my mother was not cheerful,

and 1 can tell you right now, the reason was that she was over-worked.”^
The secure parents showed little evidence of self-deception, seemed will-
ing to depend on others, had a balanced view about their own role in
their relationships, recognized that they were similar to their parents in
various ways, not all of them positive, and generally seemed to accept the
 The Residue of Our Parents 365

importance of relationships in their lives. ^ The great majority of the chih

dren of these parents turned out to have been rated securely attached in
the Strange Situation five years earlier.

Adults categorized as “dismissing of attachment” seemed unable or

unwilling to take attachment issues seriously. They answered questions

in a guarded way, without much elaboration, and often had trouble

remembering their childhoods. They seemed to dislike and distrust look'

ing inward. Some exhibited an underlying animosity that seemed to

imply: “Why are you asking me to dredge up this stuff?” or “The whole
point of this interview is stupid!” The dismissing adults spoke vaguely

about their parents, frequently describing them in idealized terms. But

when pressed for incidents that might illustrate such descriptions, their
memories contradicted their assessments, as negative facts leaked into

their narratives. Thus, one parent called his mother “nice” but eventU'
ally revealed that she was often drunk and swore at him. When asked if

that bothered him, he replied, “Not at all. That’s what made me the

strong person I am today. I’m not like those people at work who have to

hold [each other’s] hands before making a decision.”^

This stalwart, anti'Sniveling response was typical of the way dismiss^
ing subjects played down the affect of early hurts or embraced them as

having built their character. Another dismissing father described his

mother as “loving,” “caring,” “the world’s most affectionate person,”
“invariably available to her children,” “an institution.” But pressed for
details, he could not recall a single instance of his mother’s warmth or

nurturance.^ A third father described his mother as “excellent” and his

relationship with her as “fine.” Yet when probed on the point, he eventU'

ally recalled having broken his hand as a child and being too afraid to

tell her about it.^ A dismissing mother said she had never been upset as a
child.Never? “No, because nobody died.” As the interview progressed,
however, she revealed a number of painful facts. Her father had been
upset that she had not been a boy and she’d once overheard him say that

the family would not have lost their farm if they had had a son. She also

reported that she had often been left alone as a child and typically came

home to an empty house. Finally, it emerged that her father was an alco-

he beat her mother, and that one of those beatings, delivered

holic, that

when her mother was pregnant with her, caused her to be born with a
crippled foot that required surgery.
The dismissing adults could equally have been called avoidant. They
were reminiscent of the children Bowlby had studied decades earlier who
366 The Legacy of Attachment in Adult Life

had suffered terrible separations from their families at an early age and
then became indifferent to affection or hurt. And they were reminiscent
of the avoidant children in the Strange Situation who turned away from
their mothers on reunion, as if their mother’s return was no big deal,

despite in many cases their having cried and searched for her desperately
when she was gone. Like the avoidant children, the dismissing adults
seemed to have warded off crucial parts of their feeling world through
various defensive processes, so that they no longer felt the pain of re^
jection or the longing for love. Approximately three quarters of the
children of these dismissing adults turn out to be avoidantly attached to
them.
A third group of parents, whom Main classified as “pre^occupied with
early attachments,” seemed like the ambivalent child grown up. Often
they spoke as if the feelings of hurt and anger they had had as children
were as alive in them today as they had been twenty or thirty years
before. The childhoods they described were often characterized by in^
tense efforts to please their parents, considerable anger and disappoint'
ment, and by role reversals in which the child had tried to parent the
adult. But these memories were expressed in a confused and incoherent
manner, as if they had never been able to get a grip on what happened
to them and integrate it into a comprehensible picture. They seemed still

so enmeshed with their parents that infantile feelings flooded and be'
wildered them as they recalled the past. At times they would launch into
an angry diatribe, seeming to lose themselves in feelings of resentment
and to forget that they were participating in a formal interview with a
stranger. (“1 thought here 1 am getting married and she’s not bloody
prepared to give. I thought every mother would sort of want to give the
best — but not her!”)^* Unable to see the forest for the trees, these pre'

occupied parents seemed to have no sense of their own role in any of

their relationship difficulties.'^ The vast majority of their children were
ambivalently attached to them.*
In subsequent studies, similar correlations were found. Meanwhile,
several researchers, in the United States, England, and Canada, have
turned the research around and studied not parents of six'yeat'olds but
expectant parents. Peter Fonagy, working in London with Howard and
«

*Main also identified a fourth j^roup of adults all of whom

had suffered a profound early loss, like the
death of a parent, that they seemed unable to understand; in some cases they seemed unable to inte-
grate the knowledge that the long-lost parent was truly dead. Their relationships with their children
were the most distressed, usually falling into the disorganized category.
 —
The Residue of Our Parents 367

Miriam Steele, was able to correctly predict infant Strange Situation

classifications in 75 percent of the cases —
a figure in keeping with re^

search done elsewhere — based on interviews with the mothers prior to

giving birthd^
Arietta Slade, who’s begun analyzing data from a similar study, has
found many differences in the way women face the emotional upheavals

of pregnancy. Women classified as secure in the adult interview are for

the most part more open psychologically to becoming a mother. They

can accept that the child will be dependent on them and that they
themselves will experience pain and distress. “They expressed anxiety
about the changes they were facing, they complained about the physical
discomfort, they worried about their marriages, and they fretted over
their competence as parents.” But, Slade says, these concerns were hah

anced with a sense of hopefulness and joyful anticipation.”^"^ Indeed, the

secure women had already begun a relationship with their unborn babies,

had nicknames for them, and felt pleasure and fulfillment in imagining

what their babies were like.

On the other hand, women from the two anxious categories found it

difficult to derive pleasure from the knowledge of becoming a mother.

They either pushed aside any anxiety and insisted that all would he well,
or became overwrought with fears. Their descriptions of joy or excite-
ment were often constrained. And their fantasies of what their babies
would be like suggested that the unborn baby had already become a fig-
ure in old struggles. One pregnant woman, who had had terrible power
struggles withboth her parents, already imagined her baby as willful and
uncontrollable. Another, who had always felt distant from her own
mother, could not imagine anything about her unborn child. Discuss-

ing this, Slade told me, “It’s so interesting that a mother who’s felt con-
trolled by her mother her entire life is already preoccupied with whether

she can control her fetus. Or that a mother whose own mother pushed
her aside is already convinced that her child won’t have any needs, won’t
change her life, and will be autonomous early on. Chills go up and down
your spine when you hear these things.” (Anyone who has found himself

behaving as a parent in the very way he detested one of his own parents

for behaving — fretful, intrusive, moralistic, overprotective, imperious

can certainly appreciate the dilemma of these expectant mothers.)

Slade has found a need for control typical among the anxious mothers
she has studied thus has sometimes showed up in rigid ideas about
far. It

whether they wanted to have a boy or a girl. “It’s another example of

368 The Legacy of Attachment in Adult Life

how these kids don’t have much autonomy, even before they’re born.
You should be a boy, you should do this, you should do that — the baby
doesn’t really have a chance to come out and be heard and be listened to

and be seen. The secure mothers have much more curiosity about who
the baby’s going to he. They bring the past into their expectations, too,
but they do it in more flexible ways.”^^

We don’t know from Main’s work exactly why the adults become
secure or insecure. We certainly don’t know how likely it is that a secure
adult, as measured by Main’s interview, was secure with his mother as

baby, although giving the AAI to people who went through the Strange
Situation in infancy promises some answers in that regard. We don’t
even know for sure how well Main is tapping into with the adult what
Ainsworth tapped into with the child. It should be remembered in this
regard that while the Strange Situation assessed a relationship, the AAI
focuses on an individual, apparently assessing some aspect of his psycho-
logical structure. Some believe the AAI is strictly cognitive. Judith
Crowell, the child psychiatrist at the State University of New York at
Stony Brook, who has given the AAI to adults who were assessed with
the Strange Situation as children (see Chapter 19), argues that the inter-
view measures how well and how much attachment issues were discussed
in the family as the person was growing up. “That’s not quite the same
thing” she says, “as measuring an attachment relationship. I think it’s a
closely parallel thing, in that people who can do that well are the people
who can help their child be secure, but I think it’s more cognitive and
not so rooted in the behavioral system.”
Like others, I tend to believe that the AAI taps more deeply into the
subject’s psychological dynamics, probably reflecting the cumulative his-
tory of the adult’s attachment-related experiences, or perhaps, for some,

the aspects of that history that are most activated, most dominant when
the interview is taking place. If it were possible to plot a graph charting
the quality of a child’s attachments by evaluating them every few years
from infancy onward, I assume we would see a varied tapestry, composed
of winding and changing paths, eventually coalescing into certain domi-
nant and subordinate themes. One of the problems with Main’s classifica-
tion system, however, is that it does not allow for such multiplicity. Even
some of her staunchest supporters would eventually take issue with this.

The adult classification system and the conclusions Main and others
have drawn from it have not gone uncontested. The interview is com-
plex and difficult to rate. Unlike the Strange Situation, which shows
 The Residue of Our Parents 369

over 90 percent reliability between raters, the reliability of the adult

interview is in the 70 to 80 percent range, still high but with many more
disagreements between raters over how to classify any one individual.
Thus, although the correlations between the parent and child classifica^

tions are impressive, the number of adult transcripts that seem fuzzy and
difficult to categorize call into question the rigorousness of the correla^
tions. Finally, no effort has been made to understand the significant

minority of parent'child pairs whose attachment classifications do not

match up. Meanwhile, not surprisingly, skeptics in the field find it hard
to believe that how a parent talks about the past can indicate anything
about the quality of his relationship with his child. Despite such objeC'
tions the Berkeley interview has rapidly become the most influential

piece of attachment research since Ainsworth’s Baltimore study.

Main has devoted much of the past ten years to poring over the tran-
scripts of adult interviews. She has become particularly obsessed with

modes of speech, seeing them as a gateway into that invisible realm that
both psychoanalysis and attachment theory have uniquely sought to
understand —
the internal representations, or models, we have of our-
selves, of significant others, and of our relationships. Does the person
speak in run-on sentences without taking a breath? Do his sentences run

out of steam at the end as if he’s lost faith in what he has to say or hopes

that his listener will finish the sentence for him? Does he speak in the

person (immediate), the second person (more removed), or the third

first

person (detached)? Does he use certain vague phrases that are supposed
to convey a lot of meaning? Is he prone to slips of the tongue (“1 died

when my father was ten”)?

“The more 1 look at the interviews,” Main told me, “the more I’m
astonished by the degree to which aspects of adults’ speech will correlate
precisely with their child’s score on some variable in the Strange
Situation, like resistance to the mother on reunion.”^® Ainsworth, mean-
while, has found that in certain cases she can predict not only the child’s
general attachment classification but its subcategory as well by studying

the parent’s use of language.

Main has attempted to understand how linguistic mannerisms and
quirks fit into a person’s entire psychological profile — not only the qual-

ity of his relationships, but his access to feeling and to memory, his abil-
ity to direct his attention toward or away from certain subjects, his

capacity to engage in the normal, cooperative give-and-take ot conver-

370 The Legacy of Attachment in Adult Life

sation. A sensitivity to metaphor has enabled her to make subtle assO'

ciations — for instance, that the avoidant infant’s shifting its attention
away from the mother on reunion is equivalent to the dismissing adult’s
changing the subject away from attachment themes. No one has toiled
with equal fervor to make such links.

“She absolutely believes that there are these connections,” Jude CasS'
idy says. “Annie Dillard talks about carving up the Panama Canal with a
teaspoon, and I think that’s what Mary Main would do. She believes that
there’s this kernel of truth underneath there, that the way parents talk
about their own childhood attachment relationships has got to be related
to the way their babies are attached to them. And if it’s covered up by
the Panama Canal and she has to dig away with a teaspoon until she
finds that kernel of truth, she’s going to keep digging. And if it takes her
five or six or seven years, that’s all right.”^^ (This portrait recalls Main’s
own description of Ainsworth in Chapter 11.)
Main’s work supports an assumption on which much of psychoanalytic
treatment is based: that being able to put feelings, especially unwanted
feelings, into words makes them available for review and transformation.
(Indeed, some of the more reflective, introspective secure adults sound
like people often do when they have experienced some years of intrO'
spective therapy.) To have this ability means, in effect, that your internal
model is still a “working” model — open, flexible, able to assimilate new
information. It means not only an ability to rethink the past but to rec-
ognize that people can be different and that their behavior doesn’t
always mean what we think it does. The criticism of a husband or wife,
for example, may feel like an intentional assault on one’s identity. If you
are able to attend to the feeling and put it into words
— “I feel worthless
and debased, convinced he wants to get rid of me, and insanely anxious
and needy” — then you are in a much better position than the person
who reflexively acts out such feelings by becoming depressed and over-
eating or becoming uncontrollably rageful. It gives you the space to rec-
ognize that your feelings may be somewhat irrational in the current
context. That Main found this capacity present in secure adults hut

absent in those who still seemed controlled by their early anxious attach-
ments helped convince her that without reflection there is little oppor-
tunity for change.
To the younger generation of attachment workers, like Roger Kobak,
such findings have been a source of inspiration. Kohak, who studied phi-
losophy and history as an undergraduate, went to divinity school after
 The Residue of Our Parents 371

that, took a year to study psychoanalytic theory at Yale, then began a

Ph.D. in community psychology, seemed in danger (as Ainsworth, a
strong admirer of his, once remarked) of becoming a perpetual student.
He turned to attachment research and completed his doctorate
finally

with Ainsworth “because it was the only place in academic psychology

where psychoanalytic ideas were being utilized. In Main’s work those

and integrated to an unprecedented extent.

ideas are being utilized
“She created a window,” Kobak says, “on these mental processes that
have been talked about for years but never studied empirically.”^^ He
believes that it will take some time before her work is understood and

assimilated by others, that it is as alien as Ainsworth’s findings initially

were, but that it is equally revolutionary. The excitement it has gener-

ated, meanwhile, has given attachment studies a third wind, one that is

pulling in analysts and other clinicians. Arietta Slade is one example.

“Hearing Mary Main present some of her preliminary findings in
1985,” Slade says, “was a pivotal experience for me. As a clinician, I got
very excited. It was the first time I’d heard these kinds of phenomena
addressed in an empirical fashion.” Slade, inspired to begin her own
attachment research, eventually took Main’s training workshop. “My

training in the Adult Attachment Interview,” she says, “has completely

changed the way I listen to how people talk. 1 hear language differently,

because the scoring system of the interview is focused on language — on

the kinds of words people use and the ways they back up what they say.
These subtle things can tell you a great deal about a patient’s defenses
and the way he constructs his reality.” And, of course, it also suggests
something about the patient’s past. “If I hear someone being really diS'

missing, I tell myself I better find out about rejection. I know this is

someone who’s probably been really shut out in some profound way,
someone who must have felt terribly alone.

of the
In 1956, as part of the British Psychoanalytic Society’s celebration
centenary of Freud’s birth, Bowlby gave a lecture on “Psychoanalysis and
Child Care” in which he stated:

Time and again we hear it said by teachers and others that a

child is suffering because of the attitude of one ot his parents,

usually the mother. We are told that she is overanxious or

down on the child, overpossessive or rejecting, and time and

again such comments are justified. But what the critics have
 —

372 The Legacy of Attachment in Adult Life

usually failed to take into account is the unconscious origin of

these unfavorable attitudes. As a result, all too often the err^

ing parents are subjected to a mixture of exhortation and crit'

icism, each as unhelpful as the other.^"^

Now, almost four decades later, attachment research has begun ad-
dressing itself to some of the psychological forces driving the parents of
anxious children. Although new to developmentalists, understanding
parental psychology and how it affects the child was hardly new to psy-
choanalysis. But the understanding had never been systematic before,
providing an overview of certain broad tendencies in parental psychol-
ogy and their counterparts in the child’s developing psychology.
Both Fonagy and Main believe that the most important quality distin-
guishing the secure from the anxious adults* is their capacity to under-
stand what makes themselves and others tick. They are better able to
recognize their own inner conflicts and to have a sense of why their par-
ents behaved as they did. One secure adult, for instance, when asked
how she thought her experiences as a child affected her behavior as a
parent, responded with a clear sense of her own inner workings. She
explained that she feared her tendency to fluctuate from one extreme to
the other — from being too withdrawn with her child, as her own mother
had been, to “smothering him with constant attention.”^^
Anxious adults either failed to have insights into themselves and their
parents or offered explanations that were platitudinous, self-deceptive, or
self-serving. Thus, one anxious mother, when asked about the relation-
ships between her parents, responded: “I am the apple of my father’s eye
and ... he does absolutely idolize me . . . and I think it’s amazing that my
mother has never been remotely jealous of me in any way at all!. . . She’s
just genuinely never held that against me at all and is fantastic!’’^^
Compare this to the deeper, more aware, less denying response of a
secure parent to a related question: “It was difficult for her to like me. He
adored me so much and treated me as someone so special. It would have
been super-human of her not to feel jealous.

Fonagy found that neither education, intelligence, social or economic

status, nor ethnic background have any relationship to this capacity to

*As with the labeling of children, the terms “secure” and “anxious” represent a convenient but inac-
curate shorthand. An adult may be secure in his attitude toward attachment issues hut be insecure
and exhibit all sorts of prominent anxieties — in other respects, and vice versa.
 The Residue of Our Parents 373

reflect openly and clearly on the inner state of oneself or others. With
studies like this, which do the indispensable work of redintegrating
attachment theory with its analytic roots, Fonagy, who is himself an ana'

lyst, is rapidly becoming one of the most prominent and important

attachment thinkers.
Winnicott had emphasized the importance of the caregiver in helping

the infant to understand his own feelings and impulses. In the beginning
the mother does the reflecting for the baby; gradually he takes it over

himself. The work of infant researchers like Stern seemed to corroborate

Winnicott on this point. Fonagy now related this to secure attachment.

For clearly only the parents who had the capacity to reflect on inner
states could provide this function for their children. He concluded that
the child’s confidence that its emotional and mental states would be
accurately assessed, clearly reflected, and appropriately responded to
were a central feature of secure attachment.
What happens when a parent lacks this capacity? First, of course, the

may fail to become familiar and comfortable with his inner world.
child
He may not develop the selFreflective ability that will enable him to
understand and work through at a later date faulty models of relating he
learned at home. (See Chapter 15.) Meanwhile, the parent, unable to be
self'teflective himself, is more likely to instill such faulty models. For he

will continue to be a slave to his own early patterns, persisting in strate'

gies and maneuvers with his own child that he first learned when dealing

with a rejecting or erratic parent.

Both Ainsworth and Main found the mother of the avoidant child to
be distant —
rejecting of the infant’s attachment needs, hostile to signs of

dependency, and disliking affectionate, face'tO'face physical contact,

especially when the baby desired it. Her aversion to nurturance would

seem to be a logical outgrowth of the neglect she probably experienced

when she herself was young. Needs and longings that were painfully
unmet have become a source of hurt and shame for her. Having cut bet'
self off from them, they make her angry, depressed, or disgusted when she
sees them in her child.

Unfortunately, unreflective parents don’t realize what is happening to

them. They may love their baby, they may talk about it in glowing terms,

they may be good playmates, they may readily compliment the child
when doing well, they may be diligently concerned with the baby’s
it is

feeding, sleeping, and physical comfort. But when attachment needs

arise, they find themselves being impinged upon by an intolerable
sense
374 The Legacy of Attachment in Adult Life

of threat. Perhaps because to be fully open to the baby’s emotional needs

is to become reacquainted with oneself as a baby, to reexperience the
pain of being totally dependent and desperately in love and yet being
shut out and feeling unwanted. People construct their defenses in order
to prevent being reengulfed by such feelings. But when one becomes a
parent, the buried, unresolved pain is shaken loose, the defensive wall

is breached, and new defensive efforts are required, which, in the case
of the dismissing parent, means keeping the baby and its needs at some
distance.
A dismissing mother may find a solution to the clash between the
baby’s emotional needs and her own — and whatever to she may guilt

unconsciously feel about this conflict — by seeking the haven of a belief

system that disdains introspection as naval gazing, favors early training in
autonomy, and stands guard against spoiling. Meanwhile, if, like so many
avoidant children, she is somewhat grandiose, the idea of a superior kid,
who has no needs, will feed her own superiority.

Jay Belsky found mothers of avoidant children to be intrusive with

their babies. Given the understanding of the avoidant mother that has
been developing since Ainsworth’s Baltimore study, his data have been
received with some skepticism.^^ Belsky acknowledges that intrusiveness
may be an artifact of the research process behavior that comes about —
when a certain type of mother finds herself being observed by a psychoh
ogistand wants to look the perfect parent. But Jude Cassidy has at'
tempted to make sense of it by stressing Main’s theme of unconscious
strategies.

“An avoidant mother,” Cassidy says, “wants to keep her baby from
getting too close. She feels really uncomfortable when the baby comes
over crying. She doesn’t like the baby crawling on her lap. So why would
she become intrusive? I think two things are going on. One is an uncon'
scious recognition that if you badger somebody repeatedly, they’ll go
away. It’s like this idea of dysfunctional marriages where the wife nags
and the husband pulls back. In some unconscious way the wife really
wants the husband to do just that. Another reason is that by coming at

her hahy repeatedly the mother can control the interaction and keep her
baby from initiating care seeking. It’s like you’re out on a date and you
don’t want the guy to kiss you, so you just talk constantly and don’t give
him a chance to get a kiss in. I think it’s the same with these mothers.
Here, play with this, no play with this — it’s directing the focus, and she
wants the focus away from attachment issues.
 The Residue of Our Parents 375

A certain type of ambivalent, or preoccupied, mother is haunted by

different demons. The inconsistency and incompetence that may charac'
terize her caregiving suggest a lack of involvement with her baby, but
something more complex is probably occurring. For example: A mother
who has never worked through her own ambivalent attachment has

probably been struggling all her life to find stable love. When she was a

child, she may have been pained by the competent, steady caring that
she saw friends’ parents give to them. As an adult she may be prone to a

nagging, uncontrollable jealousy in any close relationships where she

feels cause for doubt. She may want to love deeply and steadily, but it is

hard for her because she’s never been filled up enough with patient, reli'

able love to be in a position to give it. When she becomes a mother, such

unresolved, tormenting issues may play havoc with her emotional life.

This mother may care for her baby as much as any other mother, but
she finds her caregiving impaired by her own rankling needs, which make
it hard for her to be consistently available. If, unconsciously, she envies

the baby’s position as the one who is meant to be coddled and cared for,

that may further impede her ability to give of herself freely. If she harbors

an underlying wish that the child remain tightly enmeshed with her, that
he beome for her an attachment figure who will never separate or betray,
she may find herself interfering with her infant’s efforts toward auton-
omy. The existence of such inner conflicts seems to be corroborated by
the observation that some preoccupied mothers frequently intrude when
the baby is happily exploring on his own and push for interaction even
when the baby resists it.^^ A mother doesn’t do such things out of a lack

of love. But to see her child separate — become his own person — in

these small ways is painful to someone who feels deprived of love. It is to

feel the knife in the chest all over again, of being a small person, desper-
ately hungry for care, tormented by a mother who is careless about her

needs and who thus makes every separation a daily torment.

Cassidy believes that such conflicts best explain the apparent inconi'
petence of many mothers of ambivalent children. For if a mother uncon'
sciously wishes to keep a baby addicted to her, there is no better strategy

than being inconsistently available. Nothing makes a laboratory rat push

a pedal more furiously than an inconsistent reward. Without
being fully

aware of it, some preoccupied mothers may know this.

The immature, dependent, babyish behavior that Sroute observed in

some ambivalent children may, thus, represent the sort ot child his pat'
ent unconsciously wishes for, one who will not grow up and
separate
376 The Legacy of Attachment in Adult Life

from mom, who will always be dingily demonstrating his need for her,
and who will anxiously seek to appease her. The child, meanwhile, sus'

pended perpetually in his attachment anxieties may, if he gets stuck in

his mother’s orbit, grow into a similar sort of person, who constantly
seeks succor and devotion from another — much more than the average
person is likely to put up with. Without enough countervailing influ'

ence, he may eventually become a parent who repeats the pattern with
his own children in order to finally get the devotion he missed. This is

certainly as a good a theory as any as to why some parents become so

invested in keeping their children preoccupied with them. Throughout
all this, a lack of experience with selTreflection — and, in all likelihood,

an aversion to it — keeps these dynamics unconscious and therefore un^

resolvable.
As I write this, however, I am aware of how much is missing and how
much effort I am making to qualify my statements. For one thing, we are
talking here about a certain tendency which may exist among mothers
of ambivalent children; how prominent it is, we do not know. It is a
compelling explanation, it resonates with our understanding of human
nature, it seems to capture a broad trend in personality as we know it, but
it is too speculative, it is too simple, and it cannot account for everyone
in this category. Other children may sense an underlying depression in
the parent and feel that only their love and attention and good per^
formance can keep that parent from emotional collapse. This, too, could
generate an unhealthy pre-occupation, even in a child who has received
all the appropriate adoration, and may be carried through to adulthood
in the form of a haunting, perhaps unnameable, separation anxiety.
Main has found that many children classified with the most severe form
of anxious attachment, disorganized/disoriented, had mothers who had
failed to mourn a significant loss, like the death of a parent. But I suspect
that unmourned losses, perhaps of a lesser degree, are rampant among
adults with anxiously attached children and that they have a thousand
ways of passing on their own, often hidden, separation anxieties to their
kids, such that they grow up terrified of abandonment, fearful that they

will disintegrate if left alone, convinced that the people they love will
die if they are not always present to save them, or some other variation
on this theme.
It should he remembered, too, that anxious children do not always
replicate the anxious category of their parents. They are often anxious
but in different ways. The mismatch rate, some 30 to 40 percent in most
 The Residue of Our Parents 377

samples, may be related in part to the child’s temperament, hut it may

also be accounted for by other factors, like gender. Some research sug'

gests that the emotional systems of boys and girls are innately dissimilar

in certain respects. Boys and girls may also react differently to the

mother for psychological reasons. “Mightn’t it be the case, as I have

observed in our sample,” Slade asks, “that a preoccupied mother might
drive her son to withdraw and avoid? Mightn’t it also be the case that a

dismissing mother’s withdrawal could inspire clinging, neediness, and

ambivalence in her daughter?”^^
Also, in trying to show how a certain attachment experience with the
mother can reverberate throughout a child’s interpersonal existence, we
inevitably leave out many other complexities, including the complexities
of family life as a whole, and particularly the effect of the father. Whom,
for instance, does the child most identify with? (To raise again one of the
eminent missing questions of attachment theory.) What is the nature of

that identification? And what course does it take over time? If the child

is a boy who feels strongly identified with an emotionally withdrawn and

avoidant father, he may start out in an ambivalent relationship with his
mother but then move to a place where his relationship with her is much
like his father’s. The underlying ambivalent current may persist, but his

predominant style will now look avoidant-dismissive. All this will be fur-
ther complicated by the boy’s struggles and disappointments with his
father, the way his oedipal conflicts get worked out, the degree to which
his love, despite such disappointments, is able to thrive in these relation-

ships or with other family members or intimates, not to mention betray-

als, separations, unresolved losses, and so on. It will be complicated to

the point where trying to tell a paradigmatic story ot attachment-related

development looks more and more pointless, even more so when illus-

trating it with the child’s relationship with just one parent.

The efforts to understand what makes unresolved parents raise insecure

kids, or, put another way, why love falters, hurts, gets wasted or forgotten,

or becomes a chalice that can never be retrieved, underscores a subtext

that runs throughout this literature: Being a parent is ditticult. It is easy

to judge parents, and some are certainly awful, making little or no effort

to protect their children from themselves. But, as self-aware parents

know, the effort to shield our children from the damaged parts of our-
selves (the fears, the unmourned injuries, the unbearable longings) is

arduous. And it is constantly undermined by our entrenched dissocia-

378 The Legacy of Attachment in Adult Life

tions and indispensable armor, which have kept us as we are for so long.
We love our children, we dread harming them, and we dread losing them.
This is true for the educated and the uneducated, the rich and the poor,
those who have some understanding ot their own psychology and those
who do not. “The poor single mothers in our study,” Sroufe told me, “all

want the best for their kids. They maybe can’t do it. They may be so
beaten down by their histories and their circumstances that they’re doing
a terrible job, but I’ve never seen one that didn’t want to do it right.”
Doing it right is, of course, a major concern for Americans of all social
levels. Millions of books on child care are consumed every year, not to
mention magazine articles, television shows, and video cassettes. Parents
learn about how the baby is likely to behave during each new week or
month, what it needs to eat, what the variations are in physical and cog-
nitive development, how to sleep train it, how to toilet train it, and so
on down to the last detail. But, as useful as this is, none of it will help
parents do the one thing they most need to do — gain a deeper under^
standing of their own motivations, conflicts, and inner needs. In the selh
help literature directed at parents virtually no attention is paid to the
emotional upheavals that the parent is likely to face — the disturbing
return of long festering feelings, the sense of being driven to behave in
ways that one would rather not think about, the haunting sensation of
being inhabited by the ghost of one’s own mother or father as one tries to
relate to one’s child.

“It’s very American,” says Slade of the child care advice now avaih
able. “You just have to do X, Y, and Z, and your child will be fine.
Parents have a fantasy that they’re going to do it differently, that they’re
not going to repeat the past and they are going to overcome the obsta-
cles. And 1 think most people don’t exactly repeat the past and they do
certainly do things differently. But to really deal with these emotions
requires a willingness to let the past into your consciousness.”^^ That not
only means giving up the fantasy that we as parents are completely
unlike our parents, hut struggling with those very things in ourselves we
have always hated and have not wanted to admit to consciousness.
The problem we have as parents, then, is not usually a lack of love
or good intentions, hut more often an unwillingness to face who we
are. Obviously, this failure has other implications for adult psychology,
as well.
 25
Attachment in Adulthood:
The Secure Base vs.
The Desperate Child Within

In 1992, a prominent New York jurist, considered a likely nominee for

candi'
the United States Supreme Court and a potential gubernatorial
date, was arrested on charges of extortion and harassment.
Married for

thirtyYive years, he seemed a model family man, but he had been

having

an affair for many years with a younger woman. When she broke it
off,

he began a bizarre campaign to terrorize her with anonymous phone

calls, letters threatening embarrassing disclosures and harm to her daugh'

ter,and extortion demands. This went on long enough and became

egregious enough for the FBI to become involved and to
apprehend him.
to resign immediately from the state’s highest court,
He was compelled
an emotional collapse afterward, and was later sen'
seemed to suffer

tenced to fifteen months in prison.

a powerful, respected, and welh
To many people such behavior by
liked figure seems utterly inexplicable. How could a man who had so
much go berserk over the loss of a mistress? Why would a man known for

become so brutal and infantile? From

his good character and fair dealings

a distance, it is impossible to say anything about this particular judge.

But the tragedy of a successful, late middk'aged man ruining his life for a
380 The Legacy of Attachment in Adult Life

younger woman is very familiar in our politics, folklore, and literature. It

is, 1 think, evocative of attachment issues gone awry.

An imaginary example: An avoidant child grows up undervaluing
attachment relationships and experiences. He is intellectually gifted and
applies himself to school. He has rough edges at first but he learns to be
charming. Although everyone admires and likes him, no one feels par-

ticularly close to him. And because he is isolated emotionally, the depri'

vation that began in infancy persists into young adulthood. Starved for
affection, though unaware of it, he fantasizes about a great, alhencom'
passing love. Once perhaps, in his early twenties, he thought he found it

and was so bitterly wounded afterward that, without being conscious of

his decision, he only allowed himself to date women who were immedi'
ately enthralled by him. Unfortunately, he didn’t respect such women.
Gradually, he concluded that the love he once dreamed of did not exist;
in the end, most people are too flawed or bound to disappoint. Be that as
it may, it was good to have someone around. He made a “sensible” mar-
riage, much like his parents, to a woman whose presence gave him the
emotional support he needed to pursue his career, a need he would have
been embarrassed to admit to himself. He became a father, continued to
be charming and admired, rose rapidly in his profession, and gained con-
siderable power. Although sometimes taken aback by a streak of insensi-
tivity or dictatorialness, people valued him and took him seriously, but,

emotionally speaking, he did not take himself seriously. He fed himself

on a stimulating but lopsided diet of power, achievement, and praise,
meanwhile allowing his deeper self to starve. He was attached to his wife
and children and would have been deeply hurt to lose them. But at
home he was a star, a figure to be more venerated than intimate with,
and they were frightened of his capacity to turn cold when disappointed
with them.
Sometimes, in the morning, he awoke from dreams in which he was
disturbingly alone. When he went on business trips and was stuck in the
airport without any work and no admirer or person of influence to chat
with, he felt strangely isolated and detached, as if he didn’t have a wife
and family back home, as if he was the only man on earth without a
friend. This loneliness of the soul distressed him, and he pushed it out of
his mind.
Then, at the height of his powers, he met an impressive younger
woman who worshipped him and by whom he was sexually captivated.
 Attachment in Adulthood 381

Unlike the first time he fell in love, he was now starting from a much
more secure He was confident. His hated feelings of neediness,
position.

which subverted him when he fell in love as a young man, were kept at
bay because of the secure base provided by his marriage. And because
power is a great aphrodisiac, his new lover perceived him in the most
positive glow, almost like the adoring, unquestioning love a child has for
her father. What’s more, because he was loyal to his wife and had no
intention of disrupting his comfortable home life, the new relationship

was never tested in the cauldron of mundane daily existence or the rigors
of true commitment. He remained the impressive charming man of
power. She remained the wide-eyed acolyte. Neither ever knew the other

in three dimensions. In effect, there was no main course in this relation-

ship, only dessert.

But, like so much else in his life, the dessert became addictive, the

affairanother way to escape from his haunting loneliness, even, in this

case, while seeming to feed it. It was as if he had finally found true
love,

the total, accepting, maternal love he’d never had enough of. But it was
not true love. It was an exciting, tantalizing, delicious mirage in which
he invested very little of his genuine self. His partner, who wanted some-
thing more lasting, eventually outgrew her infatuation and left.
The affair lost, he was thrown back on his emptiness, confronted with
the terrible aloneness he an infant before he had perfected his
felt as

avoidant rituals. Indeed, he now seemed in some ways more ambivalent

than avoidant, consumed by jealousy, hunger, and desperation, like the
crazily dependent people who murder a lover over an infidelity
and get

immortalized in country and western songs. And, accustomed as he was

vengeful, and
to getting his way, he behaved like an infant, wretched,
full of narcissistic rage.

In 1970 Bowlby gave a lecture on “Self-reliance and Some Conditions

That Promote It,” which made it plain that he saw secure attachment in
the qual-
adulthood affecting not only the quality of one’s parenting but
ity of one’s entire emotional life. “Evidence is
accumulating, he said,

“that human beings of all ages are happiest and able to deploy their tal-

ents to best advantage when they are confident that, standing behind

them there are one or more trusted persons who will come to their aid

should difficulties arise. He called the trusted person an attachment

the
figure and, borrowing from Ainsworth, said that such a figure offered
382 The Legacy of Attachment in Adult Life

companion a secure base. He believed that the ties to the parent gradu^
ally weaken as the child gets older and that the secure base function is

slowly shifted to other figures, eventually resting fully on one’s mate.

Sifting through evidence from several disciplines, Bowlby argued that

selhreliance and healthy or mutual dependence were inexorably linked.
Studies of the American astronauts had found that these most selh
reliant and capable young men were “comfortable when dependence on
others is required’’ and that they were able to maintain their trust in

those they depended on even in trying circumstances. They had grown

up “in relatively small welhorganized communities, with considerable
family solidarity and strong identification with the father.’’^ Other stud-
ies done in the 1960s also suggested that emotionally healthy young peO'
pie, who were both selhreliant and able to rely on others, had had home
lives in which both parents were loving and emotionally generous and
the mother had given them a feeling of complete security.

The importance of being able to rely on someone — and of having

someone to rely on — is illustrated by the behavior of women who cope
successfully with the immense demands of pregnancy and early mother^
hood. One study found that the women who fared best were able to ask
for help from appropriate people and to do so directly, without hints or
manipulations. They had relationships with their husbands whose sup'
port they happily sought, and they themselves had the capacity to give
spontaneously to others, including their babies.
Those women whose pregnancies were marred by emotional difficuh
ties had a much harder time with dependency. They either did not ask
for support or did it in demanding and aggressive ways, suggestive,
Bowlby argued, of their lack of confidence that true support would ever
be forthcoming. Typically, they were dissatisfied with what they received
and, in the end, not adept at giving themselves.^
Although Bowlby rarely spoke of solitude, of the importance of being
able to be happily alone, of the creativity and self-knowledge that can
come in times of stillness, he did believe that, like self-reliance, the

capacity for healthy solitude in adult life arises from being secure in the

realm of attachment and having a secure base to return to."^

But implicit in all this is that the one builds on the other, that having
an internalized secure base, a strong sense of having been loved and of
having confidence in one’s essential ability to love and be loved, enables
one to both enjoy solitude and to confidently seek nourishment when
one needs it.
 Attachment in Adulthood 383

The ways in which an internalized secure base operates in the passage

to adulthood has been studied by Roger Kobak, who has used the Adult
Attachment Interview to assess attachment in teenagers. Kobak found
that secure teens —
those who were able to speak coherently and thought'
fully about their experiences with their parents were better able to—
handle conflicts with both mother and father. They were more assertive

and more capable of listening to their parents’ point of view. And they

showed less dysfunctional, critical anger. They also made an easier transi'

tion to college. “We found,” Kobak says, “that the secure kids were rated
by their friends, who had known them for the first seven months of the
college year, as much more ego resilient —
that is, much better able to

cope with stress, study when they needed to study, have a good time
when they needed to have a good time.” The dismissing freshmen who
had trouble remembering early experiences with their parents and played
down the importance of attachment issues in their interviews
— “were
seen by their peers,” Kobak says, “as more hostile, more condescending,
more distant.” The preoccupied students —— embroiled, angry, and inco'

herent when speaking about their parents “were seen by their peers as

more anxious, introspective, ruminative.”^

This study suggests that young adults, like the astronauts and the preg'
nant women, continue to be aided by the secure base they have had
at

home. It gives them the strength to do the adult equivalent of explo'

ration— take risks, face challenges, be open to the new. In all likelihood,
it also puts them in a better position
to find a new attachment figure

and thus a new secure base — and to serve that role themselves.

The lack of a secure base would seem to leave one struggling with a
ambivalent
profound and painful loneliness. The person with a largely
if on fire and convinced
style knows it’s there and is driven nuts by it, as
disavows it.
he can never put it out. The predominantly avoidant person
form of sepa'
But both, I think, are haunted by a fear of loneliness, some
ration anxiety, occasioned by panic attacks or depressions, and a hungry
would speculate that for that
search for a sense of internal goodness. I

types of addic'
reason the two attachment styles seem prone to certain
tion, the ambivalent becoming addicted to people,
the avoidant to work,
Ultimately, the
power, acquisition, achievement, or obsessive rituals.

them shows up in surprising ways, such

control their loneliness has over
as in news stories about famous people, like the ruined New York judge,

who seem to have it all and then bring disaster upon themselves through
their behavior in passionate attachments.
 ^

384 The Legacy of Attachment in Adult Life

* * *

To Bowlby, mourning in adulthood remained as important as mourning

in infancy, for even in adulthood the loss of an attachment figure, such
as a parent or a spouse, is a powerful blow that rips away our secure base
and, in a sense, undoes our world. His work with Colin Parkes, a psychic
atrist known for his work on bereavement who joined Bowlby’s staff at

the Tavistock in 1962, further extended attachment research and think'

ing on the subject and eventually influenced Elisabeth Kiibler'Ross, who
incorporated attachment concepts into her famous 1970 book On Death
and Dying. Parkes has since used attachment principles in both bereave'
ment counseling and in developing programs to help people through the
process of dying.^ Both Parkes and Robert Weiss, a Boston sociologist
who studied loss due to divorce, found that protest, despair, and detach'
ment, which James Robertson saw in hospitalized children, are the stan'
dard responses to loss of an attachment figure at every age. Healthy
reactions to loss in adulthood entail a gradual emotional reorganization
and a refocusing of one’s attachment feelings to new figures.^
When mourning fails, and such a reorganization does not take place,
it may well be related to early anxious attachment. If someone grew up in
a home where attachment needs were played down and considered baby'
ish, feelings of grief are likely to be stifled and not worked through. The
approach to loss, whether it’s of a pet, a best friend, or parent, is an
immediate and premature ‘We can’t dwell on these things; let’s move
on”; and so the loss is never truly experienced. A man like this stifles all
his pain when his wife dies because the only means he has of coping with
emotional pain is avoidance. His loss does not seem to make him suffer;
he just becomes a more distant, compulsive, and strangely flattened pet'
sonality. If the avoidant personality says “Let’s move on” prematurely,
someone who grew up in an enmeshed home may never feel ready to
say it. Instead, he is more likely to suffer from chronic, unresolvable
grief. Tears and laments spill out on every occasion, hut they never get
resolved so that life can begin anew. The ambivalent clutching that
characterized the relationship with the living person now characterizes
the relationship with the dead. Years later, tears of despair come just as
readily as before. Nothing is changed.
If a spouse is away for a period of time, it is natural to miss him. If a
move is made to a new place, it is natural to feel a loss over friends and
family who have heen left behind and to work assiduously to create new
ties to replace the old. But with separations, too, anxious attachment can
 Attachment in Adulthood 385

deform the process. Clinical work suggests that people with what appears
to be an avoidant or dismissing psychology often fail to recognize that
separations have an emotional impact on them. We see this when a
ther-

apist goes on vacation and the patient becomes depressed without hav'

ing any idea that this may be related to the therapist’s absence. Instead,

he feels a worsening of his emotional condition and an inexplicable

indifference to or anger with the therapist upon his return. When a

spouse is away, a person with this psychology may become obsessively

focused on work, may even celebrate the separation as an opportunity to

getmore work done, but then be strangely, perhaps even cruelly, distant
from the spouse when he or she returns. Young people like this go to
col-

lege farfrom home, emigrate to a new country, readily leave friends

behind to take a job in a new place, never anticipating that they will suf-
fer any feelings of And, because they don t take attach-
loss as a result.

ment needs seriously, they don’t take care of them. They fail to make the
concerted needed to maintain old connections or build new ones.
effort

In the end they may be brought low by their unmet

attachment needs
as a sense of de-
and by an attack of separation anxiety, experienced
desperate, and too
pression and inner collapse. Feeling too weak, too
ashamed to approach anyone for love, they may allow this condition
to

worsen and persist.

Others, who display what appears to be an ambivalent or preoccupied

attachment style, are hypervigilant about separations, likely to become

anxious or even panicky when left, and to become overwhelmed by feel-

and impotent rage. They do not readily venture forth

ings of clinginess
do not believe their attachment needs will ever
or take chances, for they
often using guilt and
be met. They cling tenaciously to what they have,
blame to keep their attachment figures on a short leash.

Cindy Hazan and Phillip Shaver of

On July 26, 1985, social psychologists
in the Rocky Mountain
the University of Denver published a “love quiz’’
attachment quality
News, a local paper. Designed to test the impact of
themselves with one
on romantic love, the quiz asked readers to identify

of three statements that Hazan and Shaver believed would represent the
of secure, avoidant, and ambivalent
attachment in
emotional styles

adulthood:

• 1 find it relatively easy to get close to others and am com-

depending on them and having them depend on me.
1
fortable
 P 1

386 The Legacy of Attachment in Adult Life

don’t often worry about being abandoned or about someone

getting too close to me. [Secure.]
• I am somewhat uncomfortable being close to others; 1 find
it difficult to trust them completely, difficult to allow myself
to depend on them. 1 am nervous when anyone gets too close,
and often, love partners want me to be more intimate than 1

feel comfortable being. [Avoidant.

• 1 find that others are reluctant to get as close as 1 would
like. 1 often worry that my partner doesn’t really love me or
won’t want to stay with me. 1 want to merge completely with
another person, and this desire sometimes scares people away.
[Ambivalent.

Respondents were then asked about their childhood relationships, the

emotional styles of their parents, and fifty^six questions regarding their
most important romantic relationship. Hazan and Shaver found that the
subjects whose responses suggested a secure attachment style were hap-
pier in their love relationship, saw their partner as a good and trusted
friend, were able to accept him despite his flaws. Their relationships
lasted longer and they had far fewer divorces. Subjects whose responses
suggested an avoidant style revealed a fear of intimacy. They did not
believe in head^over^heels romantic love, were often doubtful of whether
romantic love existed at all, and found their relationships to be full of
emotional highs and lows and jealousy. Ambivalent'looking subjects
reported more intense roller-coaster romances, tended to fall in love eas-
ily, and became obsessively involved with their lover, so that it became
difficult to take their minds off him.
Unfortunately, there was a tautological quality to these findings — if

people think of themselves as secure or anxious, they are going to say a

lot of things about themselves that are consistent with that view. The

results were nevertheless intriguing, partly because when the three cate-
gories were studied in detail, patterns emerged that fit well with one’s
expectations and that seemed to add a fuller picture of attachment quality
in the adult years. If the avoidant style seemed very much like the distant,

self-sufficient man “who refuses to commit,” the ambivalent style seemed

to embody such popular phenomena as “women who love too much.”
The Hazan-Shaver report caused a huge stir in social psychology that
continues to this day. Numerous studies have been conducted using their
self-report measure or modifications of it. In one study, reminiscent of
 Attachment in Adulthood 387

into the lab. The

the Strange Situation, dating couples were brought
woman was separated and told she was about to go through an anxiety'
waiting room to be
provoking procedure. She was then sent back to the
the selTdescribed
with her partner. Sharp differences emerged between
the secure group more
secures and the avoidants, with the women in
likely to seek emotional support and physical
contact and the secure men

much more likely to give it.^°

Hazan and Shaver, meanwhile, wanted to see if they could find a link

between security and exploration in adults, as Ainsworth had found in

infants. To that end they launched their “love and work” study. Once
while the
again, those in the secure group seemed to be doing fairly well,
ambivalent group, for
anxious subjects had problems. Those in the
difficulty concentrating, and
instance, tended to procrastinate, they had
were most distracted by interpersonal concerns.
They also had the lowest
Situation,
average income. “It’s very parallel to infants in the Strange
in ex'
Hazan “where the ambivalent kids are not able to engage
says,

ploratory behavior because they’re so

preoccupied with where their

mother and what she’s doing.”

is
workaholics and
The avoidants, meanwhile, were most likely to be
social life. Some said
most inclined to allow work to interfere with their
others that they pre'
they worked too hard to have time for socializing,
ferred to Not surprisingly, their incomes were as high as the
work alone.
was as low as the ambivalents. Again,
secures, but their satisfaction

Hazan says, “it’s like the avoidant infants — they’re not exploring hap'

pily, but they’re putting all their energy into it.

in the field about

There has naturally been a great deal of skepticism
assessed by the
whether one’s attachment pattern could he meaningfully
simple self' report procedure that Hazan and
Shaver used. Nevertheless,
found
it came something of a shock when researchers subsequently
as
the Hazan'Shaver methodology
that when adults are assessed using both
there was virtually no
and Mary Main’s Adult Attachment Interview,
correlation between the results. No one has as yet been able to figure
problem, of course, with the selhreport,
which
out exactly why. One
is that people are
deten'
both Hazan and Shaver readily acknowledge,
things about themselves, but unless
they
sive, they say all sorts of positive
to remain hidden. This seems to
be
are probed in depth the truth is likely

a major difficulty with avoidant

adults, many of whom rate themselves as

to be dismissing on the AAl.

secure on the questionnaire but turn out
When this trend is accounted for. Shaver believes, the two assessments
388 The Legacy of Attachment in Adult Life

show much more congruence. Meanwhile, studies by social psycholo'

gists all over the world have found consistent results with the Hazan^
Shaver measure, even, surprisingly, in the avoidant category: If someone
calls himself avoidant in the selhreport, he tends to feel and behave in
ways one would expect — to be low in selhdisclosure, low in relationship
enjoyment, more likely to get divorced, and more likely to be promiscu-
ous (and thereby evade intimate attachments) than people with either of
the other two relational styles. So, there is a real muddle here.
Kobak believes that what Hazan and Shaver tapped in to for many
subjects was the specific attachment feelings aroused by their current
relationship. The idea is that when we’re with a noncommittal person
who’s always threatening to leave, we are prone to feel like a clingy,
ambivalent child — regardless of our previous attachment experience.
When we’re with an insecure, punishing nag, we’re prone to become
somewhat distant and avoidant. Indeed, Kobak has found that in mar-
riages people don’t always act the way their Berkeley adult attachment
category would predict. As a result he believes that “adult romantic style

may be more the product of adult romantic experience than of childhood

relationships.”^"^

To the extent that this is true, it also suggests a possible limitation in

the classification system devised by Main. For unlike the childhood
attachment categories Ainsworth developed, which label not the indi-
vidual but the relationship (secure relationship with mother, avoidant
relationship with father). Main’s system identifies each adult with a sin-
gle attachment style. This may he legitimate as far as it goes. It could, as
Klaus Grossmann suggests, represent an integration of past experience
with all attachment figures, as well as the tendency to revert to the style
learned with the primary caregiver when under stress. But it neglects
the possibility that we can draw on different models when relating to dif-
ferent people, and it may not, therefore, be enough to capture the vari-
eties of the adult experience.

It thus seems reasonable to assume that there is more to adult attach-

ment status than the AAl tells us. A predominantly dismissing adult, for
instance, may have hidden realms of security or pre-occupation that the
interview, at least in its current form, is not yet able detect. Also, in most
research like this, people do not fall neatly into the available categories;
there is usually some squeezing involved on the part of the scorer. The
assumption is that individual psychology exists on a continuum, that var-
ious people might fall somewhere between dismissing and secure, or
 Attachment in Adulthood 389

between secure and pre^occupied, and that lines must be drawn in order
place them into some category. But maybe it is not just a
continuum.
pick up on
Different selhstates may be involved. The AAI may even
these different selTstates, but it does not yet know what to do with

such data.
An example here may help. I have a patient who much of the time
seems remote and unattainable. He does not make connections
easily,

and most of his friends do not know him deeply. He does not trust
that

lovers he demands
people’s interest in him is sincere, and with new
extraordinary displays of devotion before he will consider making
a felt

commitment. It is not as if he does not care for these people. He does.

But, as a mode of selhprotection, he does not allow himself to experience

how connected he is with those he cares about, including me. Thus,

the
throughout most of his relationships, he superficially displays
avoidantHismissive style, and I think an AAI scorer might be
tempted

to categorizehim as such if he was interviewed at the right time. But

to himself that someone really matters to him,
once he acknowledges
especially a woman, it is almost as if another personality or selhstate
separation or
emerges: he becomes desperately dependent and fearful of
loss. In our work together, it has become
clear that he is fundamentally

puppy who wants a lot of attention all the time, but in most
rela-
like a
actively, preferring to starve himself than
tionships he does not seek it

child.
risk the sorts of wounds and humiliations he suffered as a
to the current system of classifying adults,”
Arietta Slade
“According
says, “you’re eitheran apple, an orange, or a banana, and there s no such
clinical experience. It
thing as mixed fruit. But that doesn’t jibe with
in one relation^
doesn’t jibe with the way people are. You see somebody

ship who’s been clinging and demanding and needy, and then you put
another relationship with a clingy, demanding, and needy
pet'
them in

son, and you watch them turn around and become unavailable and
closed off and rigid. There’s an interrelationship
between the styles that

is not accounted for in the

categorical system.

No one equates anxious attachment in infancy with emotional diS'

severe disruptions of early attachment prolonged

turbance. True,
death ot a primary
maternal deprivation, harsh separations, the untimely
lead to serious
attachment figure— can, under certain circumstances,
or depres-
disorders,including psychopathic personality, agoraphobia,
to establish in great detail.^”^ But in
childhood.
sion, as Bowlby attempted
 390 The Legacy of Attachment in Adult Life

anxious attachment alone does not predict behavior problems in school

unless it is also accompanied by high stress, such as experienced by many
of Sroufe’s poverty childrend® And even where behavior problems do
arise, the child will still have many new influences and many opportuni-
ties to take a different path.
How then are we to understand anxious attachment’s relationship to
emotional disturbance? To begin with, anxious attachment is connected
to those issues which psychoanalysts have defined as “pre-oedipal.” That
is, they arise in infancy and toddlerhood, before the pre-school eia when
triangular problems with mom and dad gain in importance and when a
whole new array of potential psychological disturbances come to the
fore. Indeed, at some point when psychoanalysis and attachment are
rewoven together, pre-oedipal and attachment disturbances may come to
be synonymous. Either way, we are dealing with a vast realm, from peo-
ple who are essentially whole but have certain types of struggles in the
realm of intimacy to those who are so badly damaged by crazy, violent,
depressed, rejecting, or absent parents that they were never able to inter-
nalize enough sense of goodness about themselves and others to keep
rage, paranoia, and panic from spoiling their lives.
When the anxious quality is severe enough to warrant thinking in
terms of a pathology, it may still be impossible to draw meaningful corre-
lations between attachment category and diagnosis. Nevertheless, the
anxious categories are suggestive of the personality distortions to which
adults are prone, and it is interesting to see how well the attachment idea
overlaps with our understandings of character disorder. Bowlby, for in-
stance, believed that an entrenched avoidant attachment lay at the core
of the narcissistic personality disorder, one of the primary diagnoses of
our era. And, indeed, the overlap is evident in the way clinicians think
about the origins of this condition. Like Winnicott, Heinz Kohut, one of
the foremost explicators of narcissistic disturbances, believed that infants
need deep and consistent appreciation. Babies crave having their per-
formance validated, they need to be seen and loved for who they truly
are, and they need to be given an ongoing sense of belonging, of being a
valued fellow being in the family.*^ If a mother fails consistently to
attune to her bahy in this way and to respond to his complex emotional
needs, the young child, feeling unknown and unappreciated, is unahle to
know or appreciate himself. He shrinks back into a sense of helplessness,
smallness, defectiveness, and shame, which he may then defend against
by clinging to his infantile grandiosity, a grandiosity one or both of his
 Attachment in Adulthood 391

parents may promote.^*^ Without corrective experience, he grows into an

adult who constantly seeks the adulation and perfect union he never had
as an Outwardly selfdmportant, prone to pomposity, selhadora-
infant.

tion, and an annoying attitude of entitlement, he is haunted by

a fragile

selhesteem. His friends complain that he’s only interested in talking

about himself, his boss that he takes frustrations too personally, his
neighbors that he’s pushy and conceited. None of this is inconsistent

with Sroufe’s descriptions of obnoxious, entitled, and wounded pre^

schoolers with avoidant attachment histories.
Avoidant attachment would also seem to be a component of compuh
At extreme, the compulsive personality is the
sive personality traits. its

nightmare version of the uptight, authoritarian father who is determined

to banish all emotions. He lives in a constricted world, his attentions

narrowed to schedules, rules, and tidinessi and he is obsessed with trivia.

He continually peeved by what he considers to be the unethical or

is

substandard behavior of others, and he gets more pleasure out of

making

lists, cataloguing his belongings, or going to the toilet than in comrade^

puts his
ship or play. If he lives in an apartment building, he always
with
garbage out in the proper bags at the correct time and remembers
pinched disapproval the garbage lapses of each of his neighbors. We
can
child s narrow^
see a kernel of the compulsive character in the avoidant
to escape his
ing of his focus to a toy or some other object in order
painful attachment conflicts.*
The isolated preschooler Sroufe describes, who also has an avoidant
things
attachment history and perhaps certain genetic leanings, may, if
continue to go poorly, develop into a schizoid personality. A
loner who

has no friends, the schizoid adult prays there will be no

one on the eleva^

tor when it lands at his floor. Annoyed by casual social encounters and

the demands of small he cannot understand what pleasure people

talk,
entanglements
can possibly extract from their tedious and demanding
with one another. For him, avoidance seems to have
become a religion
and his unconscious wish for connection a disturbing and tightly
hemmed'in desire.
perhaps to the
At the other end of the emotional spectrum, married
maker who balances her account and monitors her behav
-
compulsive list

obsessive or compulsive personality should not be

confused with someone who suffers from
*The
severe condition usually involving
obsessive-compulsive disorder, or OCD. The latter is a more
is often treated with medication,
and is no
phobias (e.g., of germs), rituals (e.g., hand washing),
personality issue. Similarly, the hysterical character
should not be confused with hyster-
considered a
were prominent in breuds day.
ical paralysis and other such anxiety-induced physical symptoms that
 392 The Legacy of Attachment in Adult Life

ior, is the live wire. Seductive and engaging, she can often make people
teel there’s no one on earth they’d rather be with. Often diagnosed as an
hysterical character, she is scattered, charmingly incompetent, and easily
thrown into a tizzy by schedules, details, and responsibilities. Her dra'
matic flairmakes her popular. Male neighbors delight in meeting her in
the hall, and they wonder: “Why is she married to But she flees
from intimacy, and, like the ambivalent child, she tends to be demanding
or clingy, immature, and easily overwhelmed by her own emotions.*
Perhaps the most extreme condition to which early ambivalent
attachment may be a contributing factor is the borderline personality.
Unlike the schizoid, had an intense relationship with almost every^
she’s

one in the building but they’ve ended badly and she now considers them
all shits. Impulsive, selPdestructive, and ragetul, patients with this
diagnosis are often dreaded by psychotherapists, as Peter Fonagy suggests:

It takes but one patient with severe borderline pathology to

shatter the equilibrium of the therapist’s life with unending
demands for “special” treatment, round-the-clock availability,
physical and sometimes sexual contact, perfect attunement,
heroic efforts to prevent self-injury or suicide — only to be
repaid by contempt, reproach, hostility and at times outright
physical attack.^^

They seem extreme edge of ambivalent attachment, where

to be at the

preoccupation with attachment signals Do you really care for me? I’m—
going to make you prove it! What’s the meaning of what you just did?
You’ve turned against me! I’m, utterly alone and unwanted! — readily
reaches the point of panic.
Few people, of course, perfectly fit these stereotypes (although some
come very close). Narcissistic problems wend their way through all the
personality disorders, and many people exhibit a mixture of compulsive
and hysterical traits. As we’ve seen, there are myriad contributors to per-
sonality,and much depends upon each person’s inborn temperamental
leanings, the unique circumstances in which he was raised, and what

*Just as the compulsive personality seems to he rtiore common among men, the hysterical seems to
he more common among women. This may he reflective of a biological gender difference or of the
fact that such traits are more acceptable in women and therefore get played out in
a less guarded
way. The same gender issues may hework in avoidant and ambivalent attachment patterns in
at
adulthood, with fretful ambivalence being more plainly apparent in women and stoic avoidance
more apparent in men.
 Attachment in Adulthood 393

transpired at each developmental stage from infancy through adoles'

cence and perhaps beyond.
The intensity of personality problems also varies greatly. There are, for

instance, welhfunctioning people who have some narcissistic traits and

those who and shame. Roger Kobak has

are crippled hy their grandiosity

proposed a similar continuum for quality of attachment, ranging from the

highly dismissive to the highly preoccupied on one scale and from the
very secure to the very insecure on another. This is a matter of great con^

troversy in attachment circles, but 1 find it appealing partly because it

suggests what many readers of this hook will have already concluded
the extent to which the problems of anxious attachment are in some way
relevant to all.
 26
Repetition and Change:
Working Through
Insecure Attachment

Early in Freud’s development of psychoanalytic technique, he discovered

that his patients relived aspects of their childhoods through their rela-
tionship with him, often responding to him as if he had been one of their
parents. It soon became a standard assumption of analytic treatment that
the patient frequently relates to the analyst the way he would to an
important early no matter how unlike that figure the analyst may
figure,

he in most respects. Working through this distorted perspective other- —

wise known as the transference, hut obviously related to what Bowlhy
would later call the internal model — is considered a key feature of ana-
lytic work.
How transference manifests itself is still a matter of debate, hut a num-
ber of features are widely recognized: Transference does not only apply to
the analytic relationship hut to virtually all relationships that are close or
involve a figure of authority; it can be triggered by certain types of inter-
actions that are especially charged for the individual; it can he positive
(irrational love or veneration) or negative (irrational hate or contempt);
and valence can change, accounting not only for the intensity of
its

falling or being in love hut also for the intensity of mistrust or loathing
 ^

Repetition and Change 395

that can be directed at the same person. Finally, it is not only our early

self that we replay in adult relationships, but the parental role as well, a

fact that is particularly evident when, inexplicably and seemingly against

our will, we act like one of our parents with our own children.
Transference feelings emerge in different ways with different people.
Harry Stack Sullivan, a founder of the interpersonal school of American
psychoanalysis, believed that, as children, we develop different senses of

self in each of our primary relationships and that in later life we relate to

others through these selves. (This corresponds nicely with subsequent

attachment findings that a child can have different qualities of attach'
ment with each of its attachment figures.) There may be a selTin'
relatiori'tO'mother pattern, a selt-in'relation'tO'father pattern, a selhin'
relation'tO'older'brother pattern, a self'in'relation'to-nanny pattern,
depending on who was important to the child when he was young.
A girl who is brought up by a cold, domineering mother and a kind
but ineffectual father will tend to see similar qualities in others, with
the

result that she behaves toward them at times as if they were

emotional

and behavioral copies of her parents. An act of coolness by a close female

friend causes her whole painful relationship with her mother to
be acti'

vated within her: Suddenly she is drenched in feelings of being

small,

oppressed, and helpless, while her friend appears alhpowerful and

coldly

dismissing. The friend may indeed have been cool, even somewhat
rejecting, but there is an irrational intensity to the womans reaction.

Meanwhile, she persistently sees the men in her life as weak despite their

evident strengths and accomplishments. And the power of her expecta-

tions is such that she actually brings out the weakness in many of the

men who are close to her, so that their weak side is all she ever knows or

relates to. At the same time, she may be unaware of the rage she teels

over dad’s ineffectualness and the subtle abandonment this

constitutes.

longings, we
Freud observed that, despite our conscious protests and
experience, no
often seem to prefer relationships that repeat our early
matter how unsatisfying. People can, of course, change, and their early

by the
models can be modified. But, as Mary Main’s work has shown,
time some people reach adulthood their models have
become relatively
closed and their capacity for selTreflection and
change compromised.

see Freud’s “repetition compulsion’’ in people who seek the same sort
We
of partners again and again, exhibiting behavior that remains dis-

turbingly consistent across relationships. B(.)wlby

explained this as a nat-

ural bias in favor of what we already know: No

matter how painful or
396 The Legacy of Attachment in Adult Life

unfulfilling it may be, it offers the second-rate security of being familiar.

Fairbairn explained it as an unconscious commitment to those we wor-
shipfully loved as children. No matter how much they hurt us, we don’t

want to give them up. Many, including Freud, have seen in repetition an
effort to master an early trauma. In the process of repetition, however,
the very meaning of love becomes subtly perverted. It may come to mean
pining after the unavailable; being seduced and rejected; being treated
with contempt; being engaged in sado-masochistic warfare; being wor-
shipped and adored as a paragon of strength even as one’s true self lies

hidden, glutted with ugly needs and emotions; and so on.

The fact that many people find romantic excitement in a lover who
displays the qualities of a rejecting parent, an excitement that they do
not find in others, suggests the degree to which they remain not just

committed to but enthralled by early attachment figures. They can’t let

go of the mother or father who didn’t love them the way they needed to

be loved. And they continue to be bewitched by the hurtfulness that

compromised their care. They are caught in the parental orbit, a hurt

child still leaning out for a love that can never be, and blinded to what
they are doing by the belief that they have no feelings toward their par-
ents at all or have washed their hands of them.
The issue of loss is important here. Bowlby talked a great deal about
the child’s need to be able to mourn a lost parent and how hard that can
be if conditions are not right (see Chapter 8.) But losses come in many
shapes: We once had an adoring, available mom, but when we became
two and over-taxed mom’s patience, the good mom was replaced by an
irritable, rejecting disciplinarian. She only became more so when the sec-
ond baby was bom, which increased the demands on her and made us

furiously jealous. That kind of loss also needs to be mourned, but the child
is rarely able to do so. Instead, a twisted attachment fomis, in which the
child may become magnetized by the rejecting quality, and to give that up
feels like giving up love itself. And so one seeks love in repetition.

But that is not all. An obvious corollary is that the prospect of being
given to in a truly loving way is undermined at every turn; indeed it feels

perversely unacceptable. There are many layers to this: We don’t deserve

it. We’ll be rejected. It’s not the way mom did it. And getting it now
would not only deprive one of one’s beloved bitterness but would also
activate the long dissociated pain of that early loss.

For many people, therefore, adult sexual attachments do not give birth
 Repetition and Change 397

to a renewal of trust; or, if they do, the trust recedes as it is overpowered

by earlier patterns.
What implications does this have for the a sense of having an inter^
nalized secure base in adulthood? That is, for having a feeling of okay'
ness, of healthy independence, of trust, self' reliance, and confidence in

one’s ability to love and be loved — all those qualities that are the halh

mark of secure attachment? A secure base is not a yeS'Ot'no proposition;

that is, for most people, the issue is not that you either have it or you

don’t. An exception might be psychopathic personalities, like orphans

raised in institutions, who never had a loving caregiver and therefore
never internalized a loving and self'loving aspect, or some of Bowlby s

juvenile thieves who, because of terrible losses or separations, are so

trapped in their rage that they cannot access part of themselves that feels
loved and capable of love. But for others who are not so fundamentally
disturbed, including those who would be characterized as avoidant'diS'
missing or amhivalent'preoccupied, the capacity to love and be loved in
a nourishing way is more like a channel that one has trouble tuning in

to. In intimate relationships, perhaps especially with the opposite sex,

one is fixated on another, more compelling, channel. Here, always play'

ing, is the drama of the rejecting parent and the longing child who is

some combination of angry, bad, inadequate, manipulative, and spurned.

Living for this drama, one moves into each relationship playing one or
both sides of the pattern. This obsession, this repetition compulsion,
inevitably intensifies one’s sense of being bereft of internal goodness, of
seeing such goodness as only existing in desperately sought after others,
whom one hates and envies as well as loves and desires. The result is that
the internalized secure base, which is indeed there in some form or

shape, does not have a chance to develop. It lies dormant or underuti'

lized and thus fails to become a foundation that can be built upon and
strengthened by adult experiences of secure relating. This pattern, 1
of
think, is what accounts for the sense of total collapse and emptiness,
depression and doom, that often occasions separation or loss — or the

fear of it — in adulthood. One feels as if there is no internalized security

or goodness at all.

Needless to say, shame also plays a part in all this. For someone whose
predominant attachment orientation is ambivalent, core feelings of

shame are probably closer to consciousness, and the desperateness he

398 The Legacy of Attachment in Adult Life

feels about relationships is probably heightened by the sense that he is

not worthy of love. The person whose dominant attachment mode is

avoidant has more likely cut himself off from such anxieties. But he is no
less motivated by them — because staying cut off from shame, which is

crucial to him, absorbs a lot of his interpersonal energy.

A number of years ago I wrote a book. Top Dog/Bottom Dog, about the
ways in which these processes of desperation and cutoff operate in daily
life, and, in particular, how people exploit each other’s feelings of inade^
quacy in order to alleviate their own selhdoubts. The book told the story
of several fictional characters, including a couple named Martin and
Georgette Grodin. Although I wrote their story before I became acquaint'
ed with attachment theory, it is plain to me now that they each represent,

to different degrees, a version of an anxious child grown up. Their ways

of being with each other illustrates, I think, some of the problems of anx'
ious attachment in adult relationships.
Georgette is a popular, attractive, thirtyTive^year^old working mother,
whom friends like to confide in, hut who doesn’t get the respect she
should, especially from her boss and husband and even on occasion from
her best friend, all of whom are very attached to her. Martin is a dy^

namic, successful man, supremely confident, addicted to work, intolerant

of his wife’s depressions, and intolerant of his own feelings and needs if

they disrupt his otherwise steady course.

It is the evening of Georgette’s thirty^fifth birthday. She has been feeh
ing depressed for several weeks. She and Martin have not been getting
along, distance has been growing between them, and, to make matters
worse, Martin comes home late from work. Hoping that sex might bring
them closer together. Georgette prods Martin into bed. They make love,
but it’s not satisfying because they remain unconnected; and afterward
there’s an uncomfortable silence.

Martin: “How was it?’’ 11 can’t bear the thought that I’m not

a good lover.]

Georgette: “Fine.’’ ll’m ashamed to admit I’m dissatisfied.]

Martin: “You don’t sound fine.” [You’re ruining it again.

Georgette.]
Georgette: “I’m sorry.” [1 hate myself.]^

Both people in this relationship suffer from selFdoubts — Martin, that

he’s not a good lover, that he’s a disloyal husband, that he’s selfish;
 Repetition and Change 399

Georgette that she’s a bringdown, that she’s never satisfied, that she’s

pathetically needy. But whereas Georgette ruminates and torments het'

self about her inadequacies, Martin thinks he’s superior to her and most
other people in innumerable ways. He is particularly proud of his inde^

pendence and the fact that he is not ridden with the pitiful insecurities

he sees in his wife, his child, and his subordinates at work. But to main^
tain his superior posture, he has had to perfect strategies to keep his inse^

curities and selhdoubts at bay, in this case by making Georgette the

problem. Such maneuvering around feelings of shame is, 1 think, a fre^

quent legacy of avoidant attachment.

Anxious children learn to manipulate to get their needs met, and
inevitably their manipulations get carried over into adulthood. The
ambivalent child may become seductive or cute, act fretful, or make oth'

ers feel guilty for not giving him the attention he wants, all depending
on what strategic styles are modeled or succeed in the family. Georgette

makes Martin feel guilty about working late and neglecting their daugh'

ter; she stops doing things that he likes and that had always been a part
of their life together, under the ruse that she no longer has the time; she

becomes depressed and overeats. Such passive aggressive behavior is

partly a retaliation for the way he treats her, partly the only way she
knows how to make demands of a withholding person.

An avoidantly attached boy, like the Minnesota child who hacked

gradually into a teacher’s lap when we needed attention, will probably
learn to disguise his care seeking. He may become adept at using various
forms of control to get another person to be there for him; he may seek
out people whose needs are more apparent and who give without having
to be asked. When Martin is feeling low and wants attention from his

wife, he cannot be direct about it. He expects her to magically anticipate

his and, because of her keen sensitivity, she otten does. It

needs,
Georgette is planning to be busy with a friend on a Saturday, Martin does
not say, “1 don’t want you to leave me today because I’m feeling dejected
and blue.” First, because he does not want to know that that is his condi'

tion and, second, because, if he did know, he would be too ashamed to

disclose it. So he cajoles and wheedles, expresses irritation over the way

she spends her time, maligns her friend, finds some pretext to get angry,
all with the idea of maneuvering her to stay with him.
Having grown up with parents who could not tolerate any signs of
weakness or dependency, Martin has trained himself to he invulnerable.
Somewhere within him may he the potential to feel and behave like an
400 The Legacy of Attachment in Adult Life

overwrought ambivalent child, but, like a child in the hospital who has
shifted from protest to detachment, Martin has steeled himself and
moved on. He learned through repeated rebuff as a young child that
there is nothing more shameful than being an anxious apron-strings
puller with needs hanging out all over the place. It may be okay for other

people, he may even be drawn to it in the people he chooses for sexual

partners, but, although he can be attracted to such people at times,

although hemay become attached to them and care for them, he has a
hard time respecting them —
because, unconsciously, he hates those
qualities in himself. Meanwhile, to maintain his rigid invulnerability, he
is willing to jettison almost anyone who threatens to disturb his equilib-
rium by challenging his domination or his view of himself. This need for

control is one aspect of his addictive life-style.

Georgette, whose father died when she was young and who had an
enmeshed and caretaking relationship with her alcoholic mother, is nev-
ertheless a warm and attractive woman with a spontaneous streak that
draws people to her. She has a greater capacity for self-reflection than
Martin, and she has managed not to repeat her mother’s mistakes with
her own daughter. But when it comes to her marriage, her insecurity and
befuddlement is pronounced. In her relationship with Martin she is pre-
occupied with attachment issues, easily convinced of her own shortcom-
ings, and quick to take the blame.
Georgette sees Martin’s hidden, hurt side; she sees the unrealized
potential, the creativity that’s been suppressed along with his cut-off self;
she believes she could heal him. But, while she wishes the best for
Martin and wants to see him develop into what she knows he can be, she
is confused about her motives. Does she want him to open up for her
benefit or for his? Martin is confused, too. He loves having someone
attuned to him,who knows him well, who is so appreciative of what
makes him happy and why. But when she pushes too far, starts messing
around with his supposed insecurities, he distrusts her, convinced that
her Delilah-like ministrations are designed to tame, declaw, and enmesh
him. And there’s something legitimate to his distrust, for there’s a com-
pulsive, self-serving quality to what Georgette does, a repetition in some
ways of what she did with her mother.
As a child. Georgette felt that if she could he her mother’s therapist,
she could tame the beast in her. She could induce her mother to he
steadier and more nurturing, cure her of her alcoholism, and put an end
to her unpredictable rages. Her efforts did gain her a measure of safety
 Repetition and Change 401

and control, but they became her addiction, and that has carried over to
her marriage, as she hopes to succeed with her husband where she failed
with her mother. But, of course, she never does quite succeed. Because
she fails to heal Martin and make him into a more caring, devoted part-

ner. Georgette believes she is an inadequate wife, just as she once

believed she was an inadequate daughter. Nevertheless, she drives on.

Georgette assumes that by struggling with Martin this way, she will

eventually work something out for herself; that if she can just convince
him to be sensitive, caring, and consistent she will have achieved her
mission and be at peace. But she has little sense of the magnetic allure

his inconsistency has for her, of how her reactions to him are amplified
and distorted by her unresolved feelings toward both her mother and the
loss of her father, or how her own maneuvers and retaliations keep their
pattern alive. Meanwhile, the power struggles that she and Martin
engage in re-create the flavor of their childhood homes, Martin taking
the role of his disapproving mother. Georgette that of her own childhood
self. Their failure to achieve a satisfying intimacy with each other ap'
pears to be typical of many couples with anxious attachment histories.^

If we look again at the findings of Main and others who have assessed

adult attachment, it is important to remember that the key quality of the

secure-autonomous adults is not that they had secure attachments with

their parents. Rather they were all distinguished by an open and coher-

ent way of reflecting on their attachments. To the extent that they felt
wounded by a parent, they had managed to work it through, so that they
were no longer either rigidly cut off from their true feelings about that
relationship or still embroiled with hurt, rage, and blame. Somehow they
had arrived at the point where they could let the past rest and move on

with their lives.

Some of them appear to have been able to go beyond what they expe-
rienced with their parents because they had other relationships in child-
hood that kept an alternate perspective alive in them. According to
Main, some of these adults spoke of

the kindness of relatives who they saw only occasionally, the

parents of friends who provided alternative models of par-
enting when they were very young, or teachers who took a

personal interest. One father cited a cold night in early child-

hood when his visiting grandfather had gotten up to bring

402 The Legacy of Attachment in Adult Life

him an extra blanket: ''Nobody had ever done anything like

that for me before, and, at that moment, I realized that not

everybody was alike.

But for many children, such alternative models do not exist, which
makes their task much more difficult. It is, therefore, not surprising that
some of Main’s subjects attributed their change to psychotherapy.
Therapy can do many things. It can provide a new model of what a
close relationship can be; it can teach one to reflect on feelings, events,

and the patterns of one’s own behavior in a way that one was unable to
do before; it can compensate to some degree for nurturing experiences

one never had as a child; it can provide the guidance, persuasion, and
pressure one needs to break an addictive pattern and attempt something
new (such as recognizing, tuning in to, and beginning to use the strong,
nourishing parts of oneself that have been disavowed and seen as existing
only in yearnedTor others); it can be an opportunity to face some un^
pleasant facts about how one really operates in relationships; it can pro-
vide a context where that portion of the self that has always been ready
to relate in a new, more trusting, more direct and healthy way can
emerge and take what may be its first tentative steps; and it can offer a
safe haven where feelings of shame no longer present such a terrible bar-
rier to self-exploration. It is the last of these functions that I would like
to comment on here, because I think shame is a critical barrier to the
entire working-through process.
I had a patient in treatment who was prone to rageful fits with her
husband and children. She wanted to know who “the real me” was but
was afraid to find out. She didn’t think she’d like whatever true self she’d
find. And yet the experience of therapy suggests that being able to come
out of hiding and speak of hated aspects of oneself can he a healing
process. A patient who is able to face the shameful fact that she is a
shrew to her husband and children is free to stop the endless litany of
blame and to feel again. The constant complaints, the incessant de-
mands that the therapist see things from her point of view, the guilt-
inducing accusations that he always takes “their side” — such defenses
are part of her flight from feeling, fueled by the desperate fear that she

will he found in the wrong. An early goal of treatment will thus be to get
her to relinquish these defenses and to know what she feels, including
her self-hatred.
 Repetition and Change 403

To stop running and experience the shame is painful, but it is also an

opportunity. It gives her the chance to recognize that being in the wrong
for acting like a shrew does not mean that her husband or daughter aren’t
also wrong in their way, nor does it make her into a poisonously de^

formed and unlovable thing. That’s a legacy of how she experienced

being wrong when she was a child. But it’s a legacy she cannot overcome
so long as the shame remains unconscious and unspoken. Once she
speaks about it, to someone who is able to listen and absorb without
becoming anxious, judgmental, or compulsively helpful, something
changes. She is able to view herself from a freer, less tyrannical perspec-

tive, able, perhaps for the first time, to feel some sympathy for herself and
her predicament. She is able to see that her cruel lack of sympathy for

herself is in part what fuels her rages and her desperate need to blame.
Gradually she may find that she is able to look at deeper issues of shame,

closer to her core — of feeling, as a little girl, unwanted, a piece of excess

baggage who constantly had to prove her worth. For the first time, she is

able to see herself as she was as a child and to grasp the importance of her
attachment needs, which somehow got thwarted at that time. The clari-
fying, sympathetic, clean relationship with the therapist helps guide and

contain this process, and may help liberate her self-love. If it goes well,

her therapy will gradually enable her to be self-reflective in a way that

was impossible before and to be more appreciative of the mental states of
others — both of which have been found to be key ingredients of the

secure adult.
How might such a therapeutic experience affect Martin or Georgette?
In Martin’s case, it would encourage him to relinquish his addictions and
power games and risk standing still with his unfelt pain, first in the ther-

apy office, gradually on the would mean recognizing how

outside. It

wrapped up he still is in his mother’s orbit, despite the fact that he rarely
thinks about her and doesn’t even like her very much. It would mean
embracing the rejected child he once was, the insecure little boy who
couldn’t take his needs to his parents. It would mean giving up his pose
of unruffled poise and revealing some of his self-doubts and anxieties
about shame to his wife — a process that would not only help balance the
power between them but would also enable her to serve as a much better

attachment figure for him.

For Georgette such a process would mean learning to stand next to

her fear of abandonment without becoming panicked. It would mean

 —
404 The Legacy of Attachment in Adult Life

reexamining her hated inadequacies — weakness, neediness, perpetual

dissatisfaction — from an adult perspective. It would mean recognizing

that not only is she not hopelessly deformed but that she’s really quite
normal in these respects and that her relentless selhhatred is an act of
unremitting cruelty against herself. It would mean seeing that her con'
stant fear of abandonment — and the appeasements that grow out of it

allows people to take her for granted, to act disrespectful, and to con'
stantly reinflame her self'hatred and shame. It would mean trading in

her obsession with Martin’s acceptance and approval for a firmer, clean'
er, less manipulative, less caretaking, less clutching stance with him, by
which she gains the best chance of winning the respect and steady caring
she wants and deserves.
Becoming conscious of shame and other early unresolved anxieties in
this way, allowing oneself access to needs and parts of the self that were

put off limits ages ago, can also go on outside the therapy room. Close
friends and marriage partners can sometimes serve this function, when
the trust is such that they can disclose things to one another and explore
them in ways they haven’t before. This is perhaps taking the secure base
function to the highest possible level in adulthood.
In Main’s study, some of the secure adults who believed they had been
anxious children attributed the change to their relationship with their
spouse. Something new and special happened there that overcame pulls

they may have experienced toward transference and repetition. As

Alicia Lieberman puts it, “With a love partner one can go over and over
the kinds of things that one’s mother and father didn’t respond to well,
and if the partner responds well, it’s like a release, and one can finally lay

old conflicts to rest.’’^ The preoccupied wife who had ambivalent attach'
ments to her parents cannot believe her husband when he says, despite
their fights and mutual dissatisfactions, that he genuinely loves her and
wants to stay with her. She cannot assimilate it to her worldview, her

internal model. She is sure that he will abandon her, either because he
already wants to or because her impossible and anxious neediness will
eventually drive him out. But his steadfastness over the years builds her
trust. It causes her to remember her relationship with a great uncle,
whose love was precious and unwavering, and to think more and more
about him and how good she felt about herself around him. Gradually,
she assimilates her marriage to this model, and it becomes more central.

Feeling more secure, she finds herself freer to reflect on the past — about
how her parents made her feel, how it affected her behavior with others.
 —
Repetition and Change 405

She is able to consider what she’s feeling when her husband criticizes her,

because she is not unconsciously convinced that she is worthless and he

wants to get rid of her. And she is able to use her husband as a secure

base for this process.

A growing body of evidence indicates that these three variables
having had a loving, supportive figure available in early childhood, hav'
ing undergone in-depth psychotherapy, and/or being in a stable relation^
ship with a supportive spouse — are perhaps the most important elements
in breaking the intergenerational cycle of emotional damage. Several
studies have shown this to be the case with women who were abused as

children but have not become abusing mothers.^

Selma Fraiberg believed that having a baby is a key developmental

stage for adults and that it too represents an opportunity for change.
Having a baby can offer a new perspective on one’s own childhood
what one felt, what one’s parents must have felt, how psychologically
delicate the process is, how inevitably imperfect. Finding oneself dealing

well with the emotional challenges of parenthood can also be a tranS'

forming experience. “The wounds can really be healed in the context of
raising a baby,’’ Lieberman says. She believes that one’s sense of self can

be lifted by being with a baby who seems to be saying, “You’re doing

okay. Mom, you’re really doing okay.’’^

Given the considerable struggle that seems to be involved in resolving

the past and becoming our own person, where do we, as adults, stand in
relation to our parents, especially if our childhood attachments were not
secure?
That our parents have such immense impact on our lives, that they
affect the types of people we become and the way we behave with our

mates, that their anxieties often live within us in some fashion and
emerge in the way we experience, or fail to experience, our children

all this is naturally the cause for much emotion. In our society impas'

sioned attitudes toward parental flaws tend to flow in one of two direc'
tions, which, not surprisingly, echo the two major styles
of anxious

attachment. Some people hold that we should never speak ill of our par-
ents. They may not be perfect, but they did give us life, they did struggle
to feed us and take care of us, and to carp or complain reflects ingratitude

and betrayal. In my experience, people who hold this view rarely have a

warm and open relationship with their parents. The form may be there,

the dutiful visits, the gifts, the cards, the helping hand, hut the feeling is
406 The Legacy of Attachment in Adult Life

not. They seem to be defending so rigidly against feeling any anger that
they often have difficulty experiencing other feelings as well.
second response to parental flaws is unrelenting anger and blame.
A
In recent years Alice Miller, whose excellent hook, The Dvcnna of the
Gifted Child, introduced many people to the ways in which selTabsorbed
parents wound their sensitive children, has become a proponent of this
response. In a series of works, culminating a few years ago in Banished
Knowledge, Miller has become committed to exposing parental abuse and
denouncing mental health workers who whitewash or make excuses for
it. Her message is that parents do a great deal of harm and that we, the

children, must be willing to remember and embrace our rage if we are

ever to be whole.
A friend, a woman in her forties, attended a workshop in which she
had an important insight on this subject. She had been speaking bitterly
of the ways in which her mother had disappointed her since she was a
child, when another woman, in her seventies, asked her, “What would

you think of me if I attacked my mother at such length?” My friend said,

“I guess I would hope that by your age you would have come to terms
with all that.” The older woman said, “That’s exactly what the women

here in their twenties are thinking about you.”

I relate this incident because nowhere in Miller’s recent work is there

the suggestion that persistent anger at parents can work against us, even
when its roots are valid. In The Drama of the Gifted Child, Miller equated

a person’s ability to mourn with the ability “to give up the illusion of his

‘happy’ childhood.” This makes good sense; it is a useful prescription for

the avoidant adult. But holding on to anger and blame is another imped'
iment to mourning. It prevents the grown-up child from feeling the loss
of what he never had. His anger, in effect, shields him from his sadness,

and it also, paradoxically, keeps him enthralled with the very parent he
blames and hates and denounces. This is the ambivalent, enmeshed posi-

tion, and, in adopting it. Miller, who in her old age is seething over what

her mother did to her, has fallen into a writing style typical ot the inco-
herent, muddled interview transcripts of preoccupied adults.

Miller is right, I think, that early wounds must be fully experienced at

some point in one’s life, but carrying them like a banner is not a symbol
of health or maturity. Eventually, one must separate, in the positive sense
of becoming one’s own person, which means not just letting go of the
unconscious, neurotic tie to the parent, hut letting go of the wound that
 Repetition and Change 407

perversely sanctifies that tie and letting go of the ways in which one’s

own behavior with others (including with one’s children) replicates it.

The cards that people get dealt — in terms of the blessings and curses
of heredity, the supportiveness of their families, the quality of their early
social encounters — are terribly unequal. And yet no one comes away
with the perfect childhood. Even those that attachment theory labels
secure are not without troubles, anxieties, annoying habits, selhdoubts,
phobias, depressions, or obsessions. Securely attached children can have
an overly sexualized relationship with a parent, be too encouraged in
their aggression, fail to experience adequate limit setting, have their
intellects overemphasized or dismissed, feel tormented by parental con^
diet, and by the inner conflicts that result from allegiances, betrayals,
and identifications with a damaged or hurtful parent. They may experi'
ence unfortunate sibling rivalries, suffer inadequately mourned losses, be

troubled by conflicts outside the home, and so on. Various estimates put
the securely attached as high as two thirds of the middle^class popular
tion. As Arietta Slade and Larry Aber have written:

Given our knowledge of human nature and of our fellow men

and women, it would be foolish to assume that such a large

proportion of individuals would be free of neuroticism or char^

acter pathology. In fact, in reading the adult attachment tran^
scripts of secure individuals, neurotic conflicts, problematic

identifications with parents, evidence of character problems,

etc., are quite evident. And, yet, at a fundamental level these

do not interfere with the capacity to recognize and integrate
the importance of early relational experience, to remember
the effects of that experience, even if it was negative, and to
respond to their children in a way that is relatively free of the

burdens of the past.^

The willingness to remember and reflect is at least as important in

keeping the past from weighing too heavily on our relationships with our
children as it is with our mates. This was brought home to me recently
in

my own experience as a new father. In the early months of my son s life 1

found myself haunted by the fear of losing him. It often came upon me

just at those moments when 1 felt most in love, when

he was most

unbearably cute. 1 would be hit by a fantasy that something terrible

408 The Legacy of Attachment in Adult Life

would happen to one of us. Initially, I tried to shut this fear out and con-
vince myself it wasn’t important, just a “rogue” thought, as a friend’s psy-
chologist once put it. When I finally decided to face it, I remembered
what actually happened to me and my father. We had a warm connec-
tion until I turned nine or ten. Then we became antagonistic, and for
several years our relationship was rent by battles.
Recently, my father commented that he liked little children mainly
until they reached what he called the end of the age of innocence. I
thought, yes, after that their independence can become trying and
thence begins the struggle for control. For the first time I recognized that
I had already begun to fear repeating this process with my son, who is

willful, assertive, and full of beans. Unconsciously, I expected that as-

pects of his personality would represent a poor fit with mine (which is

quite similar) and, rationally or not, I was anticipating struggles that

would put an angry wedge between us. As this issue became more con-
scious for me, the concern did not go away. I still recognize it as some-
thing I may have to deal with in one form or another. But being able to

reflect on it and talk about it has helped dissipate some of the haunting
anxiety about permanent loss, and it gives me more confidence that
when we do come into conflict my reactions are less likely to emanate
from old destructive patterns.
An important lesson emerging from adult attachment research, then,
would seem to be this: We cannot change our childhood. But we can let

go of the defensive and obsessive postures formed at that time. We can

make sense of what has been repressed and forgotten. We can reexperi-
ence dissociated feelings with a new appreciation for ourselves as we
were as children, for the situation that existed at the time, for the par-

ents who may have caused us to suffer. And we can successfully mourn
our losses. If we’ve managed to hold on to an alternative model, and if

we are wise or lucky in love, we may be able to work through our child-
hood experience in the context of a marriage or something like it. If the
grip of the past is too strong, we may work it through in therapy. In

either case, if we remain conscious of ourselves and of the pull of early

models, even if hang-ups of various kinds remain, as inevitably they
must, we have a better chance of creating satisfying relationships with
our mates and secure relationships with our children. To that extent it

seems that in emotional life, much as in history, we are only doomed to

repeat what has not been remembered, reflected upon, and worked
through.
 PART VI

The Odyssey of an Idea

!

I
 27
Avoidant Society:
Cultural Roots of
Anxious Attachment

Before the modern era, most life tended to be family life. In preindustrial

Europe, production was carried on in family farms or small shops. Hired

laborers lived with the farmer, the baker, or the cobbler and his next
of

kin and were considered family members themselves. Servants fit into
the family affectional system. Their relationships with their master’s
chib

dren were often like that of an uncle or aunt, and their own children like

that of siblings or cousins. Apprentices were treated like sons,

obliged to

obey the master, and were completely dependent on him until they
turned twenty^one. Few people went out to work. The old did not be-
come segregated in retirement villages or institutions, hut lived out their
days within the family they’d been a part of all their lives. The setting

may have been claustrophobic, with the intense hatred, resentment, and

murderous jealousies that inevitably arise under such conditiems; but

there was also love, familiarity, and unquestioned belonging.

Time was [Peter Laslett has written], and it was all time up to

200 years ago, when the whole of life moved forward in the
family, in a circle of loved, familiar faces, known and fondled
412 The Odyssey of an Idea

objects, all to human size. That time has gone forever. It makes
us very different from our ancestors.^

Preindustrial society was hierarchical, authoritarian, and unjust and

characterized by a degree of bullying and submission that we would find

intolerable. Personal freedom was severely limited. And the opportunity

to rise above the conditions into which one was born, to even hope for a

materially better future, was almost entirely absent. But although exis-
tence was precarious in many ways, with poor nutrition and primitive
medical practice placing life and health at risk at every turn, and
although personal misery and psychological disturbance could no more
be averted than today, people experienced a kind of security that until
our era could be taken for granted in most stable societies. They were
guided by traditions and values that they accepted for the most part
without question. And they belonged; they had people they could count
on and a milieu they felt a part of.
It wasn’t just that parents stayed together and were always in or near
the home, or that extended families were often tightly knit, an ever-pre-
sent factor in one’s life. There were larger communities as well. People
identified strongly not only with their family or clan but with their

church, their village, their guild if they belonged to one, and their social
class. They worked, worshipped, celebrated, and amused themselves in
the context of these groups. They did not consider themselves to be spe-
cial or apart, but cut from the same cloth as their peers. Indeed, personal
identity was inconceivable outside these affiliations,^ which were emo-
tionally and spiritually sustaining and saturated with mutual help. They
were in every respect attachments, social attachments, as imbued with
love, rage, hurt, and happiness as a child’s attachment to his parents.

Inmany traditional societies, there was a familiarity and intimacy

even among nonequals. People lived in such close daily proximity and
were so mutually dependent that little of what one was could be hidden.
In late-nineteenth-century Russian novels, it is apparent that even the
roles of master and servant or landlord and peasant did not preclude feel-

ings of affection, devotion, dependency, and deep familiarity. Goncha-

rov’s Ohlomov is particularly poignant on this score, lamenting the
passing of the traditional order with its warmth, its certainties, and its

timeless routines in favor of the energy, opportunity, and calculations of a

newly mobile and enterprising society.
 Avoidant Society 413

The industrial revolution and the shift in customs and ethics that
went along with it caused the world to gradually change from one in
which people were born into a God'given place, with clearly ordered

roles and duties, to one in which they had to make their place. Com-
petition became a new organizing principle, and in one sphere of life

after another it was becoming impossible to feel quite the same unity

with one’s fellows. No longer cradled by unchanging, familiar connec-

tions, by homey surroundings, and by the known rhythms of nature, men
went out to work in impersonal, often crushing environments. Fre'

quently, they had to steel themselves and swallow their emotional needs.

“Dependency,” an unnecessary concept before, became an undesirable

quality in men and, much later, in women. Meanwhile, as old bonds diS'
solved or became less central to one’s life, the sphere of intimacy that

surrounded each person shrank.^

In the world that commerce and industry were creating, there was less

time for nonproductive pleasures. Daydreaming, chatting away the after'

noon, taking long midday naps, and other idle pastimes no longer fit the
quickened schedules of life and were frequently stigmatized as lazy or

lower-class.^ As the opportunity for success and the fear of failure became
a bigger factor in people’s lives, parents became anxious about formerly
acceptable qualities in their children that were now deemed slothful,

clingy, or immature. The passivity that allows one’s feelings to come to

the surface and get worked through in fantasy and contemplation was
now more likely to be seen as a waste of time. One had goals to achieve.
If one became cut off from one’s feelings, passivity became not just a

waste of time but a threat, and one’s busyness something of an addiction.

The history of the family is so complex with so many contradictions
and so many variations from era to era, from place to place, and from one
social group to the next, that it is difficult to make generalizations. But

if we think of the family as the child of society, we could say that modern
conditions have often tended to make the increasingly isolated nuclear

family anxiously attached to the culture that gave birth to it. Meanwhile,

as each home became a more private and unique place, without the pro-

tocols and rituals that had governed life before, parental psychology

became far more critical in determining whether a child would he

securely or anxiously attached.^
In the century our society has become more child-conscious per-
last
aware of
haps than ever before, concerned with what children need, and
414 The Odyssey of an Idea

the importance of early environments. We have struggled to eliminate

the routine mistreatment of children, the diseases that once ravaged them
(and that in prior times caused some parents to hold off becoming too
attached to them when they were little lest they die), and the wretched
poverty and ignorance that often deformed family life. Our accomplish'
ments in these areas have been huge. As historian John Sommerville has
observed, “Some children are probably better cared for now than ever

before in history.”^ But the social organization of the modern era has

thrown up new sorts of challenges, many of them stemming from isola^

tion and alienation, which present new barriers to secure attachment.

The middle'class patients who sought help from Freud, Janet, Jung, and
the other pioneering psychotherapists at the turn of the century and for
several decades afterwards tended to know who they were, or at least
who they thought they should be, and to feel a natural sense of belonging
and of responsibility to both their families and their social milieux. They
suffered from annoying symptoms — compulsions, hysterical paralyses,

phobias, fetishes — that they hoped to discard, hut otherwise experienced

themselves as emotionally intact.^ The new analytic patients, as the

reader may recall, did not display these obvious symptoms. They were
haunted by feelings of selTdoubt and emptiness, were prone to dislike or

hate themselves, and were unclear about who they were or should be.
During this era many of the character neuroses we discussed in the previ'
ous chapter — schizoid, narcissistic, borderline, as well as Winnicott’s

false self — came to the fore.

Heinz Kohut believed that during the early part of this century, when
psychoanalysis was first coming into being, problems in child care were

more apt to arise when the child’s first sexual stirrings began. Hence
Freud’s emphasis on oedipal conflicts and the neurotic problems that
sometimes developed in their wake. In recent times, Kohut argued,
parental failures start earlier when the parents lack the time, the energy,
the inclination, and the cultural support to provide the attuned, baby'
oriented environment infants need in the earlier years.* Such failures are

more likely to impair the child’s developing sense of self.'^

As early as his World Health report Bowlhy struggled with the societal

*Kohut was probably mainly of middle- and upper-class families. Very poor or financially
speakinfi;
destirute families, especially those with a lot of youny children or where the mother had to take in
work, would have had a hard time providing a hahy-oriented environment either in Freud’s day or
earlier. The poor, however, were rarely seen in treatment.
 —

Avoidant Society 415

changes that militated against secure attachment. In previous genera^

tions, Bowlby noted, mothers were surrounded by teenage sisters, couS'

ins, and their own mothers, all of whom pitched in with the child care.

Such helpers not only gave mothers much needed breaks from the ordeal
of taking care of one or more children, hut also companionship, guid'
ance, and confidence. The child, meanwhile, had alternate attachment
figures; and less pressure was, therefore, placed on the bond with the

mother to provide him with all his needs. “It has taken the world’s rich'

est societies,” Bowlby said, “to ignore these basic facts.

The loss of community ties also disturbed Bowlby. “Mobility is the

great enemy,” he told me. “In days gone by, people stuck around, they

saw a lot of each other. But this business of moving every five years from

one place to another is exceedingly destructive. I’ll give you an example.

The kibbutzim look after small children in an awfully silly way.”

Israeli

(Infants have traditionally slept in a dorm overseen by a caretaker

and

only spent limited time with their parents. An unusually high number of
them have turned out to be ambivalently attached.^ But once the

children are older, it’s a wonderfully stable community where everyone

knows each other and where there s a great deal of mutual support, like
a small town. Any small town in days gone by was a
community where
people knew each other, and you constantly met all your friends
when you went shopping or to church or to school or whatever. And
there was a lot of mutual help. If people know each other and have long'
help you
term relationships, mutual help makes sense, because I can
going to be
today and five years hence you can help me. But if you aren t
five years, and the community is constantly
changing, it is by
here in
definition not mutually helpful.
The pace of life can also be a hardship. Speaking of a time no more
1970s, Ainsworth says, “People used to have more
distant than the
leisure, more time for fun, sociability. Now everybody’s too busy to be

sociable. It’s sad.”^^ We seem to be, more and more, a busy, time'Con'
scious people, occupied with goals of all kinds and habituated enough to
for much of
work and advancement or other forms of stimulation that,
pleasure.
the population, sociability has lost some of its former
The effect on children of our increasing emphasis on ambition
is per'

haps most apparent in athletic prodigies, who often lose their childhoods

in training and competition, reaches most children in some way.

but it

sometimes pressured
Even infants, naturally joyful learning machines, are
learn faster and hetter.^^ New devices are being
marketed to improve
to
416 The Odyssey of an Idea

mental capacity, including a strapped'On'thc'belly tape player for preg'

nant mothers to stimulate the nervous systems of the unborn. In Phila-

delphia the Better Baby Institute trains infants who can’t yet crawl to
become intellectual powerhouses and instructs their parents on how to

carry on the process at home.^^

“I don’t think it’s a very healthy thing to be at the child too much,”
Ainsworth says, “to have him taste this and smell that and feel this, try-

ing to enrich all aspects of his life. It’s intrusive. The normal kind of
interactions that take place in the course of routines, where there is some
conversation and smiling back and forth and perhaps a little play — or in
periods that are consciously devoted to play — I think that’s what the
child needs in terms of stimulation. That doesn’t mean that the child’s

interest in other things shouldn’t be encouraged, but he’ll have that

interest if he just has a chance to explore. Stimulation is something you
do to somebody else. It’s experience the child needs.”
There’s a simplicity and naturalness to Ainsworth’s message. All your
child needs in order to thrive both emotionally and intellectually is your
availability and responsiveness. You don’t need to be rich or smart or tal-

ented or funny; you just have to be there, in both senses of the phrase. To
your child, none of the rest matters, except inasmuch as it enables you to
give of yourself. And yet this does not seem to be an easy message for our

culture to absorb. For many parents it seems simpler to apply themselves

to beefing up the baby’s IQ or working overtime to ensure that they can
pay for the best schools.

In 1975 an American writer, Jean Liedloff, wrote a book called The Con-
tinuum Concept. Liedloff spent two and a half years in a South American
jungle living with the Yequana, a stone-age tribe, and was transformed by
the experience. She saw people who were at ease with themselves, who
took great pleasure in whatever they did — even ordeals, like rowing a
canoe against an impossible current, which we might consider intolera-
bly frustrating. She saw a graceful people, living in the moment, who
were not preoccupied with everything they had to do.
The Yequana infants were carried everywhere by their mother and in
constant physical contact with her until they started to crawl. Whether
doing daily chores, dancing excitedly at a party, or sleeping at night, the

mother kept the hahy attached to her. Children were not fretted over or
given training in obedience. But they were never far from loving atten-
tion. And they grew up to he extraordinarily self-reliant and at ease with
 Avoidant Society 417

both themselves and their environment. They were also well behaved,
“never fought, were never punished, always obeyed happily and
in^

stantly; the deprecation ‘Boys will be boys’ did not apply to them.
The society Liedloff described was a world of Bowlby’s theories
come
to where attachments were deep and natural, where children fob
life,
took
lowed their mother like ducks, where “the younger generation”
behaviors
“pride in becoming like its elders,”^^ and where attachment
persisted fullyand openly over the course of a lifetime. The capacity of
Yequana adults for mutual caring and dependence is captured by an inci'
dent Liedloff describes with a young man of twenty:

I was doing my best to excise the beginnings of gangrene from

his toe by flashlight. The pain must have been excruciating.

While offering no resistance to my scraping the wound with a

hunting knife, he wept without any sign of restraint on his

wife’s lap. She was completely relaxed, not putting herself
. . .

in her husband’s place at all, but serenely accessible, as he

buried his face in her body when the pain was greatest or

rolled his head from side to side on her lap as he sobbed. The
did
eventual appearance of about half the village at the scene
or
not appear to affect his reaction either toward selTcontrol
dramatization.^^

our
Her encounter with the Yequana caused Liedloff to question
practices. With a
entire way of life but especially our child^rearing
Bowlby-like emphasis on the ancient conditions under which
our species
called the con'
evolved, Liedloff urged her readers to return to what she
into being. She
tinuum,” the natural biological flow in which we came
possible, during the babys first year, to
urged mothers not to work, if

carry him next to their body while doing their chores; to instantly

respond to his cries, just as their heart naturally dictates; and to sleep

with him at night, as mammals and primitive peoples who haven t lost

touch with their nature have always done.

moment in history, with our customs as they are,

At this

sleeping with one’s baby seems a wildly radical

thing to ad'
or hav'
vocate. And so, of course, does carrying him around,
or awake.
ing him held by someone, at every moment, asleep
But in light of the continuum and its millions of years, it is
418 The Odyssey of an Idea

only our tiny history which appears radical in its departures

from the long^established norms of human and pre-human
existence.

Liedloff and similar thinkers have inspired many people. More moth'
ers now use Snugli'like devices to carry their babies rather than strollers
or baby carriages, and one recent study suggests that this practice alone
may help promote secure attachment. Some parents have even begun
taking their babies to bed, and the influential developmentalist VIelvin
Konner has endorsed it.^^ And yet, how much can we import the meth'
ods of a primitive tribe to our own culture where the demands of social

life and the meanings we attach to what we do are so different? While

there may be limits to human adaptability, it seems to me that letting a
baby be on his own at times, letting him nap and sleep alone, do not
stretch those limits too far, as long as his care is, as Ainsworth has said,

consistently warm and responsive.

Karin Grossmann agrees with Liedloff that a very close mother'infant
connection in the first year is ideal. “It’s what the Japanese do, it’s what
many other cultures do. These children are very secure. When they are
distressed, they will go and seek comfort right away; they will not be very
afraid of strange people. But, of course, this is not enough to make you a
capable member of our society. You have to learn to be on your own.
That’s the way life is structured here. So it would be silly to raise a child
in America for the first three years in this style. He would not be able to
go to preschool. He would not be able to negotiate with his mother and
to give a little when she does return to work. He would not be fit to live
in this culture.

To emphasize this point, Karin refers to the five weeks she and Klaus
spent working among a tribe on Trobriand Island in Papua New Guinea.
The children of the Trobrianders were secure and confident in their

home environment, but when they tried to move to the more bustling
and competitive society next door, they generally failed.

It is important to remember, too, that we come to parenthood not as

blank slates but as people with specific experiences and feelings. A thou'
sand years of therapy would not turn any of us into a stone-age parent
even if we were certain that was a worthy goal. We can be inspired by
other peoples, but change has to be organic and begin with who we are,

both culturally and psychologically. Without that we are bound to stuiri'

ble. That is why when a mother decides to hehave in a totally different

 Avoidant Society 419

nianner from her own mother, she often finds her mother s shadow hang'
As Karin Grossmann observes, a mother
ing over her in unforeseen ways.
may give up her ownand career on the child s behalf, in attempted
life

imitation of primitive child care, because of her own need

for a cuddly

child that she

object of security. She then becomes “so invested in the
expects him to bring her instant and eternal happiness in return. And
when needs to explore. Such
she has a hard time letting the child go it

parents may also find impossible to accept or react constructively to

it

the child’s natural aggression. None of this is to disparage the efforts of

their
American parents who are attempting to adopt Liedloff’s ideas to
own homes. We do not yet know how well this is working, and each fam^
ily no doubt goes about it in different
ways. But replicating the Yequana

in contemporary society is virtually impossible,

and numerous compro'
mises will have to be made.
To read about the Yequana is nevertheless to feel a sense of loss.
We
ourselves; and, despite
miss having a deeper, more satisfying contact with
our belief in individualism, our tendency to look
down on those who
social support, and our disdain for government
programs, we miss
need
having a community that surrounds us with love and security.
Some of
that loss can be dealt with on the personal level
—
by coming to terms
with the past in ways described in the previous chapter.
Some of it can be
dealt with through social adjustments, by
mobilizing ourselves to make

our society more supportive and caring, especially in light ot what we

now know about attachment needs.

his followers have

Since Bowlby’s World Health report in 1951, he and
the world.
had huge and steady influence on child care practices around
a

Adoption policy, hospital policy, sociabwork policy, not to mention the

been affected. That infants
everyday behaviors of parents, have all

to parents or parental figures, that they

love
become profoundly attached
that they need to be loved and
them with every ounce of their being,
affirmed by them and are wounded by and lengthy separations is
losses
this enlightenment
perhaps better understood now than ever before. But
contlict with other social
has been patchy, inconsistent, and often in
mothers, industry’s wish
needs or wants, like the importance of careers to
the general reluctance
not to be disrupted by lengthy parental leaves, and
to spend lavishly to assure high-quality
child care programs. That a U.S.
child (the famous
court in 1993 could order a two-and-a-half-year-old
thereby
“Baby returned to biological parents she did not know,
Jessica”)
 —
420 The Odyssey of an Idea

severing her ties to the adoptive parents who raised her, without a con^

sideration of the child’s feelings, suggests that our courts and our laws
remain ignorant of important issues in child development.
In the World Health report Bowlby stated, with what now sounds like

utopian optimism, that, having conquered most of the diseases caused by

malnutrition and infection, society was free to turn its attention to men^
tal health. He advocated supporting families with a helping hand or
financial assistance in times of need, with psychotherapy, and with other
services, so that children would not have to suffer the deprivations and
other torments that are both harmful and preventable, and so that they
would not grow up to be the sorts of parents and fellow citizens one
would not wish to have in one’s midst. In the post-war period many
industrialized countries have developed such supportive family services.

Nevertheless, some developmentalists warn that child welfare is in crisis,

perhaps especially in the United States, which has lagged significantly

behind on all fronts.

That social policy and social interventions can make a society more
humane and give parents a better chance to raise secure children has
been demonstrated in numerous studies. People respond in remarkable
ways when they feel that someone is there to help them. A series of stud'
ies have now found that when mothers have solid social supports

whether from the extended family or an outside helper — the likelihood

of secure attachment is enhanced, particularly, one study found, when
the infant has a trying temperament.^^ Even Dymphna van den Boom’s
simple six hours of helpful instruction over the course of several months
to poor Dutch mothers of temperamentally difficult children (see Chap'
ter 21) seem to have been a godsend to them, enabling them to form
secure attachments where avoidance would otherwise have been pre'
dieted. Alicia Lieberman’s therapeutic work with troubled mother'infant
pairs, where the mother learns not so much better techniques of child
care but how her own emotions, based on her own childhood experience,
may he getting in the way of providing responsive care, has also been
shown to make a huge difference in the development of secure attach'
ment. Studies of women giving birth in a Guatemala hospital — where
they had previously been left alone for much of the time — demonstrated
that the constant friendly support of an untrained female helper enabled
them to have a much shorter labor (8.7 hours on the average as opposed
to 19.3), to have half the perinatal complications, and to be better able
to enjoy the early hours of their baby’s life.^^ Such support — educative.
 Avoidant Society 421

therapeutic, and warmly attentive — if provided to children, parents, the

sick, and the dying, could help relieve the sense of isolation, alienation,

and insecurity that troubles our society.

In Syracuse, New York, an extensive program of child care and
family
children
support was offered to a group of low'income families until their
reached the age of five. A
follow'up study when the children were fifteen

found them doing much better than a control group of kids from similar

backgrounds. They performed better in school and were more apt to

families who
report liking their teachers. Adolescent boys from similar
did not have access to the program during their early years
committed

four times as many offenses, and the offenses were more severe. A similar
fami'
program in Connecticut provided health and social services to poor
lies with infants and toddlers. Ten years later
the children and their

j-QOthers were doing far better than those who had not
received such early

support. The other families were costing the state an average of almost
three thousand dollars more each year in welfaie and special education.
Most parents have some social support. It takes the form of friends,

close relatives, including brothers and sisters who may also he parents

and can give them hand-meMowns and advice; and their own parents,

who may be in the position to help out with child care and to help
them
financially. They may not have a lot of support, but
enough to get by and

to give their children the kind of care they need.

Other parents lack

these supports entirely. Many would benefit from community assistance

programs. But when

of the sort offered in the Syracuse and Connecticut
it comes to providing such services, the United States is particularly
instituted a
impoverished, the only industrialized country that has not
system of child care for working parents, a guaranteed
minimum family
income, and national health care for families with small children. While

a parental leave law, small in comparison to Sweden

we now have it is

where mothers can take a year off with full pay and the guarantee that

their job will be there for them when they return.^^ A report in the New
York Times on the Australian health care system offers
one small example

of the routine support other nations provide:

When a baby is the hospital notifies an infant health

born ...
parents
center in the family’s neighborhood. Within days, the
these “sisters”
are visited by a highly trained nurse. Typically

try to get to know the mother, offer a phone number where

bathing to
they can be reached for advice on everything from
422 The Odyssey of an Idea

breast-feeding, and make an appointment for the baby at the

health center. These nurses take care of all the baby’s routine
health needs.

In such an environment, parents are probably more secure and less likely

to let anxieties impinge upon their relationships with their children.

Poverty per se does not breed disturbed family relationships. Attach-
ment research has tended to show that people can have very little mate-
rially and yet be excellent caregivers as long as they are not overwhelmed
by deprivation. But if they have to deal with the stresses of poverty, have
suffered early maltreatment or neglect themselves, and are isolated as

well from sources of support, then the chance of their children becoming
securely attached and well functioning declines sharply. Even the ability

to learn is impaired by such conditions. According to a ZERO TO THREE

report, which echoes two decades of attachment research and other stud-
ies in early development, students do well in school if they arrive having
already developed in their early years confidence, curiosity, self-control,
a sense of effectiveness, the ability to relate well to others, the capacity

to communicate, and the ability to cooperate. Children from over-

stressed homes, with few resources and little social support, often fail to

develop these qualities, leaving them with bad feelings about themselves
and little hope of success later.

And this is not a marginal issue. According to Urie Bronfenbrenner,

in the United States, unique among developed nations, a quarter of all

young children, from newborns to five-year-olds, have the added burden

of poverty. Studies, like Glen Elder’s Children of the Great Depression, as
well as a host of more recent investigations, have documented the toll

that poverty and family instability can take on the lives of such children.
(Paternal unemployment alone has been associated with unstable, un-
happy, unproductive lives among sons.)^^ Economic privation has been
associated with serious medical, psychological, and social problems that
could have been prevented. These problems include disturbed family
relationships; child abuse; lonely children coming home to empty
houses, being kept company for hours on end by television, and coalesc-
ing into destructive groups with similarly deprived youngsters; and even-
tually, of course, more alienated, dysfunctional adults who have little

stake in society, who, it could be said, lack a secure attachment to the

society, and act out their anger by being uncaring, uninvolved citizens or
worse.
 Avoidant Society 423

Developmentalists have made numerous proposals to ameliorate these

conditions. ZERO TO XhreE whose hoard of directors includes many
of the nation’s leading child development experts — Berry Brazelton,

Robert Emde, Stanley Greenspan, Joy Osofsky, Arnold Sameroff, Edward

Zigler, among others — recently put forward a series of recommendations
parent have
that are modest by Western European standards: that every
the opportunity to take at least six months off from work after
the birth

of a child to help promote responsive caregiving; that the

government

establish strong, meaningful standards for day care, mandating appropri'

ate staff ratios and training and including family subsidies to bridge the
gap between what families can afford to pay and what child care
really

costs if staff are paid appropriately”;^^ that comprehensive

health care he

available to all families, as well as decent affordable housing,

and income
Equally important, recom-
support for families below the poverty line. it

mends an integrated network of services that would enable parents

including foster parents, some of whom are caring for babies handicapped
cocaine to
by traumatic early experiences, like being born addicted to
benefit from all that we have learned about what children
need in their

Such services would provide easy access to parenting educa^

early years.
developmental specialist who works with their pediatrician and
tion; to a
or why it
could help them understand what their baby is experiencing
provides to
behaves as it does; to treatment of the kind Alicia Lieberman
disturbed mother'infant pairs; and to a friendly person who
is familiar

of the var-
with their circumstances and can point them in the direction
ious social services that are available to them.
Bronfenbrenner has lamented the fact that in the United States age

segregation of children makes it impossible for many adolescents to

relate to small children. He believes that not only should they have con-
a “curriculum in
stant opportunities to be with younger kids, hut that
caring” should be instituted from grade school on up
and that this should

include special tasks, like taking care of old people

and sick people,

which children are naturally good at if given the chance.

Teachers could also he given better training in the meaning

ot the dis-

in their classrooms.
turbing, sometimes maddening, behavior they see
anxious
They could work more closely with school psychologists so that
don’t get driven more deeply into their early attachment
mod-
children
els, as is now typically the case, hut are encouraged to open up and see
adults in the schools,
the world of people anew. Having additional lay
to serve as poten-
hired on the basis of their ability to relate to children,
424 The Odyssey of an Idea

tial attachment figures to the children who need them could, as Sroufe
has suggested, facilitate this process and take a huge burden off teachers.
The school years are a unique opportunity for troubled children to be
redirected emotionally, but that opportunity has not been seized.
There are other ways, of course, in which our society can become
more responsive, especially in relation to how we care for the sick, the
dying, and the bereaved. In each of these conditions, attachment needs
soar and we hunger for someone to hold our hand or show us the way.
Such support was once more available, provided to varying degrees by
the extended family, by the church, or by neighbors whom one knew and
trusted. The recent success of hospices in helping people to die in a posi'
tive frame of mind, the success of bereavement counseling in preventing
suicide,^^ the rage people experience over the unfeelingness of doctors all

suggest the importance we place on being cared for in a kindly way when
we need it.

Governmental efforts, of course, can never adequately fill a gap where

warmth and connection are wanted. The emotional challenges of the
current era will still need to be faced more personal, ways. But
in other,
social programs can perhaps create conditions where human connections

have a better chance to flourish.

Implementing programs that take attachment needs seriously would
be very expensive —
a lot more than the supercollider, though perhaps
less than the Savings and Loan bailout or going to war. There would be
savings, too, just as there always are when preventive programs are insti'
tuted, especially where children are concerned. The hills for later welfare
services would probably he reduced, as would the destruction caused by
alienated youths and the cost of expensive incarcerations. We would also
have a bettet'earning, tax^paying work force among the groups at risk.
But, most important, it would offer us a different quality of life, where we
would be taking care of human needs that most societies attended to
before the industrial era; strengthening the family, which is the best
institution we have for meeting the needs of children; enjoying a greater
social cohesion, with less cause for segregation and fear.

But we are nowhere near making any of these commitments because

our social policies are shortsighted. Americans, for instance, are livid
about crime, infuriated by beggars, indignant at the antisocial behavior
of minority or poverty youths. But our response to these social ills is like
that of Aladdin’s uncle who, you may recall, sent the hoy into the under'
ground chamber in search of the magic lamp. With Aladdin struggling
 Avoidant Society 425

up the stairs, the uncle made louder and louder demands that he hand up
the lamp while refusing to give Aladdin the hand he needed to emerge

from the cramped stone stairwell. When Aladdin could not produce the
lamp because he was too weighed down, his uncle, impervious to the
obvious solution and incensed at being thwarted, ragefully slammed shut
the overhead door, thereby abandoning his nephew and the lamp in one
swat. This is America today, stubbornly committed to threats and pun-

ishments (and moralistic exhortations regarding family values) as the

preferred method to bring about changes to problems whose causes are in
large part social, economic, and psychological. Although police action
and criminal always important, no amount of threat or punish'
justice is

ment can reverse the alienation and hostility of those who have paranoid
of
or infantile character structures, who are overwhelmed by feelings
being cheated and deprived, who have no faith in their ability to nourish
or be nourished, and who face every encounter with a lover, a child, or

society itself with the expectation that they will exploit or be

exploited.

People like this will cost their communities a bundle whether or

not they

arehandled in an enlightened way. But at least by providing psychiatric

and social supports, we can hope to improve their behavior, lessen future
damage, and perhaps give them some release from their inner demons.
costs to
Meanwhile, nothing would serve us better than to pay whdtcvcT it

help families give their children adequate, loving care so that such
pet'

sonalities will be less prevalent in the next generation

and to keep —
some unworthies take advantage. Investing in secure
doing so even if

attachment is not a cure'all by any means, but it is a win-win proposition

for everyone involved. Unfortunately, like Aladdin s uncle, we would

rather pay through the nose in punishing.
 28
Looking Back:
Bowlbv and Ainsworth

John Bowlby died of a stroke on September 2, 1990, at the age of eighty^

three. He was staying at his beloved vacation home on the Isle of Skye in
Scotland where he had spent many summers with his family, gardening
and taking long walks over the rugged, majestic landscape. He was buried
there, as he wished, before a group of family and local friends, in the

graveyard of a ruined pre-Reformation church. At the service, which

took place at another church nearby, which he and Ursula had attended
for many years, the minister read a passage from Pil^im’s Progress about
the passing over the river of the character. Valiant for Truth, as “the
trumpets sounded for him on the other side.”
Earlier that year Bowlhy had seen his last work, a biography of Darwin
— a surprising departure into a new realm — published to near universal
acclaim. Frank Sulloway, himself a Darwin scholar, called it “a remark^
ably sensitive and revealing portrait.of Darwin by a selhtrained admirer
who came to know more about the man than many Darwin scholars
do.“‘ He considered it a model psychohiography, a significant improve-

ment over Freud’s efforts in that genre.

 Looking Back 427

Bowlby’s Charles Darwin is both a biography and an attempt to clear

up a medical mystery, for Darwin had all his life suffered from a debilitate

ing anxious condition, which included bouts of nausea and vomiting and
which had never been, in Bowlhy’s mind, satisfactorily explained. Bowh
by attributed Darwin’s troubles to a sequence of unhappy childhood
experiences, of which the most telling was the loss of his mother at age
eight. After her death his father and sisters were so aggrieved they
refused to speak of her or allow her name to be uttered, and the wall of
silence” made it impossible for Darwin to mourn her loss. Years later he

seemed to have even forgotten that she’d existed or that the trauma had
ever occurred. Add to this, difficult relationships with his father, of

whose love and respect he never felt certain, and his reproving older siS'

ter, who undertook to fill the maternal role, and Darwin becomes a

poignant study in spoiled attachment.^

Bowlby had many reasons to feel connected to Darwin. He believed

that much of his struggle with organized psychoanalysis had represented

his effort to “recast” analysis “in terms of modern evolution theory.”^ He
saw himself as an evolutionist and believed that the formation of affeC'
tional ties — the development of love —
was one of the greatest achieve-
ments of human evolution. Bowlby also admired Darwin s personal
qualities, many of which foreshadowed his own — the courage, the tenac-
ity in the face of staunch and hostile opposition, and the warm and gen-

erous style with which Darwin conducted himself in his writings.^

But

knowledge of Bowlby’s own childhood experience the coldness and —

emotional unavailability of his parents, his being sent away to boarding
school at age eight —
might lead one to believe that the connection he
felt to Darwin was also based on a shared sense of loss, a shared sense ot

not having had his deepest self responded to. One can imagine that he

made much the same connection with the maladjusted children at the
progressive schools where he briefly worked as a young man and with the
child patients he treated for many years thereafter. Their pain had some-

thing in common with his.

Bowlby acknowledged that he was not an intuitive therapist and that

he had to rely on theory to discern what his patients were
struggling

with.^ But despite this and despite his cocksure intellect, he

was not kill
without irony, a
of himself, he was not puffed up, he was, as his wife says
very plain, simple man.”^ A colleague, Hyla Holden, similarly referred to

his tormida-
“the straight-forward and loving simplicity which lay behind
428 The Odyssey of an Idea

hie intellect.”^ Surely, his patients must have appreciated this in him. He
rated himself a good listener, and one can imagine his asking earnest and
sympathetic questions as he homed in on a point.

Friends and colleagues remember him as having been reliable. “You

could count on him in every way,” wrote his former subordinate, psychia^
trist Freda Martin.^ “1 have images of him, even last winter,” Tavistock

familty therapist John Byng^Hall wrote the year Bowlby died, “shaking
the rain off his green mackintosh and hat as he arrived on time for some
evening meeting, while others sent their apologies. Despite his severe
and sober countenance, he could readily loosen up and be fun. “I can
vividly see him,” yet another colleague recalled, “leading the Scottish
reel with Mattie Harris at the magnificent Christmas parties held at the
Institute of Psychoanalysis.”^^ But he was not emotionally accessible, was
unable to express affection verbally, and outside the therapy room with
its carefully prescribed roles, he was not drawn to the emotional affairs of
others. He valued his many friends, thought only well of them, and pre-
ferred not to go any deeper. Regarding his personal relationships, Ursula
Bowlby has said, “John wasn’t curious about people, about ‘how they
ticked,’ and he was very incurious about himself. This seems an absurd
statement, considering his career, but anyway it is true. . . . Perhaps he
was fascinated by patterns — how the pieces fitted into the jigsaw.”^^ All

this, of course, is a hallmark of the avoidant personality. But, no doubt

like many avoidant personalities, Bowlby did manage successful and fuP
filling relationships. His well of feeling, shapeless and unarticulated as it

may have been, could be touched by others. And perhaps especially with
children — including those who had become detached, mistrustful, and
mean — he could resonate with their suffering, even if he was not able to
suffer himself.

Bowlhy lived long many of his once heretical views widely

enough to see

accepted. Hotly disputed for many years, his early work on separation
and deprivation is barely questioned now, except by those who argue
not that such experiences aren’t damaging but that the damage may he
reversible under optimal conditions. In psychoanalysis, the idea that

parental neglect or mistreatment is a source of later emotional distut'

bance is no longer questioned by anyone. And in developmental circles,

behaviorist objections to other aspects of attachment theory no longer

command attention. It is perhaps a measure of how thoroughly some of
his concepts have been absorbed that a review in The Psychoanalytic
 Looking Back 429

Quarterly of a collection of his short pieces faults him for stating the

obvious, the reviewer apparently unaware that what is obvious now was
anathema when Bowlhy first spoke it.^^ He took pride in the fact that
even some former detractors quietly admitted to him that he had been
correct all along.
In his last decade, his status as a significant innovator, reformer, and
theory builder was recognized throughout the West, while his recom^
mendations were built into the practice of social work and became part
of the standard advice that baby doctors give to parents and lawmakers.
Many honors were bestowed on him, the one he valued most being the
honorary degree he received from his own university, Cambridge, in
“1 come from a
1977. “Fortunately,” he said of the belated acceptance,
long'lived family.”^"^
After completing the third volume of his trilogy, Bowlhy offered a

reporter the following, typically understated assessment of his accom^

plishment: “Look back thirty years, and there were four theoretical para^
digms that were totally distinct from each other” psychoanalysis, —
learning theory, cognitive psychology, and ethology. “There was no con^
versation between any of these four fields, and they each had a different

set of assumptions. . . . One of the things I’ve been concerned about is to

try todevelop some sort of theoretical framework that does justice to the
contributions each has made, so that there could be interchange and
communication, and data drawn from different fields. 1 think that’s a use^

ful thing to have attempted.

In a more formal, if less restrained, review of the final volume Rycroft
wrote: “Few people, it seems to me, are today as persistent and con'
sistent, as singlc'minded, as Dr. Bowlhy has shown himself to he; his

attitude towards his life’s work and achievement has a Victorian monu'
mentality about it which is enviable and all too rare.”^^

Bowlhy ’s influence has extended to many disciplines, from ethology,

where animal studies based on his theories have been conducted, to din'
ical work with patients, to personality psychology, to social psychology,

where ever more basic attitudes and conditions, such as the way one con'
ceives of one’s relationship to God, are found to relate to attachment
style. (One can easily imagine that the quality of one’s patriotism is

related as well.)
Bowlhy was perhaps most proud of his gradual acceptance by his psy'
choanalytic family, which had once driven him trom home. He always
wanted to he considered an analyst and must have been hurt by those
430 The Odyssey of an Idea

who he wasn’t (one analytic journal referred to him disingenuously

said

as an ethologist)d^ It pleased him that, especially in later years, he was

asked to speak repeatedly to analytic audiences around the world, includ'

ing the Anna Freud Center in 1988. When he turned eighty the British
Psychoanalytic Society, somewhat belatedly, held a celebration in his
honor, which at least one former Bowlby antagonist recalls as “a con-
vincing and moving reunion.”^® Bowlby’s name is still distasteful to some
stalwarts in the Kleinian camp, so that one can still hear dismissing com-
ments about him at the Tavistock Clinic, where they remain in the
majority. But even there the mental framework has changed, and the
influence of the eminent Kleinian analyst Wilfred Bion has allowed a
focus on the early real-life experiences that were Bowlby’s passion. Some
analysts, however, continue to believe that he abandoned psychoanalysis
or turned hatefully against it. And only a tiny few are aware that he and
his followers have succeeded in making a science of something that ana-
lysts deal with every day in their clinical work but have tended to ignore

in their theory: the influence of the early environment. Most, as of this

writing, have not even heard of Mary Ainsworth.

And yet this state of ignorance has begun to change, especially among
the younger generation. Articles have begun appearing in some analytic

and psychiatric journals attempting to integrate attachment concepts

into the main body of analytic thinking or offering introductory over-

views of attachment research. Eminent Freudians, like Americans Fred

Pine, Otto Kernberg, and Albert Solnit refer to attachment contribu-
tions with respect. And others have begun to realize that far from being
a threat to psychoanalysis, attachment research offers a degree of scien-
tific support that it has never before enjoyed. Toronto analyst Morris
Eagle, for instance, notes that research on avoidant attachment provides
the best empirical evidence on repression and its consequences.^^ The
work that Main initiated on how we internalize early experience offers

similarly powerful confirmation of traditional analytic concepts like

transference and repetition, while simultaneously providing a means to

explore them further scientifically. As more children from the Sroufe,

Grossmann, and Main samples are given the Adult Attachment Inter-

view,new sources of data, of special interest to analysts, will be available

about how early experience is incorporated psychologically. If some of
these subjects, whose early attachment histories are known, can eventu-
ally he seen in some form of research-oriented psychotherapy, as Nancy
 Looking Back 431

Kaplan hopes to do one day with the Berkeley sample, the results might
certainly captivate the psychoanalytic world. Meanwhile, Peter Fonagy
and his colleagues in London, with their work on selLreflection, internal

representations, borderline states, and other forms of psychopathology,

has pioneered the long'overdue integration of attachment theory with
more traditional analytic thinking, yielding important developmental

and clinical insights.

Politically, things have changed, too. Anna Freud and Melanie Klein
are dead, as is the sanctity of drive theory (see Chapter 7). As many ana-

lysts, unencumbered by the old debates, come across Bowlby’s ideas

today, they find in them a natural extension of their own thinking. The
third issue of the New York journal Psychoanalytic Dialogues noted BowL
by’s passing with a strong appreciation. The statement, written by Lewis

Aaron, ends, “The editors dedicate this issue to John Bowlby, who taught
so much about the importance of mourning our losses.”^^
Indeed, those who champion his views in the analytic community
believe not only that he deserves to be considered a genuine psychoana^
lyst but an invaluable one. “To my mind he is one of the three or four
really great psychiatrists of the twentieth century,” the British analyst

Anthony Storr wrote in The Lancet . Storr later told me: “1 think he

really saved psychoanalysis from being totally discredited. By writing as

he did, and giving scientific status to everything he said, he took his

findings out of the morass that psychoanalysis got into and gave it some
scientific respectability. He had a much more commonsense approach to

many of the things that Freud postulated, and where he disagreed with
Freud, I think he was nearly always right. This is, needless to say, a

minority view, especially in England, where Bowlby is still less appreci'

ated among analysts than elsewhere, but one that is no longer unthink'
able.

In 1984 Jeffrey Masson published a book called The Assault on Truth in

which he accused Freud of having been dishonest and cowardly in aban'

doning his original theory of neurosis —
and in particular that hysterical
symptoms adulthood were caused by childhood sexual molestation. In
in

reviewing the book, which he called “distasteful, misguided and at times

silly,” Rycroft wrote, “If Masson had really wanted to reintroduce a trau'
matic theory of neurosis, he would have done better to base it on
Bowlby’s work on separation and loss, instead of attempting to resuscitate
Freud’s seduction theory.”^^
432 The Odyssey of an Idea

On October 16, 1991, the British Psycho^ Analytic Society held a me'
morial in Bowlhy’s honor. Three people spoke, but, according to Ursula
Bowlby, “It was the middle paper, by Eric Rayner, that hit the jackpot.”
Rayner reviewed the development of Bowlby’s ideas, his struggle with

Melanie Klein, his report on juvenile thieves, his explosive monograph

for the World Health Organization on maternal deprivation, his deveh

opment of attachment theory, the supporting research on secure and

insecure attachment coming out of the United States, the writing of his

Attachment and Loss Rayner also spoke of the shameful organized

trilogy. .

attack that drove Bowlby from any further presentation of his views at
the Society’s scientific meetings; the extraordinary value that child ther^
apists had found in attachment ideas; and the usefulness of the attach'
ment metaphor in understanding social issues — such as the disaffection

that a whole portion of society can experience when it feels left out,

demeaned, uncared for. He noted that in terms of social policy Bowlby

was the most influential psychoanalyst that had ever lived, but that most
psychoanalysts had ignored him because his scientific interests didn’t

seem to have much relevance to their clinical work. Rayner challenged

his fellow analysts to decide whether an evolutionary, biological orienta'
tion and an openness to scientific research should have a place in ana'

lytic thinking. “Perhaps we should . . . debate his theories again,” Rayner

concluded. “If nothing else, our therapeutic frame ot mind is altered by

theory. John Bowlby was a great alterer of frames of mind.”^^

Rayner’s was a modest “when he sat down,” Ursula Bowlby

eulogy. But

recalls, “there was spontaneous applause. Even the ranks of Tuscany,” she
adds, delightedly quoting Macaulay, “could scarce forbear to cheer.”^^
Although most of those present were analysts from the “middle group” —
that is, independents like Bowlby himself who had not aligned them'
selves with either Melanie Klein or Anna Freud^^ Rayner was amazed —
hy the strength of the response.
Bowlhy did not investigate the subtleties of clinical experience, which
is the everyday bread and butter of analytic work. Indeed, in later
decades, he largely gave up doing treatment altogether, in favor of writ'
ing, training an army of psychotherapists, assessments of incoming
patients, and clinic management. (He once told Ursula that sitting for

five hours a day and listening to patients was “no life for a man.”^^)

Given his interests, it is hardly surprising that Bowlby has not achieved
the status among working analysts of the other great innovators of his

era, like Fairbairn and Winnicott. He offers something different, some'

 Looking Back 433

thing that analysts are beginning to cautiously consider from a distance.

It’s not what they’re used to, hut with interest in infant research rising
throughout the analytic world, their appreciation is growing.

Inge Bretherton has written, “One notable talent that stood Bowlby in

great stead throughout his professional life was his ability to draw to him^
self outstanding individuals who were willing and able to help him
acquire expertise in new fields of inquiry that he needed to master in the

service of theory building.”^ ^ Ethologist Robert Hinde is the outstanding

example there. But Bowlby generally drew outstanding people to himself,

not just experts from other fields. These included Jimmy Robertson, who
had the passion and energy to tackle hospital policy; Peter Marris, whose
book Widoies and Their Families had so influenced Bowlby; Colin Parkes,

who has used attachment principles with the dying and bereaved; among
others. None, however, compared to Ainsworth.
When I met her in 1988, Ainsworth was a bright^eyed woman of sev-
enty'Six whose mode of engagement changed readily from intellectual
delight to professorial instruction, from easy laughter and fluency to a
flustered search for the right word. She lived in semiretirement near the

University of Virginia in Charlottesville, where she had taught for many

years. She still attended child development meetings, where her voice

and analysis carried great weight. (That activity was halted by a series of

small strokes in 1992.) In discussing her work, Ainsworth revealed a

strong, sometimes and an uncommon willingness to credit
bristly pride

others. The penetrating gaze she trained on me at times was suggestive of

her years as a teacher and clinician. Although more self-aware, more

interested in herself as a psychological being, Ainsworth seems to share

some of Bowlby ’s avoidant features and, like him, has never found it easy
to express feelings. Her former student Robert Marvin, also of
Char-

lottesville, who, along with his wife, Cheri, has been like family to

Ainsworth, feels that he was aided in becoming close to her because he

himself came from a family in which feelings had to be read between the
lines.

With her two pioneering longitudinal studies, Ainsworth made the

most dramatic effort to take attachment theory into the field. But she did

much more than that. She gave Bowlby early theoretical support — one
thinks of her seminal papers on maternal deprivation (1962) and depen-
dency (1969) —
and was instrumental in introducing his ideas to devel-
opmental psychologists. With her clinical and diagnostic skills (it should
434 The Odyssey of an Idea

be renieiTibered that she was an established Rorschach expert before she

ever met Bowlby) she led the way to a practical study of attachment
themes, enabling other scientists to consider and explore them further.
That was the great meaning of her ingenious creation, the Strange
Situation. With that in hand, she was able to mobilize the better part of

an entire discipline to follow in her footsteps.

Bowlby described his extraordinary connection with Ainsworth as one

of the happiest he’s had (“a mutual admiration society,” in the words of

his wife^^). Her visits to London were a treat for him, as the two of them
barricaded themselves in his office in the Bowlby home and read each

other’s work. He saw their four decades of nearly unbroken collaboration,

with manuscripts regularly flying back and forth across the Atlantic for
critiquing, as having been invaluable. He did seem to harbor a touch of

guilt about her long comparative obscurity. When we spoke in 1989, he

still recalled with regret a ground-breaking report she wrote for him that,

because of prior commitments he had made, was published in an obscure

journal. His final book on attachment, A Secure Base, was dedicated to

her.

Although she never had children of her own, Ainsworth did become
the matriarch of the far-flung but close-knit family of attachment re-
searchers, many of whom were intellectually nurtured by her since their
graduate school days and continued to see her as a guiding force in their
work. They, in turn, have helped make her one of the biggest names in

developmental psychology since Piaget. Although still unknown to the

general public and to most clinicians, her fame in the world of infant
development exceeds that of Bowlby himself.
“Bowlby does the theory,”^"^ Ainsworth used to say, as if he were the
commander in chief and she the field marshal. It was an arrangement
that reflected the side of her that emerged in this relationship. Despite

her own prominence, with many followers who barely know Bowlby’s

work, to Ainsworth he remained the senior partner. That they were not
riven by the jealousies and competitiveness that have destroyed so many
other scientific enterprises may have something to do with a supportive
femininity that came out with Bowlby and that men like Bowlby thrive
on. “I think that women on the whole are much readier to take the lead
from a male mentor than the other way around,” Ainsworth says. “It’s
not just Bowlby. saw this with Bill Blatz in Toronto, way hack when. He
1

had a great following of women who worked in the nursery school or

 Looking Back 435

with parent education and who seemed to he disciples. But he could

never find a man to work with him for any length of time.”^^
Like Bowlby, Ainsworth is an impressive thinker and writer. The suC'

cess of attachment theory no doubt owes a great deal to this concentra'

tion of communicative talent. She writes as if the material has personal

meaning must have contributed to the influx of stU'

to her, a quality that

dents who came to study with her and carry on her work. That she was
also a creative teacher with a unique approach to child development

caused some of her undergraduates to want to continue in the field. And,

like Bowlby, too, she was able to attract outstanding people Mary —
Main, Inge Bretherton, Alicia Lieberman, Jude Cassidy, Roger Kobak, to
name but a few, and, indirectly, Alan Sroufe, who has also attracted out-

standing people. attachment research in American universities is

If

“zooming,” to use her word, it is partly the result of the personal power of
this shy, strong-minded woman.
To her students, Ainsworth has always been a formidable and domi-
nating figure, no-nonsense in her approach to the research, and hardly
self-effacing in the presentation of her views. But there has always been
the self-doubting side. “I was pretty insecure as a child, and I suppose I

never really let it was turned back with a severe criticism or

go. If a paper

a grant proposal turned down as having no value, 1 would immediately

think, well, maybe I’m just no good; maybe there isn’t anything at all to
this thing I value so much.”^^ It seems fitting, when placed alongside

Bowlby, who seemed blissfully unfamiliar with the experience of self-

doubt, that, while he did the grand synthesis, she was the one to clarify
the origins of some of the commonplace insecurities that haunt our daily
lives.

Through many of the early years, Ainsworth’s work was subject to blis-

tering attacks. The revolutionary style of her research — the shrewd

move from counting behaviors to attending to the meaning of behavior

and the context in which it occurs — not to mention the conclusions she
drew about the importance of sensitive, responsive, consistent parenting,
so broke with the conventional frame of mind in developmental
psychol-

ogy, and so challenged the popular wish that both parents

could work

full-time with little impact on their children, that many were angered by
it. Like Bowlby, who weathered vitriolic attacks tor suggesting that chil-

dren are harmed by prolonged separations from their families, many of

the attacks on Ainsworth were bitter and impassioned. Sroute,
who
 —
436 The Odyssey of an Idea

observed the vindictiveness at professional meetings, was impressed with

her “grace and dignity” under fire.^^

“One of the things that gave me more and more confidence,” she told

me, “was that things always worked out the way we hypothesized they
would. We gave negative evidence every chance to emerge if it was
there, but the whole thing has been so reliable in that sense.” But hers
was never the implacable Bowlby^style confidence. When Ainsworth
says, “Imagine the surprise I had, beginning in the eighties, when began I

to get awards it does not sound coy. She did not expect them. But by
the end of the decade, few psychologists had as many.
Entering the spotlight both pleased and embarrassed her. Unlike
Bowlby, who held the light as if he were born to it, Ainsworth never
seemed at home. “It sounds corny and modest,” she told me with a touch
of urgency, “but it’s the ideas I’ve been so enthusiastic about and so eager
to put forward, not myself. You ask whether it took a lot of patience to do

those longitudinal studies. Well, yeah, it takes patience, because 1 don’t

think there are any useful shortcuts. But it never felt that way to me,
because I find the firsthand details so awfully interesting. The data cob

lections for those longitudinal studies were among the most interesting

things I’ve ever been into in my life.”^^

The change that Ainsworthism has wrought in developmental psy^

chology in the last fifteen years can hardly be measured. “Our whole
developmental approach was cognitive until she came along,” Berry
Brazelton says, referring to psychology’s prC'Ainsworth emphasis on such
functions as perception, memory, and abstraction. “She enabled psychoh
ogy to look at the emotional development of children in a reliable, quan^
tifiable way.”"^^ Sroufe goes further: “I’ve always said that if there were a

Nohel Prize for this kind of thing, she would get it. It’ll never happen.
Some of the implications are too subtle. It’s not characterized by a spe-
cific, in^time breakthrough, like the isolation of a gene. But when I went
to school I was taught that only behaviors were real, not relationships
they didn’t exist. Ainsworth demonstrated that there can be a psychoh
ogy of relationships and that relationships can be measured. That’s why
you get Nohel Prizes, isn’t it?”'^'

Even those who strongly support the attachment idea have reservations
about various aspects of the theory or research. Roger Kobak has stressed
to me repeatedly his concern that attachment theory does not take
enough account of later developments in childhood and adolescence.
 Looking Back 437

and the ways in which each new stage brings new energies and new
opportunities to work through earlier problems. Morris Eagle, one of a
handful of psychoanalysts who have made a serious study of attachment
themes, believes, as 1 myself do, that in focusing so much on experience

Bowlby eventually lost touch with the child’s fantasy life and its in^
evitably irrational components. Also, partly as a legacy of Bowlby ’s bat'
ties with Kleinians and orthodox Freudians,who were very interested in
such matters, and partly because of Bowlby own limitations, attach' ’s

ment theory, as it now stands, does not take much account of the child’s
sexuality or aggression.
Bowlby complained that he could hardly be faulted for not studying
everything. He studied what interested him and he took it as far as he
could. But, as Rycroft notes, one does get the impression at times that he
sees attachment not as being one factor in emotional life, but the key to

it all."^^ And when this happens one becomes more troubled by the ele'

ments it leaves out — from the complexities of fantasy life to the turmoils

of the oedipal era — and thus more troubled by the possibility of oversim'

plification. Chicago analyst Michael Basch sees development as progress'

ing in sectors — attachment, autonomy, creativity, psychosexuality, and

the feeling'thinking system — that sectors influence one another but

develop along separate lines. “1 think it is important to emphasize that

attachment is only one very important aspect of development,” he says,

“albeit one that significantly affects the other sectors.

Clearly, then, attachment theory is by no means without flaws, holes,

and huge unanswered questions. Various studies suggest that we cannot
be as confident as we once were about the parenting styles that lead to
ambivalent vs. avoidant attachment or the degree to which inborn tern'
perament or cultural mores may also play a part. (Is the ambivalent child
constitutionally irritable? Is the mother erratic in her attentions? Is she

overly attuned to the baby’s signals of fear?) We still do not know enough
about the critical role of the father — a legacy of Bowlby ’s mother'centric
views and of the fact that mothers have been much more accessible to

study. We do not know how quality of attachment to same'sex or oppO'

site'sex parent plays out in later development or how the child’s iden'
tifications affect his attachment patterns. We know even less about

attachment issues with siblings and other nonparental figures, like grand'

parents and nannies. We do not know what goes on in the parent'child

relationship in the years after age one that may contribute to the security

or insecurity of the older child’s attachment.

438 The Odyssey of an Idea

How is attachment played out during the second year, which must
still be loaded with attachment issues, including the period that Mahler
has labeled “rapprochement,” when the child is full of internal conflict

over his growing independence? Is it still a fundamental factor, and

therefore still readily accessible to change, during the oedipal years, or is

it now more entrenched and acting to help or hinder the child’s and par-
ent’s capacity to deal with new struggles that arise? What about the
school years of middle childhood? What about puberty? And how does
an insecurely attached child of one develop qualities that suggest security

later on?
How central is quality of attachment to character? How does it relate

to sexuality and aggression, two aspects of emotional life long regarded

by analysts as playing a central role in development but about which
attachment theory has so little to say? Does insecure attachment neces-
sarily lead to neurotic character? Does a consistently secure attachment
necessarily preclude it?

Quality of attachment seems to be an important factor in those prob-

lems of identity and relatedness that are believed to develop in the first

years of life (the “pre-oedipal period”). Does that mean it is not a factor
in problems that develop later on? Does it mean that early security pre-

cludes early disturbance? I think the answer to both these questions is

no; but if that is true, then pinpointing early attachment’s role in adult
psychology becomes even more complicated.
What are the implications for character development of a child who
has an ambivalent relationship to its mother, an avoidant relationship to
its father, and a secure relationship to its nanny or grandmother or sister?
Why do we recognize that such multiple internal representations can
exist in children and yet give adults only one attachment classification?

Are the four major attachment categories or the eight or nine subcate-
gories distinct and separate, as Ainsworth believes? Or do they become
fuzzy at the edges and tend to blend like colors in a spectrum? Does it

make sense to talk of the securely attached as having avoidant or

ambivalent leanings? And there is more.
But I think even the unanswered questions suggest the impressive
scope of attachment’s reach, how important future attachment research
will he, and how useful are its metaphors. Also, there is no reason to

believe that at least some of what is missing in attachment theory as it

now stands cannot he incorporated once the appropriate attention has

been paid. That longitudinal studies in the home have not been ex-
 Looking Back 439

tended yet beyond the first year means that attachment theory still has a

lot to learn not only about what sorts of parental behaviors sustain secure
attachment in later childhood or how attachment issues manifest them'
selves in the evolving parent-child relationship, but about how later

developmental issues — all the changes that are going on in the child’s
mind and body — get incorporated into its internal model. Alicia

Lieberman, author of The Emotional Life of the Toddler, has been strug-

gling with this question as it pertains to the second year. “I’ve been very

interested in the question of assertiveness in the second year of life,” she

says. “And in sexual curiosity — exploring one’s genitals and checking

out the differences between boys and girls, worries about how one was
made versus how others are made — and the effects all that has on mood

and self-concept. All these cognitive and emotional developments must

have an effect on the parent-child bond and on the internal organization
of attachment. I think attachment has become a little too isolated from
other motivational systems within the child, and this becomes particu-
larly apparent in the second year of life.”"^^

Much has been debated about Ainsworth’s categories of attachment,

which in the beginning seemed too simple to represent complex human-
ity. The idea that they were simplistic has, I think, been refuted, both by
their enormous metaphorical power and the way in which the same three
or four styles show up in so many areas of thought, feeling, and function-
ing. But, like the parallel classification of parents — and respon-
sensitive

sive, inconsistent or chaotic, hostile and rejecting — they do inevitably

miss much of the poetry and texture of any individual life.*^^ Anyone who
treats patients in psychotherapy will find a knowledge ot anxious attach-
ment useful and at times enlightening. But it is only one piece of the puz-

zle that any personality presents. In discussing the avoidant-dismissive

attachment style throughout this book, I have tried to give some breadth
to the category and acknowledge that it can encompass many different
personalities from the well-functioning person for whom certain portions

of the inner world are off limits to the person suffering with a
schizoid

disorder for whom any act of relating arouses anxiety and threat. But who
would guess might also include a John Bowlhy, who could write so
it

movingly about mother-love and the feelings of small children and of

whom a colleague could say after his death that it was “hard to imagine a
more fulfilled or completed life”?"^^ The mysteries and special adaptations

of any individual are unavoidably compromised by classification.

The fact that much of the impressive data coming out ot attachment
440 The Odyssey of an Idea

research based on correlations must also give one pause. Even high
is

correlations, like Sroufe’s data on the number of anxiously attached

babies who act overly dependent in later years or Main’s on the number
of parents whose children fall into an attachment category analogous to

their own, still include a lot of individuals who do not behave in the
expected manner. Some of this is no doubt due to the fact that measure-

ment tools in the social sciences are always imperfect. But it also seems
reasonable to assume that more is operating here than meets the eye.
Bowlby himself was well aware of the fragility of all theories, that, no
matter how good for their time, their usefulness must one day come to an
end. But, for now, the usefulness of attachment theory seems to be rising.
The work on family relations, the impact of fathers, the differential effect
of insecure attachment on boys and girls, the differential impact of
mothers and fathers on the same- and the opposite-sex child has all only
just begun. The investigation of adult attachment, how it affects person-

ality style and relationships — and the fact that the way a parent repre-

sents his own attachment experiences in his mind predicts the quality of

the relationship he will have with his child — interest and research in

this area is nothing short of explosive. The use of attachment principles

in child, adult, and family treatment is also growing. That clinical work
can be assessed to some extent by the use of the Strange Situation — as

Lieberman has done in her work with distressed Bay Area families (see

Chapter 17) — suggests the uniqueness of this tool.

Although Bowlby was a great theory builder and fascinated by all the

questions psychoanalysis and attachment theory touched upon, his first

concern seems to have remained what it was for him as a young man, the
well-being of children. “I’m the kind of person,” Bowlby told me, “who
identifies a typhoid bacillus and says. Look, if you let typhoid bacilli get
into the water supply, there’ll he trouble. That’s been my job in life.”"^®

His monumental efforts at theory building were thus mainly in the ser-

vice of social change. He hoped that the lasting value of attachment the-

ory would “he the light it throws on the conditions most likely to
promote healthy personality development. Only when those conditions
are clear beyond doubt will parents know what is best for their children

and will communities he willing to help them provide it.”"^^

As wedded as she is to scientific method, Ainsworth, too, has the

heart of a reformer. It has been her mission to prove that how we respond
to our children when they are very young is of singular importance. Of

her work and the work that has grown out of it she has said, “It’s more a
 Looking Back 441

matter of faith than anything else, but I do think it has great relevance to
the welhbeing and happiness of mankind. It sounds corny, and I don’t go

around shouting it from the rooftops, but that’s what’s behind the whole
thing as far as I’m concerned.
In his review of James Agee’s 1957 novel, A Death in the Family,

Dwight Macdonald wrote that it was “an odd book to be written by a

serious writer in this country and century,” partly because of the way it

dealt with love. “It is not sexual, not even romantic; it is domestic —
between husband, wife, children, aunts, uncles, grandparents.”^^ This
was the love that stirred Bowlby and Ainsworth.
Starting with a simple concept, which seems self-evident now that it’s

been stated — that a child needs to be lovingly attached to a reliable

parental figure and that this need is a primary motivating force in human
life — the work of Bowlby and Ainsworth and the many others they
inspired has helped fill an unseen void that wended its way through
much of social and medical science. It’s a void that still exists in many
respects, and it’s a concept that still meets with resistance: It doesn’t

always fit comfortably with the lives we have built for ourselves. Modern
society has taken many of us a long way from a life centered on the plea-

sures and pains of being connected to others. Our focus is often on other
things — achievement, power, acquisition, romance, excitement. But the

need for proximity, for felt security, for love; the need to be held, to be
understood, to work through our losses; these basic themes of attachment
are to some degree built into us biologically. We have mixed feelings

about them. But they are there.

 APPENDIX
Typical Patterns of Secure
AND Anxious Attachment

NOTE; The following chart is meant only as a convenient guide and

does not take into account many of the complexities and exceptions

found in the research. It should he remembered that anxious attachment

(avoidant and ambivalent) is not always associated with the styles of par^
enting described here, but can sometimes come about for other reasons;

that a child often has a different pattern of attachment to

mother and

father; and that attachment patterns can change, so that, while many
avoidant babies, for example, continue in their early pattern, others
do
later into a
not end up behaving like an avoidant six-year^old or develop
dismissive adult.
 —

444 Appendix

Securely Avoidantly Ambivalently

Attached Attached Attached

Mother (or primary Mother is often emO' Mother is unprc'

caregiver) is warm, tionally unavailable dictable or chaotic.

sensitively attuned, or rejecting. Dislikes Often attentive but

consistent. Quickly “neediness,” may out of synch with

responds to baby’s applaud indepen' baby. Most tuned in

cries. dence. to baby’s fear.

Baby readily explores, By end of first year, Baby cries a lot, is

using mother as baby seeks little phys' clingy and demand'

secure base. Cries ical contact with ing, often angry, upset

least of three groups, mother, randomly by small separations,

most compliant with angry with her, unrc' chronically anxious in

mother, and most eas^ sponsive to being relation to mother,

ily put down after held, but often upset limited in explo'

being held. when put down. ration.

Strange Situation: Strange Situation: Strange Situation:

Actively seeks mother Avoids mother when Difficult to soothe

when distressed, distressed, seems after separation

maintains contact on blase. angry and seeking

reunion, readily com' comfort simultanc'
forted. ously.

Preschool: Easily Preschool: Often Preschool: Fretful and

makes friends. angry, aggressive, defi' easily overwhelmed
Popular. Flexible and ant. May be isolated, by anxiety. Immature,
resilient under stress. disliked. Hangs overly dependent on
Spends more time around teachers. teacher. May be viC'
with peers. Good self' Withdraws when in timized by bullies.

esteem. pain.
 Appendix 445

Securely Avoidantly Ambivalently

Attached Attached Attached

Teachers treat in Teachers become con^ Teachers indulge,

warm, matter'of'fact, trolling and angry. excuse, and infan'

age-appropriate ways. tilize.

Age 6 with parents: Age 6 with parents: Age 6 with parents:

Warm and enthusiasm Abrupt, neutral, Mixes intimacy seek'

tic. Able to be open unenthusiastic ing with hostility.

and to engage in exchanges. Absence Affectedly cute or

meaningful of warm physical con^ ingratiating. May be

exchanges. tact. worried about mother

Comfortable with when apart.

physical contact.

Middle childhood: Middle childhood: No Middle childhood:

Forms close friend' close friends or friend' Trouble functioning

ships, and is able to ships marked by in peer groups.

sustain them in larger exclusivity, jealousy. Difficulty sustaining

peer groups. Often isolated from friendships when in

the group. larger group.

Secure Adult Dismissive Adult Preoccupied Adult

Easy access to wide Dismissing of impot' Still embroiled with

range of feelings and tance of love and anger and hurt at pat'

memories, positive connection. Often ents. Unable to see

and negative. idealizes parents, but own responsibility in

Balanced view of pat' actual memories don’t relationships. Dreads

ents. If insecure in corroborate. Shallow, abandonment.

childhood, has if any, selhreflection.

worked through hurt

and anger.

Usually has securely Usually has Usually has ambiva'

attached child. avoidantly attached lently attached child.

child.
 ACKNOWLEDGMENTS

I was fortunate in writing this book to have had the assistance of numer-
ous researchers, clinicians, and theorists, including many of the leading
attachment scholars. I am indebted to all of them and thankful for their

cooperation. Without it a hook like this would have been impossible.

Mary Ainsworth has spent countless hours discussing her work and her
life with me over the course of several years. Her open and generous assis'

tance has been an immeasurable help. The late John Bowlby welcomed
me into his office and home for two memorable afternoons, as he dis'

cussed the development of his thought and the intellectual struggles of

his career. Alan Sroufe has been available for many hours of interviews.
His ability to vividly conceptualize key issues in the attachment debates
and his feedback on the manuscript have been a great help to me. Ursula
Bowlhy, a wonderful and enthusiastic correspondent, has trusted me with
her memories of her htishand, providing me with insights into his charac'
ter and motivation 1 could not have gcuten elsewhere. Arietta Slade first

introduced me to attachment concepts in her exceptional class on

infancy in the Doctoral Program in Clinical Psychology at the City Uni-
versity of New York. She has been an invaluable supporter and adviser.
 Acknoudedgments 447

helping to clarify many of the key issues, and generously critiquing the

manuscript.
Roger Kohak, Jay Belsky, and Klaus Grossmann made themselves
available for repeated questioning. Each of them has helped me under'
stand subtle aspects of the theory and of their own work and has gener^

ously read and responded to the manuscript. Inge Bretherton, Mary

Main, Hopkins, Michael Basch, Alice Smuts, and Edward and Jean
Juliet

Mason have similarly undertaken to educate me about various important

realms that this book has touched upon. I am grateful both for their guid'
ance and their careful reading of the manuscript.
I also wish to thank Stella Chess for
the time she spent with me diS'

cussing both her own work and the early days of Child Psychiatry at

Bellevue Hospital in New York, and for reading and commenting on the
chapters relating to temperament, day care, and the critiques of attach'
ment theory. 1 am similarly indebted to Karin Grossmann, Jerome Kagan,

Jude Cassidy, Nancy Kaplan, Everett Waters, Alicia Lieberman,

Patricia

Crittenden, Eleanor Szanton, Robert Marvin, Charles Rycroft, Cindy

Hazan, Sandra Scarr, Leah Albersheim, Van Roy Pancake, Philip Shaver,
Morris Eagle, and Thomas Bouchard for the time they spent speaking

with me. Charles Rycroft, Katherine Coker, and Mary McKinney read
the book in manuscript form. They provided me with valuable
feedback,

as did Eleanor Szanton who read the chapters on day care and society.

Others who have shared their thoughts with me at various times in

ways that have enriched the book are Dymph van den Boom, Anthony
Storr, Colin Murray Parkes, Lawrence Aber, Robert
Emde, Dante
Elizabeth
Cicchetti, John Byng'Hall, Steven Ellman, Michael Lewis,
Bott Spillius, Eva SpitZ'Blum, Otto Kernberg, Fred Pine, Albert
Solnit,

Hanna Segal, Susana Isaacs Elmhirst, T. Berry Brazelton, Annelies Reiss,

Emily Fenichel, and Richard Billow.

1 am my editors at The Atlantic, Michael Curtis and
grateful, too, to

William Whitworth, who responded encouragingly to my proposal for an

present it in the
article on attachment theory and gave me the space to

length and manner 1 felt it needed. Without them and their very special
magazine this hook would not have happened. My editor at Warner
pleasure to
Books, Jamie Raab, has been supportive throughout and a
Deborah
work with. 1 would also like to thank my agent, Kris Dahl;
Huntington, Vickie Wingfield, and Wendy McCormick, who transcribed
some of my interviews; and Jennifer Patton, who assisted me in
getting

my notes and bibliography together in the harried final days.

448 Acknowledgments

My closing thanks go to my wife Thaleia who has read the manuscript

through all its stages and has helped me to understand how someone out'

side the field might respond to what I’ve written. She has also made my
life a lot easier and happier throughout.
 NOTES

INTRODUCTION
1 . Gesell, 1928, p. 328, cited in Laharba, 1981, p. 179.
2. Watson, 1928, pp. 81-82. u u
3. Bowlby, 1966, p. v. Forward in Ainsworth, 1967. Note that new studies have shown that the
of his mother and to
infant, even the neonate, does have the capacity to distinguish the smell
recognize her voice.

CHAPTER 1: Mother-Love

1 . Levy, 1937, p. 644.

2 . Cited in Spitz, 1945, p. 53.

3. Levy, 1937, p. 645.

4. Levy, 1937, p. 644.
5. Bender and Yarnell, 1941, p. 1169.
6 . Interview with Chess, October 21, 1991.
7. Bender and Yarnell, 1941, P- 1169.
8 . Bender and Yarnell, 1941, p- 1 169.
9. Bakwin, 1941, p. 31. i ,4 inm
Spitz, 1945; Bakwin 1942; Interview with
Robert Emde, March 14, IVVI.
10 .

11. Interview with Chess, October 21, 1991.

12 . Bakwin, 1942, p. 31.

13. Smuts ’s interview with Senn, September 1980.

14. McNemar, 1940, cited in Ainsworth, 1962.
that the fluctuations in IQ score do not
15. Skeels and Dye, 1939, cited in Ainsworth, 1962. Note
rather on how well one is able to function
necessarily reflect fluctuations in intelligence but
which might be affected by one’s emotional state.
intellectually,

16. Mary Ainsworth, personal communication, August 7, 1993.

450 Notes to Chapter 1

17. Spitz claimed to have been analyzed by Freud, but there was no independent record or corrobo-
ration of this, and according to his daughter, he enjoyed telling tales. Interview with Eva Spitz-
Blum, October 8, 1991.
18. Interview with Robert Emde, March 14, 1991.
19. LaBarba, 1981.

CHAPTER 2: Enter Bowlby

1. Bowlby, 1940, p. 155.
2. Interview with Chess, October 21, 1991.
3. Bowlby, 1988a.
4. Bowlby, 1940, p. 155.
5. East and Hubert, cited in Bowlby, 1940, p. 156.
6. Bowlby, 1940, p. 164.
7. Bowlby, 1940, p. 169.
8. Hamilton, 1990.
9. Interview with Colin Murray Parkes, May 8, 1992.
10. Sutherland, 1990.
1 1. Ursula Bowlby, personal communication, February 7, 1992.
12. Hamilton, 1990.
13. Ursula Bowlby, personal communication, July 6, 1992.
14. Ursula Bowlby, personal communication, February 19, 1992.
15. Kraemer, S., 1990.
16. Ursula Bowlby, personal communication, February 19, 1992.
17. Interview with Juliet Hopkins, May 7, 1992.
18. Scarf, 1976, p. 151.
19. Ursula Bowlby, personal communication, August 23, 1993.
20. Interview with Hopkins, May 7, 1992.
21. Hopkins, personal communication, August, 1992.
22. Alice Smuts’s Interview with Bowlby, June 6, 1977.
23. Bowlby, 1979, p. 13.
24. Smuts’s Interview with Bowlby, June 6, 1977.
25. Bretherton, 1992.
26. Smuts’s Interview with Bowlby, July 23, 1977.
27. Riviere, cited in Bowlby, 1985, p. 26.
28. Grosskurth, 1987, p. 396.
29. Although Bowlhy gave no indication that he respected Riviere, valued his experience with her,
or was influenced by her intellectually, he did send one of his sisters (Juliet Hopkins’ mother) to
Riviere for analysis. (Personal communication, Juliet Hopkins, 1993.) Was this in the first blush
of treatment when he still had hopes that it would be a valuable experience for him? Was this
indicative of a belief that psychotherapy or psychoanalysis was good for other people but not
nece.ssarily for him (an attitude that would seem consistent with other aspects of his personal-
ity)? Or did this suggest that Bowlby got more out of his experience with Riviere than he knew

or cared to let on?

30. Bowlby, 1985.
31. Ursula Bowlby, personal communication, July 4, 1992. According to Charles Rycroft: “Joan
Riviere impressed many people by her insistence that she was ‘county’ and ‘top drawer,’ but 1
don’t think she really was. Distinguished academics on her father’s side, but no baronets, barons,
medieval Welsh princes, Bowlby had.” Personal communication, March
etc., as 16, 1993.
32. Smuts, personal communication, August 14, 1992.
33. Bretherton, 1991.
34. Bowlby, 1988b, p. 44.
35. See Bowlby, 1979.
36. Rycroft, 1992, p. 74.
37. Freud, 1900, cited in Bowlby, 1979, p. 8.
38. Bowlby, 1988b, pp. 43-44.
 Notes to Chapter 5 451

CHAPTER 3: Bowlby and Klein

1 . Bowlhy, 1979, p. 5.
and wrote forcefully in the margin of
2 . Riviere, 1927, p. 374. Bowlhy did not much care for this

his copy, “Role of Environment = Zero.”

3. Klein, 1959, cited in Caper, 1988, p. 151.
4. Klein, 1927, cited in Caper, 1988, p. 154-
5. Klein, 1933, cited in Bloch, 1978, p. 2.
6 . Segal, 1981, p. 13.
7. Bowlhy, Figlio, and Young, 1990.
8 . Grosskurth, 1991.
9. Interview with Bowlhy, January 14-15, 1989.
10 . Klein, 1933, cited in Caper, 1988, p. 166.
11 . Riviere, 1927, pp- 376-377.
12 . Interview with Bowlhy, January 14—15, 1989.
13. Klein, 1975, p. 78.
14. Interview with Bowlhy, January 14-15, 1989.

CHAPTER 4: Psychopaths in the Making

1. Bowlhy, 1979, pp. 5—6.

2. See Vaillant, 1977, for a discussion of these themes. u •

with the conflict side of the equation.

•

He is
3. Bowlhy, 1979, p. 12. Bowlhy is working here strictly
out the equally critical issue of shame formation.
See Chapter 18.
leaving'
4. Bowlhy, 1979, p. 12.
5. Smuts’s Interview with Bowlhy, July 23, 1977.
6. Bowlhy, 1944, pp. 1 13-1 15.
7. Bowlhy, 1944, p. 40.

8. Bowlhy, 1944, p. 38.

9. Bowlhy, 1944, p. 41.

10. Bowlhy, 1944, p. 38.

11. Bowlhy, 1944, p. 38.

12. Bowlhy, 1944, p. 39.

13. Bowlhy, 1944, p. 49.

14. Bowlhy, 1944, p. 123.

15. Bowlhy, 1944, p. 35.

16. Bowlhy, 1944, p. 19.

17. Ainsworth, 1962.

18. Ainsworth, 1962; Rutter, 1981.

CHAPTER 5: Call to Arms

1. Bowlhy, 1951, p. 31.

Burlingham and Freud, 1944, cited in Bowlhy, 1979, p. 10.

2 .

in Bowlhy, 1979, pp. 89-90.

3. Burlingham and Freud, 1942, cited
4. Bowlhy, 1951.
Burlingham and Freud, 1944, cited in Bowlhy, 1951, p. 20.
5.

6 . Ainsworth, 1962.
7. A. Freud, cited in Bowlhy, 1951, p. 132.

8 . Bowlhy, 1951.
9. Bowlhy, 1951, p. 47.

10. Bowlhy, 1951, p. 76.

11 . Bowlhy, 1951, pp. 67-68.

Bowlhy, 1951, cited in Rutter, 1981, p. 15.
12 .

13. Bowlhy, 1951, p. 100.

14. Bowlhy, 1951, p. 157.

15. Bowlhy, 1951, p. 71.

had an unlucky publishing history.
16. The hook did not do well in the United States, where it

17. See Rutter, 1981.

 .

452 Notes to Chapter 6

CHAPTER 6: First Battlefield

1. Robertson, 1962, pp. 58-59.

2. Robertson and Robertson, 1989.
3. Edelston, 1953, cited in Robertson and Robertson, 1989.
4. Interview with Edward and Jean Mason, April 12, 1990.
5. Robertson, 1962, pp. 57-58.
6. Robertson, 1962, p. 110.
7. Robertson, 1962, pp. 61-62.
8. Lorenz, in Tanner and Inhelder, 1971, p. 232.
9. Robertson, 1962, pp. 67-68.
10.Robertson, 1962, p. 69.
11. Robertson, 1962, pp. 63-64.
12. Senn’s Interview with Bowlby, October 19, 1977.
13. Bowlby, Figlio, and Young, 1990.
14. Bowlby was proud of having written the first paper on family therapy, although he never had
the time to pursue the matter further and is thus not included in anyone’s pantheon of family
treatment founders.
15. Robertson and Robertson, 1989, pp. 10-11.
16. Mason, 1991. This film has unfortunately been withdrawn from distribution at the insistence of
Joyce Robertson.
17. Senn’s Interview with James Robertson, October 25, 1977.
18. Robertson and Robertson, 1989, p. 13.
19. Robertson and Robertson, 1989, p. 13.
20. Robertson and Robertson, 1989, p. 15.
21. Robertson and Robertson, 1989, pp. 16-17.
22. Robertson and Robertson, 1989, p. 20. This is Robertson’s recollection. Why was Spence so
cold to him? Perhaps he disliked someone else adopting his issue and taking it further than he
had. Or did Robertson, who could be quite touchy at times, misinterpret Spence?
23. Bretherton, 1987.
24. Senn’s Interview with James Robertson, October 25, 1977.
25. Robertson and Robertson, 1989, p. 32. Fourteen years later, when Robertson showed the film to
a gathering of Laura’s family, the sixteen-year-old Laura stood up at the end and said of the
experience she’d just watched, “That meant nothing to me.’’ But as she spoke, she leaned over
and clutched her father’s tie.
26. Robertson and Robertson, 1989, p. 41.
27. Tanner and Inhelder, 1971, pp. 230-231.
28. Robertson and Robertson, 1989, p. 42.
29. Senn’s Interview with James Robertson, October 25, 1977.
30. Mason, 1991.
31. Senn’s Inter'dew with James Robertson, October 25, 1977.
32. Robertson and Robertson, 1989, p. 46.
33. Robertson and Robertson, 1989, p. 45.
34. Senn’s Interview with James Robertson, October 25, 1977.
35. Grosskurth, 1987. Bowlby said of this episode: “We do not think this was due to pregnancy,
because we have got so many other anecdotal stories of exactly this kind.”
36. Robertson and Robertson, 1989, p. 45.
37. Smuts ’s Interview with Stone, August 1, 1977.
38. Robertson and Robertson, 1989, p. 54-
39. Tanner and Inhelder, 1971, p. 223. This is obviously not a direct quote but Bowlby’s memory of
what she said.
40. Senn’s Interview with James Robertson, October 25, 1977.
41. Robertson and Robertson, 1989, p. 67.
42. Robertson and Robertson, 1989, p. 111.
43. Robertson and Robertson, 1989.
44. Robertson and Robertson, 1973b.
45. from Mason, 1991
46. Robertson and Robertson, 1973b, p. 3.
47. Robertson and Robertson, 1973b, p. 10.
48. from Ma.son, 1991.
49. Robertson and Robertson, 1989, pp. 90-91.
 Notes to Chapter 7 453

50. Senn’s Interview with James Robertson, October 25, 1977.

51. Bowlby, 1973, p. xiii.
52. James from Robertson and Robertson,
1989.

53. Ainsworth, 1962.

CHAPTER 7: Of Goslings and Babies

1. Smuts’s Interview with Bowlby, July 23, 1977. r: ui i

written the foreword to the English

2. Smuts’s Interview with Bowlby, July 23, 1977. Huxley had
translation of King Solomon’s Ring, Lorenz’s new book, which
was due out the following winter,
and promised to get an advance copy for Bowlby.
3. See Bowlby, 1958b.
4. See, for example, Bowlby, Figlio, and Young, 1990.
5. Smuts’s Interview with Bowlby, July 23, 1977.
6. Interview with Bowlby, January 14—15, 1989.
7. Bowlby, 1958b.
Inspired by animal studies, in which— to cite one example—
mother goats rejected newbonas
8.
and John KenneU
who had been separated from them immediately after birth, Marshall Klaus
held that there is a critical period during which mother and
baby have the opportunity to bond,
of connection may never
and if they are separated by hospital policy, the necessary intensity
in the field to have been hugely
develop between them. The idea has been widely recognized
overstated and downright false in many respects. Mothers
have plenty of time to hoi^ to their
Nevertheless, the work of Klaus and
babies, even if they adopt them months after birth.
post-partum contact between
Kennell enabled a greater appreciation for how wonderful early
be to some mothers. Their
mother and newborn can be and how emotionally valuable it can
remain together rather than per-
work has convinced hospitals to allow mothers and babies to
1985.
senseless policy of isolating them after birth. See Ainsworth,
sisting in the
9. Freud, 1938, cited in Bowlhy, 1958, p. 352.
10. Sayers, 1991.
11. A. Freud, 1944, cited in Sayers, 1991, p. 169.

12. Erikson, 1950, pp. 247, 249.

13. Bowlhy, 1958b, p. 369. , , , , -r, . r-r i
worked at the lavistock L^linic
14 Bowlby noted that Ian Suttie, a British psychotherapist who had
before the war, had argued that the child’s need for his
mother was a primary motivation,
related to the desire for company and the dislike of isolation,
and was independent of the tK)L y i

that he had coupled his theory

needs that mothers satisfy. That Suttie was not an analyst and
ignored. He also died young.
with an attack on Freud probably insured that his theory would be
Suttie, 1935, Origins of Love and Hate, cited in Bowlhy,
1958b.

15. Klein, 1952, cited in Bowlby, 1958, p. 354.

16. Klein, 1952, cited in Bowlhy, 1958.
17. Fairbaim, 1952, p. 137.

18. Greenberg and Mitchell, 1983.

19. Rycroft, 1985.
20. Winnicott, 1975, p. 314.
21. Hughes, 1989.
22. Winnicott, cited in Hopkins, 1991, p. 187.

23. Grosskurth, 1987.

24. Phillips, 1988.
25. Rycroft, 1992, p. 80.
26. Hughes, 1989.
27. Rycroft, 1985. ^ u i j i -

Juliet don’t easily associate JB with the use of such an

Hopkins observes: “1
28.
‘maddening’ but that may well be correct. He once told me that
he found DWW
dinicult
scientist, could not
because he was completely content with paradox, whereas he himselt, as a
be.” Personal communication, August, 1992.
29. Winnicott, 1975, p. 239.
30. Bowlby, 1958b, pp. 366-367.
of the pleasures and health benefits breast feeding
may
31. Bowlby, 1958b, pp. 369, 370. Regardless
either of baby to mother or
provide, Bowlhy did not see it as essential to healthy attachment
hormone prolactin, secreted in substantial quantities
when a
vice versa. The fact that the
 .

454 Notes to Chapter 7

mother breast-feeds her baby, has been shown to promote nurturing behavior in laboratory ani-
mals is considered by some to be proof that breast feeding is the sine qua non of mother- infant
bonding. But the work of comparative psychologist Jay S. Rosenblatt suggests that prolactin is
not a crucial element in the bonding of mother to infant in mammals. While the hormone may
give some initial impetus to mothers who need it, the presence, appearance, and behavior of the
young seem to be more significant in sustaining the maternal bond. Personal communication,
1990.
32. Bowlby, 1958b, p. 370.
33. Yarrow, 1967.
34. Bowlby, 1958.
35. Burnham, 1965, cited in Bowlby, 1973, p. 251.
36. Bowlby, 1960.
37. Ainsworth, 1962.
38. Klein, 1935, cited in Bowlby, 1973.
39. Bowlby, 1973.

CHAPTER 8: “What’s the Use to Psychoanalyze a Goose?’’

1. Bowlby, 1958a, p. 3.
2. Bowlby, 1958a, p. 3.
3. Bretherton, 1987.
4. Bowlby, 1958a, p. 3.
5. Bowlby, 1958a, pp. 5-6.
6. Bowlby, 1958a, p. 7. Ursula Bowlby had lost her nanny at five and a half and felt that it had
scarred her for life. Ursula Bowlby, personal communication, August 23, 1993.
7. Bowlby, 1958a, p. 6.
8. Bowlby, 1958a, p. 9.
9. Bowlby, 1958a, p. 10.
10. Ursula Bowlby, personal communication, July 27, 1993.
11. Bowlby, 1958a, p. 6.
12. Bowlby, 1958a, p. 6.
13. Isaacs, quoted in Scarf, 1976, p. 151.
14. Isaacs, quoted in Scarf, 1976, p. 152.
15. Interview with Segal, January 13, 1989.
16. Interview with Rycroft, May 8, 1992.
17. Grosskurth, 1987, p. 325.
18. Interview with Hopkins, May 7, 1992.
19. Grosskurth, 1987.
20. Grosskurth, 1987.
21. Interview with Rycroft, May 8, 1992.
22. Grosskurth, 1987.
23. Charlotte Balkanyi, cited in Grosskurth, 1987, p. 406.
24. Interview with Rycroft, May 8, 1992.
25. Interview with Otto Kernberg, February, 1988.
26. Rita Frankiel, personal communication, October 1996.
27. Bowlby, 1980, p. 276.
28. See, for example, Furman (1981) and Lopez and Kliman (1979).
29. This opinion was expressed to me by both Hanna Segal (interview, January 13, 1989) and
Susana Isaacs (personal communication, June 19, 1992).
30. Bowlby, Figlio, and Young, 1990.
31. Ellenberger, 1970.
32. A. Freud, cited in Grosskurth, 1987, p. 404.
33. Interview with Bowlby, January 14-15, 1989.
34. A. Freud, 1960, p. 54.
'
35. Interview with Robert Emde, March 14, 1991.
36. Hamilton, 1990, p. 2 1

37. Grosskurth, 1987. Bowlhy characteristically refused to name the person he suspected of the
treachery.
38. Interview with Bowlby, January 14-15, 1989.
39. Bretherton, 1991.
 0

Notes to Chapter 1 455

40. Bretherton, 1991.

41. Bowlby, 1959.
42. Malan, 1990, p. 7.

43. Trowell in 1990.

44. Smuts’s Interview with Bowlby, July 23, 1977.
45. Rutter, 1981.
46. Bowlby, cited in LaBarba, 1981, p. 493.
47. LaBarba, 1981.
48. Casler, 1961, p. 30.
49. Casler, 1961, p. 30.
50. Casler, 1961.
r r>u
notes for Chapter
51. Interview with Ainsworth. (Regarding interview dates with Ainsworth, see
10 .)
52. Bowlby, cited in Ainsworth, 1962, p. 115.
53. Andry, 1960, cited in Ainsworth, 1962, pp. 103-104.
54. Ainsworth, 1962.
55. Stone, 1954, cited in LaBarba, 1981, p. 492.
56. Emde, 1983.
57. Eriksen, cited in Casler, 1961, p. 9.
While Casler’s review quoted here appeared in 1968, it was
58. Casler, cited in Rutter, 1981, p. 15.
typical of what the opposition had been already saying ten years earlier.

59. Rutter, 1981.

CHAPTER 9: Monkey Love

1. Such contact needs had been reported earlier by others.
2. Harlow, 1958.
3. Harlow, 1958, p. 673.
4. Harlow, 1958, p. 685.
5. Harlow, 1958, p. 677.
6. Smuts’s Interview with Bowlby, July 23, 1977.
7. Ainsworth, 1962.
8. Suomi and Harlow, 1978.
9. Bowlby, 1979.
10. Hinde and Spencer- Booth, 1970 and 1971,
cited in Rutter, 1981.

11. Harlow and Harlow, 1966, pp. 244-245.

12. Bowlby, 1988b, p. 23.

CHAPTER 10: Ainsworth in Uganda

following dates: June 6, 1988; June 16,
Note- Interviews with Mary Ainsworth took place on the
26, 1992; January 30, 1992;
1988- September 1, 1988; December 28, 1988; January 27, 1989; January
material relevant to chapters 10
February 2, 1992; August 11, 1993; August 20, 1993. Much of the
and 1 1 was covered in the interview of December 28, 1988.

1. Interview with Ainsworth.

2. Interview with Ainsworth.
3. Interview with Ainsworth.
4. Interview with Ainsworth.
5. Interview with Ainsworth.
6. Interview with Ainsworth.
7. Ainsworth, 1962.
8. Schaffer and Emerson, 1964, p. 5.

9. Ainsworth, 1967, p. 331.

10. Interview with Ainsworth.
11. Ainsworth, 1967, p. 270.
12. Ainsworth, 1967, p. 337.

13. Ainsworth, 1967, p. 341.

14. Ainsworth, 1967, p. 343.

 0

456 Notes to Chapter 1

15. Ainsworth, 1967, p. 345.

16. Ainsworth, 1967, p. 345.
17. Interview with Ainsworth.
18. Interview with Ainsworth.
19. Ainsworth, 1987, p. 331.
20. Ainsworth, 1987, p. 331-332.
21. Ainsworth, 1987, p. 392.
22. Ainsworth, 1967, p. 144.
23. Ainsworth, 1967, p. 141.
24. Mead, 1962.
25. Lieherman and Pawl, 1990, p. 382.
26. Ainsworth, 1967.

CHAPTER 1 1 : The Strange Situation

1. Interview with Ainsworth.

2. Interview with Ainsworth.
3. Interview with Ainsworth.
4. Interview with Ainsworth. This baby later turned out to fit into one of the less secure subgroups
in the securely attached category. Given the mother’s overall responsiveness, Ainsworth could
not have understood this outcome if she hadn’t visited the home.
5. Bowlby, 1988b, p. 49.
6. Interview with Ainsworth.
7. Interview with Ainsworth.
8. Interview with Ainsworth.
9. Interview with Ainsworth.
10. Interview with Ainsworth.
1 1. Interview with Ainsworth.
12. Ainsworth et al., 1978.
13. Both Suzie and Donnie, who follows, were not part of Ainsworth’s original study, hut were
observed — and videotaped —
some years later. Ainsworth did not videotape her original sam-
ple, but the phenomena she observed were essentially the same.
14. Interview with Ainsworth.
15. Ainsworth et al., 1978.
16. Interview with Ainsworth.
17. Main and Weston, 1982, p. 46.
18. Main and Weston, 1982, p. 45.
19. Bell and Ainsworth, 1972.
20. Ainsworth et al., 1978.
21. Interview with Main, December 18, 1988.
22. Interview with Ainsworth.
23. Interview with Ainsworth.
24- Ainsworth, 1983; interview with Ainsworth.
25. Interview with Bowlby, January 14-15, 1989.

CHAPTER 12: Second Front

1. Interview with Bretherton, January 22, 1989.

2. Interview with Waters, January 5-6, 1989.
3. Interview with Sroufe, December 20, 1988.
4. Waters, 1978.
5. For an overview of behaviorist thinking at this time see Ainsworth, 1969; Hartup, 1963; Mac-
coby and Masters, 1970; also Dollard and Miller, 1950; Gewertz, 1961.
6. Interview with Ainsworth.
7. Ainsworth, 1969, p. 1015.
8. Interview with Sroufe, December 20, 1988.
 4

Notes to Chapter 1 457

CHAPTER 13: The Minnesota Studies

1. Bowlhy, 1979, p. 109.

2. Bowlby, 1979, p. 102.

3. Clarke and Clarke (Eds.), 1976; Kagan and Klein, 1973.
4. Sroufe, 1989.
5. Sroufe, 1983.
6. Interview with Albersheim, April 23, 1993.
7. Interview with Albersheim, April 23, 1993.
8. Matas, Arend, and Sroufe, 1978. (Albersheim’s married name at the time was Matas.)

9. Mahler, Pine, and Bergman, 1975, p. 79.

10. Ainsworth, 1984, p. 568.

11. Mahler et al., 1975, p. 78.

12. Mahler et al., 1975, p. 95.
13. Mahler et al., 1975, 80.
14. Interview with Albersheim, April 23, 1993.
15. Waters, Wippman, and Sroufe, 1979.
16. Waters et al., 1979.
17. Sroufe, Carlson, and Shulman, 1993.
Sroufe found “virtually no differences” between avoidants and secures
“on Apgars, Brazelton
18.
behavioral observations at 3
neonatal exams at 7 and 10 days, nurses’ ratings in the hospital,
months, or Carey temperament questionnaire responses (derived from Thomas,
Chess, Birch, &
1968) at 6 months.” Sroufe, et al., 1993, p. 71.
19. Interview with Sroufe, December 20, 1988.
20. Sroufe, 1983.
21. Interview with Sroufe, December 20, 1988.
22. Cowen, 1973, cited in Eagle, 1992.
23. Interview with Sroufe, December 20, 1992.
24. Elicker, Englund, and Sroufe, 1992.
at times would proba-
25. Sroufe ’s third category of avoidant children who looked almost psychotic
bly have been considered part of a fourth category by other researchers,
more disturbed than the
avoidant or ambivalent children.
26. Interview with Sroufe, December 20, 1988.
27. Sroufe, 1983, pp. 74-75.
28. Sroufe, 1983.
29. Sroufe, 1983, p. 66.
30. Interview with Sroufe, December 20, 1992.
31. Sroufe, at City University of New York, February 10, 1989.
32. Sroufe, 1983, p. 76.
33. Rosenberg, 1984, cited in Sroufe et al., 1993.
34. Fury, 1992, cited in Sroufe et al., 1993.
35. Interview with Main, March 5, 1993.
36. Interview with Marvin, March 19, 1993.
37. Cicchetti and Beeghly, 1990.
38. Sroufe, 1983, p. 66.

CHAPTER 14: The Mother, the Eather, and the Outside World

1. Elicker et al., 1992, pp. 77-78.

2. Elicker et al., 1992.
3. Pastor, 1981.
Jacobson and Wille, 1986, cited in Turner, 1991. Note that this study, unlike the previous one,
4.
didn’t find such differences until age three.
5. Interview with Pancake, March 23, 1992.
6. Sroufe, personal communication, July 11, 1993.
7. Interview with Pancake, March 23, 1992.

8. Troy and Sroufe, 1987, p. 169.

9. Troy and Sroufe, 1987, p. 169.

10. Sroufe, 1989.

1 1. Elicker et al., 1992.
 4

458 Notes to Chapter 1

12. SroLife et al., 1993.

13. Sroufe et al., 1993, p. 25.

14. Erickson et al., 1985.

15. Renken et al., 1989.
16. Turner, 1991. Here are some from Turner’s study, expressed in the number of occur-
ot the data
rences of each type of behavior in the ninety-minute observation time. There were thirteen
securely attached girls, 1 1 securely attached hoys, 9 anxiously attached girls, and 7 anxiously
attached hoys in the study, all second-horns. Unique among such studies, attachment status was
based on current relationships, using a modified Strange Situation adapted especially for four-
and-a-half-year-olds.

Secure Girls Secure Boys Anxious Girls Anxious Boys

Dependent Acts 7.2 6.0 13.9 22.4
Assertive Acts 10.7 9.6 4.0 27.6
Controlling Acts 18.8 20.8 13.1 51.6
Acts of Aggression:
Mild 8.1 4.4 7.7 29.0
Strong 2.8 1.5 2.3 15.0
Playful 5.1 10.4 7.0 18.9

17. Interview with Ainsworth.

18. Sroufe et al., 1993.
19. Interview with Sroufe, December 20, 1988
20. Sroufe, personal communication, July 11, 1993. When the children in this study turned thir-
team got “every available man into the lah (N = 44)’’ and assessed his interaction
teen, Sroufe ’s
with the child and with the child and mother together.
21. See Bornstein and Lamb, 1992, for complete list of Lamb references, p. 419.
22. Klaus Grossmann, personal communication, August 10, 1993.
23. A subsequent effort to pool the data from all the mother-father attachment studies has found a
modest correlation between the child’s quality of attachment to mother and father. Sroufe, per-
sonal communication, July 11, 1993.
24. Main and Weston, 1982; Easterbrooks and Goldberg, 1991; Suess, Grossmann, Sroufe, 1992.
25 . Zero to Three, 1993.
26. See, for example, Smolak, 1986, on the differences between mothers’ and fathers’ speech pat-
terns with young children. Cited in Sroufe, Cooper, and Detlart, 1992.
27. Sroufe et al., 1992.
28. Belsky, 1984.
29. Sroufe et al., 1992.
30. Marvin and Stewart, 1991; interview with Marvin, March 19, 1993. Salvatore Minuchin, one
of the pioneers in family treatment and theory, has identified families as falling into one of three
types: adaptive, in which family members are sensitive to, open w'ith, and supportive of one
another; disengaged, where they tend to be uninvolved, insensitive, and somew'hat angry; or
enmeshed, in which they tend to ignore one another’s boundaries, behave in intrusive and dis-
respectful ways, and feel ambivalently toward one another. As Marvin points out, this bears a
remarkable similarity to secure, avoidant, and ambivalent attachment patterns.

CHAPTER 15: Structures of the Mind

1. Bowlhy, 1982.
2. Bowlhy, 1982.
3. Mahler et al., 1975.
4. Marvin, 1977; interview with Marvin, March 19, 1993.
5. See Bretherton, 1991.
6. Bowlhy, 1988; Wachtel, 1977.
7. Main, 1991.
8. Main, 1991.
9. Cain and Fast, 1972, cited in Bowlhy, 1988, p. 102.
10. Bowlhy, 1988b, pp. 101-102.
11. Main, 1991.
12. Rioch, 1988, p. 40.
 Notes to Chapter 1 7 459

13. “Piaget Sees Science Dooming Psychoanalysis,” by John L. Hess, New York Times, October 19,

1972, p. 49.

14. Rycroft, 1985, p. 149.

CHAPTER 16: The Black Box Reopened

1. Interview with Cassidy, February 9, 1993.
2. Interview with Main, March 5, 1993.
3. Interview with Nancy Kaplan, September 9, 1993.
4. Kaplan, 1987, p. 7.

5. Kaplan, 1987, p. 8.

Kaplan, 1987, p. 9. Children who had been classified secure five years earlier, however,
were not
6.

uniformly healthy or creative in their responses. Some disavowed their feelings and
lacked cre-

ative solutions, suggesting, perhaps, that their security had been compromised.
7. Kaplan, 1987, p. 1 1.

8. Kaplan, 1987, p. 13.

9. Kaplan, 1987, p. 17.

10. Main et al., 1985, p. 89.

Main et al., 1985, p. 90. Main cautions that these procedures are meaningless if attempted in
11.
the home.
12. Main, Kaplan, and Cassidy, 1985, p. 67.

13. Interview with Cassidy, February 9, 1993.

14. Main and Cassidy, 1988.

CHAPTER 1 7: They Are Leaning Out for Love

1. Campos came up with an average of 62 percent secure, 23 percent avoidant, and 15

et al, 1983,
of attach-
percent ambivalent based on a summary of all the American studies assessing quality
ment, not just middle-class samples.
2. Bowlby, 1973.
3. Cassidy and Berlin, 1992.
4. Main and Weston, 1981; Main, 1991.
5. See Sroufe, 1983. Avoidant aggression from the anger of the ambivalent children,
is different

who, as Sroufe points out, lack self-control, who are frustrated and flail about, haphazardly
seek-
more likely to be cruel and
ing connection. The avoidant child is very controlled, and his anger
bullying, a reflection of an underlying hostility that has been redirected
toward a safer target.
Longitudinal studies have yet to show that for any specific person the pattern formed
in child-
6.

hood still exists in adulthood. But adults do show the same

patterns (see Chapters 24 and 25).

7. See, for example, Egeland et al., 1988.

8. Sroufe, 1983, p. 77.
9. Sroufe, 1989.
10. Hopkins, 1991, p. 190.
1 1 . Interview with Crittenden, January 9, 1989.
12. Thirty percent, according to recent research. Cited in Fonagy, unpublished.
13. Fraiberg, Adelson, and Shapiro, 1975, p. 125.
14. Fraiberg et al., 1975, p. 1 10.

i975, p. 127. By being freed to express the negative emotions that had
been
15. Fraiberg et al.,
“momentous
walled off for so long, each mother seemed to go through what Fraiberg called a

shift in her identification with the baby. Whereas before she was
identified with the aggressors

of her childhood” —
and convinced that she was destined to behave as they did she was now —
ability to experience
the protector of her baby” (p. 127). It could also be said that the mother’s
enabled her natural
her long-walled-off attachment feelings, in the form of anger and grief, had
feelings with
attachment feelings for her baby to flow. What linked her childhood attachment
with herself as
her maternal attachment feelings seemed to be the mother’s ability to sympathize
analogous to the caring
a child. For the caring she finally felt for herself as a wounded child was
representing the
she now spontaneously gave to her baby. The therapy could thus be seen as
conquest of extreme avoidance by enabling attachment feelings to flow once more.
16. Lieberman, 1991, pp. 268-269.
17. Lieberman et al., 1991.

18. Interview with Sroufe, December 20, 1988.

 8

460 Notes to Chapter 1

CHAPTER 18: Ugly Needs, Ugly Me

1. See, for example, Sroufe, 1983.
2. Lewis, 1987.
3. Hopkins, 1991, p. 191.
4. For an overview of this subject, see Karen, 1992.
5. Cassidy, 1990. See also Ainsworth, 1985.
6. Interview with Basch, September 22, 1990.
7. Kohak, 1987.
8. Interview with Marvin, March 19, 1993.
9. Grossmann, 1989.
10. Slade and Aber, 1987, p. 21.
1 1. Lewis, 1987, p. 32.

CHAPTER 19: A New Generation of Critics

1. Interview with Kagan, January 9, 1989.

2. Kagan, 1984, p. xii.
3. Kagan, 1979, p. 9.
4. Kagan, 1984, pp. 62-63.
5. Interview with Kagan, January 1, 1989.
6. Kagan, Kearsley, and Zelazo, 1980.
7. Kagan, 1984, pp. 56-57.
8. Bowlby, 1980, pp. 7, 442.
9. Kagan, 1984, pp. 55-56.
10. Kagan, 1984, p. 63.
11. Interview with Kagan, January 1, 1989.
12. Interview with Sroufe, December 20, 1988.
13. Suomi and Harlow, 1972.
14. “It is, of course, not being argued that anxious attachment causes later overdependency,” Sroufe
wrote in 1984; “that is, were the caregiving environment substantially changed we would
expect greatly reduced continuity in adaptation. Quality of attachment in infancy does,
. .
.

however, predict later dependency behavior (as nothing else has yet been shown to do), and it
seems reasonable to assume that behavioral organization builds upon earlier foundations in a
coherent manner.” Sroufe et al., 1983, p. 1625.
15. Interview with Sroufe, December 20, 1988.
16. Erickson et al., 1985.
17. Kagan, et al., 1980, p. 147.
18. Kagan, 1980, cited in Sroufe, Egeland, and Kreutzer, 1990.
19. Sroufe et al., 1990.
20. Interview with Sroufe, December 20, 1992.
21. Sroufe, 1983, p. 74.
22. Sroufe et al., 1990.
23. One result is the lack of a truly valid method for assessing the attachment status of older chil-
dren, although several ingenious methods have been devised, most notably by Jude Cassidy and
Robert Marvin and by Pat Crittenden.
24. Sroufe et al., 1993.
25. Interview with Judith Crowell, MD, February 27, 1997. See also Crowell (1995).
26. Judith Crowell Interview, February 27, 1997.
27. Kagan, 1984, pp. 60-61. His comment overlooks the fact that many of the avoidant infants did
not just fail to approach their mother, but didn’t greet her at all, didn’t interact with her,
actively snubbed her.
28. Kagan, 1982.
29. Interview with Karin Crossmann, July 12, 1991.
30. Kagan, 1984, p. 62.
31. Interview with Klaus Grossmann, July 12, 1991.
32. Interview with Klaus Crossmann, July 12, 1991. This was in stark contrast to the mothers of
ambivalent children in Bielefeld, many of whom behaved in odd and disturbing ways — stuffing
the child at mealtimes, teasing him excessively, and engaging in other strangely in.sensitive
behaviors.
 Notes to Chapter 20 461

33. Interview with Karin Grossmann, July 12, 1991.

34. Interview with Karin Grossmann, July 12, 1991. might suppose that the repressed emo-
One
tions would then fester and become bigger, maybe emerging later in destructive ways. “That’s
what we would expect,” Klaus said. “But if you have a very highly regulated culture then all
these eventualities will be taken care of — by very rigid behavior.”

35. Leach, 1990, p. 190.

36. Interview with Karin Grossmann, July 12, 1991. Grossmann credits Ainsworth’s strict rules on
how to conduct a home observation for enabling them to notice this variable.
37. Grossmann and Grossmann, 1991.
38. Grossmann and Grossmann, 1991.
39. Klaus Grossmann, personal communication. May 24, 1993.
40. Reported to me by Klaus Grossmann, personal communication. May 24, 1993.

41. Interview with Klaus Grossmann, July 12, 1991.

42. Rutter, 1981, p. 160.
43. Interview with Sroufe, January 5, 1989.
44. Interview with Bretherton, January 22, 1989.
45. Lamb et al., 1984.
46. Interview with Bretherton, January 22, 1989.
47. Interview with Bretherton, January 22, 1989.
48. Grossmann et al., 1985.

49. Bates et al., 1985; Belsky and Isabella, 1988; Egeland and barber, 1984; Kiser et al., 1986;

Barglow and Hoffman, 1985.

CHAPTER 20: Born That Way?

1. Bruch, 1954.
2. Bruch, 1954.
3. McIntosh, cited in Bruch, 1954, p. 727.

4. Bruch, 1954, p. 726.

5. Chess et al., 1959, p. 801.
6. Chess, 1964, p. 613.
7. Chess, 1964, p. 614.
8. Levy, 1943.
9. Gabon, 1869, 1979.
10. Sargent and Stafford, 1965.
11. Interview with Chess, October 21, 1991.
12. Interview with Chess, October 21, 1991.
13. Interview with Chess, October 21, 1991.
14. Thomas, Chess, and Birch, 1969, pp. 20-24.
15. Chess now regrets that she reported that term. Personal communication, July 12,
1993.

16. Chess, Thomas, and Birch, 1965, pp. 37-38.

17. Chess et al., 1965, pp. 38-39.
18. Thomas et al., 1969, p. 79.
19. Thomas et al., 1969, p. 92.

20. Interview with Chess, October 21, 1991.

21. Thomas et al., 1969, p. 93.

22. Chess et al., 1965, pp. 35-36.

23. Chess et al., 1965, pp. 36-37.
24. Thomas et al., 1969, p. 87.

25. Chess et al., 1965, p. 40.

26. Chess et al., 1965, p. 43.

27. Chess et 1965, pp. 44-45.

al.,

28. Further analysis revealed that other combinations of inherited traits also put a child at risk for
later disorders. Other risky temperamental configurations include: 1 ) a combination
of irregu-

larity, nonadaptability, withdrawal responses, and predominantly negative mood of

high inten-
sity; 2) a combination of withdrawal and negative responses of low intensity to new situations,

followed by slow adaptability; 3) excessive persistence; 4) excessive distractibility; and 5)

markedly high or low activity level. Thomas et al., 1969.
29. Thomas et al., 1969, p. 73.

30. Chess and Thomas, 1987, p. 188.

462 Notes to Chapter 20

31. Chess and Thomas, 1987, p. 189.

32. Thomas et al., 1969.

33. Brazelton, 1983.
34. Levy, 1943.
35. Greenberg and Mitchell, 1983, p. 229.
36. Greenberg and Mitchell, 1983, p. 228.
37. Schoenewoll, 1990, p. 19.

CHAPTER 21 : Renaissance of Biological Determinism

1. Bowlby, 1940, p. 157.

2. Bowlby, 1944, pp. 39, 41.
3. Bowlby, 1973, p. 187.
4. Ainsworth et al., 1978, p. 142.
5. Neuhauer and Neubauer, 1990, p. 21.
6. Bell, 1968.
7. In the United States, among Thomas, and Birch, were the
the key studies to follow Chess,
Louisville Twin Study, a longitudinal study of some five hundred twins from birth through fif-
teen years (begun in 1963, it started assessing temperament in 1976); the Colorado Adoption
Project, another large-scale longitudinal, which compared the temperamental characteristics of
adopted children to their biologically unrelated siblings raised in the same family; and the
Minnesota Study of Twins Reared Apart.
8. Todd, 1985; Rosen, 1987.
9. Segal, 1984-
10. Interview with Bouchard, November 11, 1990.
11. Jackson, 1980; Rosen, 1987.
12. Interview with Bouchard, May 5, 1993.
13. Bouchard and McGue, 1990; Bouchard et al., 1990; Tellegen et al., 1988; Waller et al., 1990;
Arvey et al., 1989.
14- Rosen, 1987, p. 38.
15. Neuhauer and Neubauer, 1990.
16. Dunn and Plomin, 1990, p. 78.
17. Goleman, 1991.
18. Most psychologists agree that temperament is an inborn predisposition that is stable across time
and that show’s up in almost any situation. This includes genetic as well as constitutional com-
ponents, which may he influenced by such variables as the mother’s hormonal levels during
pregnancy or her having a disease like rubella.
19. Plomin, 1987.
20. Kagan estimates, based on the many different recipes that can comprise this broth, that we may
ultimately discover sixty or seventy temperamental types.
21. Kagan, 1989. (No page numbers are available for this source as 1 was using a tape transcript and
not Kagan’s paper.)
22. Quoted in Galvin, 1992, p. 41.
23. Interview with Kagan, December 21, 1990.
24- Kagan, 1989.
25. Kagan, 1989.
26. Interview with Bouchard, November 1 1, 1991.
27. Interview with Kagan, December 21, 1990.
28. Interview with Kagan, December 21, 1990.
29. Interview with Sroufe, December 20, 1988.
30. Lally and Mangione, 1 979.
31. Interview with Lieberman, March 1, 1993.
32. Bowlby, 1982, p. 368.
33. Sroufe et al., 1992, p. 217.
34. Mangelsdorf et al., 1990.
35. Interview with Suomi, November 1 1, 1991.
36. Chess and Thomas, 1982, p. 220.
37. Interview with Sroufe, January 5, 1989.
38. Sroufe, 1983, p. 42.
39. Sroufe, personal communication, July 1993.
 Notes to Chapter 22 463

40. Vaughn et al., 1987; Zeanah et al., 1987.

41. Sroufe and Waters, 1977H.

42. Interview with Kagan, January 9, 1989.
43. Belsky and Isabella, 1988, p. 52, emphasis in original. The Japanese study was reported by
Miyake, Shing-jen, and Campos, 1985.
44. Belsky and Isabella, 1988. At nine months the mothers of babies who would later be classified
as insecure rate them as significantly more fussy.

45. Interview with Karin Grossmann, July 12, 1991.

46. Vaughn et al., 1989.
47. Van den Boom, personal communication, March 1, 1993.
48. Moore and Freeth, 1993.
49. Letter to The APA Monitor, April 1993, p. 3, by Joseph Richman. See also Kohn, 1993. The
trend is such that former U.S. Surgeon General C. Everett Koop warned against the power of
the “pharmaceuticahmedical complex” in the mental health fields.

50. In a study of adults whose fathers had been heavy drinkers, it was found that 85 percent con-
sumed very little alcohol or none at all. Cited in Kohn, 1993.
51. Scarr and Weinberg, 1993, p. 94.

52. Scarr, 1991, p. 16.

53. Scarr, 1991, p. 16.
54- Jensen, 1973.
55. Mischel, 1981, cited in Tellegen et al., 1988, p. 1037.
56. Waller et al., 1990.
57. Interview with Stella Chess, October 21, 1991.
58. Bronfenbrenner, quoted in Adler, 1991, p. 8.
59. Gottesman and Shields, 1972, found the concordance rates to he anywhere from 20 percent to
60 percent depending on the study.
60. See, for example, E. Fuller Torrey’s otherwise excellent Surviving Schizophrenia: A Family Man-
ual, 1988.

61. Tienari et 1990, cited in Sroufe et al., 1992. Considering this and other studies, psychiatrist
al.,

Lyman Wynne, a schizophrenia researcher at the University of Rochester, now believes that
although heredity plays some role, “the evidence for genetic effects” in schizophrenia is not
overpoweringly strong.” Kohn, 1993, p. 30.

62. Neubauer and Neuhauer, p. 22.

63. See Schore, 1994.

CHAPTER 22: A Rage in the Nursery

1. Clarke-Stewart, 1989. Today almost a third of mothers with children under three are working
full-time.
2. Scarr, 1984, p- 98.
3. Interview with Bowlhy, January 14-15, 1989.
4. Scarr, 1984, p. 100.
5. Rutter, 1981.
6. Scarr, Phillips and McCartney, 1993, pp. 373-374.
7. Kagan et al., 1980.
8. Rutter, 1981.
9. Sroufe, 1988.
10. ZERO TO THREE, 1993.
1 1. Interview with Scarr, August 19, 1993.
12. An example of the complexities involved by a study which found that working
is illustrated

mothers of securely attached toddlers showed better emotional integration than working moth-
ers of anxious toddlers. Benn, 1985.
13. Vaughn, Gove, and Egeland, 1980.
14. Schwartz, 1983, and Vaughn et al., 1980, cited by Vaughn, Deane, and Waters, 1985.
1 5. Vivian Cadden, Letters to the Editor, The Atlantic, May 1990.
16. Belsky, quoted in Wattenherg, 1992, p. 8.
17. Interview with Bowlhy, January 14-15, 1989.
18. Interview with Bowlhy, January 14-15, 1989.
19. Interview with Cassidy, February 10, 1993.
20. Letter from Sroufe to Belsky, March 2, 1979.
 .

464 Notes to Chapter 22

21. Schwartz, Strickland, and Krolick, 1983; Willie and Jackson, 1984; Barglow, 1985. All cited in
Belsky, 1986.
22. Interview with Belsky, December 4, 1992.
23. Interview with Belsky, December 4, 1992.
24. Interview with Belsky, December 4, 1992.
25. Belsky, 1986, p. 6.
26. Bronfenbrenner, quoted in Byrne, 1977, p. 43.
27. Interview with Szanton, May 5, 1993.
28. Interview with Fenichel, May 6, 1993.
29. Interview with Belsky, December 4, 1992.
30. Phillips etal., 1987.
31. Chess, 1987, p. 25.
32. Scarr, quoted in Ricks, 1987, p. 35.
33. Collins, 1987.
34- Scarr, quoted in Evans, 1987.
35. Belsky, 1987, p. 23.
36. Kagan, Phillips, both quoted in Jackson, 1987, p. 3.
37. Interview with Cassidy, February 10, 1993.
38. Interview with Belsky, December 4, 1992.
39. Interview with Belsky, December 4, 1992.
40. Interview with Scarr, August 19, 1993.
41. Interview with Belsky, December 4, 1992.
42. Clarke-Stewart, 1989.
43. Interview with Belsky, December 4, 1992.
44. Belsky and Cassidy, in press. Note that the 40 percent figure for children in day care includes
children from all sorts of homes, not just low-risk, middle-class homes.

45. Interview with Belsky, December 4, 1992.

46. Interview with Brian Vaughn, 1994.
47. Interview with Peter Barglow, 1994.
48. Belsky, personal communication, August 31, 1993.
49. Interview with Belsky, December 4, 1992.
50. Interview with Slade, March 3, 1993.
51. Interview with Szanton, May 5, 1993.
52. Interview with Alicia Lieberman, 1994-
53 . Zero to Three, 1993.
54. Zero to Three, April/May 1994, p. 33; Chira (1994).
55. Scarr et al., 1993, p. 369.
56. Interview with Deborah Vandellm May, 1994. See also Vandell, (1990).
57. Interview with Bates, May 1994. See also Bates (1994).
58. Interview with Belsky, April 25, 1994.
59. Interview with Belsky, December 4, 1992.
60. Belsky, quoted in Wattenberg, 1992, p. 8.
61. Interview with Belsky, April 29, 1994.
62. Interview with Belsky, April 29, 1994.
63. Interview with Bowlby, January 14-15, 1989.
64. Martin, 1991 ,
p. 65.
65. Figures from Felice Schwartz, of Catalyst, interviewed on National Public Radio, Weekend
Edition, February 13, 1993.
66. Interview with Bowlby, January 14-15, 1989.
67. Interview with Scarr, August 19, 1988.
68. Interview with Bowlby, January 14-15, 1989.
69. Interview with Ainsworth.
70. Sroufe, 1992.
71. Bronfenbrenner, quoted in Byrne, 1977, p. 43. Bronfenbrenner might have added peer experi-
ence as well.
72. Interview with Lieberman, April 25, 1994.
73. Barglow (1995).
74. Interview with Barglcw, 1994.
75. Interview with Barglow, 1994.
76. Haines-Stiles, 1991
 Notes to Chapter 24 465

CHAPTER 23; Astonishing Attunements

1. Cited in Sroufe, 1979, p. 462.

2. MacFarlane, 1975, cited in Stern, 1985.
3. Siqueland and DeLucia, 1969, cited in Stem, 1985; Goodman, 1991/1992.
4. Kohut’s actual term was “self-object.”
5. Stressing the importance and healthiness of this early grandiosity, Kohut argued that the narcis-
sistic personality disorder develops in those who have never had their grandiosity adequately

supported as infants and thus are unahle to outgrow the need for it.
6. The eminent Yale developmentalist Jerome Bruner refers to mother’s ability to help the child
unfold by being just a tiny step ahead of him as “scaffolding”; the great Russian psychologist Lev
Vygotsky referred to the phenomenon as the “zone of proximal development.” Sroufe et al.,
1992.
7. Stern, 1985.
8. Tronick, quoted in Goodman, 1991/1992, p. 27.

9. Sander, 1964.
10. Brazelton et ah, 1974.
11. Stern, 1985, p. 140.
12. Stem, 1985.
13. Abramis, 1990.
14. Stem, 1985.
15. Stem, 1985, p. 148.

16. Tronick et al., 1978.

17. Stem, 1985, pp. 151-152.
18. Isabella and Belsky, 1991.
19. Winnicott, 1958, p. 34.

20. Stem, 1985.

21. Stem, 1985, p. 195.

22. Stem, 1985, pp. 196-197.

23. Stem, 1985, p. 197.

24. Stem, 1985, p. 198.

25. Stem, 1985, p. 227.

26. Stem, 1985.
27. Bowlby, 1988b, p. 132.

28. Stem, 1985, p. 214.

29. Stern, 1985.
30. Haft and Slade, 1989.
31. Grossman, 1989, p. 2.

32. Winnicott, 1958, p. 300.

33. Winnicott, 1958.
34. Winnicott, 1965.

CHAPTER 24; The Residue of Our Parents

1. Bowlby, 1979, pp. 19-20.

2. Fraiberg et al., 1975, p. 387.
3. Sroufe, 1989.
4. Morris, 1983; Ricks, 1985.
5. Main et al., 1985, p. 96.
6. Fonagy et al., unpublished.
7. Shaver and Hazan, 1992, p. 43.
8. Fonagy et al., unpublished, p. 7.
9. Main et al., 1985, p. 97.

10. Fonagy et al., unpublished.

11. Fonagy et al., unpublished, p. 8-9.
12. Main and Goldwyn, in press.
13. Fonagy, Steele, and Steele, 1991.
14. Slade, September 1992, p. 4.
15. Slade, September 1992.
16. Interview with Slade, February 18, 1993.
466 Notes to Chapter 24

17. Interview with Judith Crowell, February, 1997.

18. Interview with Main, March 5, 1993.
19. Ainsworth and Eichberg, 1991.
20. Interview with Cassidy, February 9, 1993.
21. Kohak, personal communication, August 24, 1993.
22. Interview with Kohak, July 8, 1993.
23. Interviews with Slade, January 6, 1989; February 18, 1993.
24. Bowlby, 1979, p. 17.
25. Fonagy et al., 1991b, p. 209.
26. Fonagy et al., 1991b, p. 210.
27. Fonagy et al., 1991 h, p. 1 1 1.
28. Because Ainsworth and Main were analyzing mothers’ behavior in the home and Belsky in the
lab, one way of making sense of this conflicting evidence is to assume that mothers who are nor-

mally rejecting in the home may behave in an intrusive, out-of-synch way during a short struc-
tured play period in a university setting, in order to be the interactive mother they think a
psychologist expects. Karin Grossmann (1989) also found mothers of avoidant children to be
intrusive.
29. Interview with Cassidy, February 9, 1993.
30. Ainsworth et al., 1978.
31. Slade, 1993, p. 8.
32. Interview with Sroufe, December 20, 1988.
33. Interview with Slade, March 11, 1993.

CHAPTER 25: Attachment in Adulthood

1. Bowlby, 1979, p. 103.

2. Korchin and Ruff, 1964; Ruff and Korchin, 1967; cited in Bowlby, 1979, p. 108.
3. Bowlby, 1979.
4. Interview with Anthony Storr, author of Solitude, July 14, 1993. (He spoke with Bowlby on the
subject.)
5. Interview with Kohak, June 9, 1993.
6. Bretherton, 1992.
7. Interview with Parkes, May 8, 1992.
8. Weiss, 1975, 1982.
9. Hazan and Shaver, 1987, p. 515.
10. Simpson, Rholes, Nelligan, in press, cited in Shaver and Hazan, 1992. No clear pattern
emerged for ambivalents in this study.
1 1. Interview with Hazan, July 20, 1993.
12. Interview with Kohak, June 9, 1993.
13. Shaver, personal communication, September 1, 1993.
14. Kohak, personal communication, August 24, 1993.
15. Grossmann, personal communication, September 3, 1993.
16. Interview with Slade, February 18, 1993.
17. See Bowlby, 1973; 1980.
18. Lewis et al., 1984.
19. See Basch, 1988.
20. This is a revisionist reading of Heinz Kohut’s work on narcissism. He did not see grandiosity as a
defense against shame, hut rather the other way around. See Morrison, 1989. Note that it is
commonly accepted that healthy infant development includes a grandiose sense of self. It wanes
as the young child becomes better able to handle daily challenges on his own, develops a secure
sense of himself, and is simultaneously confronted with the fact the world does not exist to
serve and applaud him.
21. Fonagy et al., 1992, p. 18.
22. It could equally he argued that borderline pathology begins with disorganized attachment. See
Slade, 1993, for a valuable discussion of these issues and in particular the way early modes of
coping with negative emotion typical of the avoidant and ambivalent child may contribute to
neurotic character. See also Liotti, 1992, for a discussion of how disorganized/disoriented
attachment may lead to an increased vulnerability to dissociative disorders, like multiple per-
sonality.
 Notes to Chapter 27 467

CHAPTER 26: Repetition and Change

1. Rioch, 1988.
2. Adapted from Karen, 1987, p. 27.

3. Slade, 1993, writes: “It is my strong impression —

having read many [Adult Attachment
Interview transcripts]
.”
—
that insecure adults generally have more difficulties with intimate rela-
tionships. . . (pp. 5-6).
4. Main and Goldwyn, in press, p. 23.
5. Interview with Lieberman, March 1, 1993.
6. See Belsky et al., 1989; also Egeland, Jacobvitz, and Sroufe (1988), and Hunter and Kilstrom
(1979), both cited in Belsky et al., 1989. Belsky studied an average population. The other two
studies were of women who had a history of abuse as children.
7. Interview with Lieberman, March 1, 1993.
8. Slade and Aber, 1987.

CHAPTER 27: Avoidant Society

1. Laslett, 1962, p. 93.

2. Sennett, 1977.
3. See Fromm, 1955, for a discussion of some of these issues.

4. Lasch, 1979.
5. Sommerville, 1990.
6. Maccoby and Martin, 1983.
7. Sommerville, 1990.
8. Erikson, 1950.
9. Kohut, 1980.
10. Bowlby, 1988b, p. 2.

11. Sagi et al., A

more recent report by Sagi and co-workers suggests that elevated levels of
1985.
ambivalent attachment only occur in those kibbutzim where children sleep separately from
their parents. With only a single caretaker listening for disturbance over a monitor at another
location, the babies’ cries often go unheeded or are only responded to after a long delay. Sagi,

1991, cited in Belsky and Cassidy (in print).

12. Interview with Bowlby, January 14-15, 1989.
13. Interview with Ainsworth.
14. Greenspan, 1983.
15. Interview with Sroufe, December 20, 1988.
16. Mitchell, 1991.
17. Interview with Ainsworth.
18. Liedloff, 1977, p. 9.
19. Liedloff, 1977, p. 139.
20. Liedloff, 1977, p. 78.
21. Liedloff, 1977, p. 157.
22. Anisfeld et al., 1990.
23. Konner, 1989.
24. Interview with Karin Grossmann, July 12, 1991.
25. Interview with Karin and Klaus Grossmann, July 12, 1991. The reader may also recall the case

of Muhamidi, one of the insecurely attached Ganda children whose mother carried him every-
where, fed him on demand, but was herself emotionally disturbed.
26. Crockenberg, 1981; Jacobson and Frye, 1991.
27. Sosa et al., 1980; Klaus et al., 1986, both cited in Bowlby, 1988b.
28. Zero TO Three, 1993.
29. Interview with Sandra Scarr, August 19, 1993.
30. Rodell, 1993.
31. Bronfenbrenner, 1988.
32. Zero to Three, 1993, p. 19.

33. Interview with Colin Murray Parkes, May 8, 1992.

468 Notes to Chapter 28

CHAPTER 28: Looking Back

1. Sulloway, 1991, p. 30.

2. Bowlby, 1991.
3. Bowlby, 1973, p. 403.
4. Liotti, 1990.
5. According to Ursula Bowlby,John was “fearless rather than brave — he did not have to struggle
with fear.” Personal communication, March 10, 1992.
6. Bowlby, 1985.
7. Ursula Bowlby, personal communication, February 19, 1992.
8. Holden, quoted by Trowell, 1990, p. 6.
9. Martin, 1991, p. 66.
10. Byng'Hall, 1991, p. 267.
1 1. Isca Wittenberg, quoted by Trowell, 1990, p. 12.
12. Ursula Bowlby, personal communication, February 19, 1992.
13. Gilmore, 1990.
14. Interview with Bowlby, January 14, 1989.
15. Bowlby, quoted by Dinnage, 1980, p. 60.
16. Rycroft, 1985, pp. 151-152.
17. At meeting of the American Psychological Association in August 1992, Yale psychologist Sid-
a
ney Blatt reported that the first question Bowlby asked him when they met was, “Do you consid-
er me a psychoanalyst?” “Ethologist” comes from Gilmore, 1990, in The Psychoanalytic Quarterly.
18. Susanna Isaacs Elmhirst, personal communication, June 19, 1992.
19. See, for instance, Silverman, 1991; Osofsky, 1988; Goldberg, 1991.
20. Interviews with Kemherg, February 11, 1989; Pine, February 1, 1989; and Solnit, August 24,
1993.
21. Interview with Eagle, September 21, 1992.
22. See Slade (in preparation, 1997a and 1997b) for a valuable assessment and overview of the
many junctures of attachment research and psychoanalytic theory, including the contributions
of Fonagy and others.
23. Aaron, 1991, p. 241.
24. Storr, cited in Martin, p. 65.
25. Interview with Storr, July 14, 1993.
26. Rycroft, 1992, p. 80.
27. Rayner, 1991, p. 8. Rayner’s paper owed “much of its contents and inspiration” to Juliet Hopkins.
28. Ursula Bowlby, private communications, July 4, 1992; February 7, 1992.
29. Juliet Hopkins, personal communication, July 1993.
30. Ursula Bowlby, personal communication, February 19, 1992.
31. Grossman, 1989, p. 2.
32. Robert Marvin, personal communication, 1993.
33. Bretherton, 1992, p. 762.
34. Ursula Bowlby, personal communication, September 1 1, 1993.
35. Interview with Waters, January 5-6, 1989.
36. Interview with Ainsworth.
37. Interview with Ainsworth.
38. Interview with Sroufe, December 1988.
39. Interview with Ainsworth.
40. Interview with Ainsworth.
41. Brazelton, personal communication, February 23, 1989.
42. Interview with Sroufe, December 1988.
43. Interview with Eagle, September 21, 1992.
44. Rycroft, 1985.
45. Basch, personal communication, June 1993. See also Basch, 1992.
46. Interview with Lieberman, March 1, 1993. See also Lieberman, 1993.
47. Rycroft also makes this point. Interview, May 8, 1992.
48. Byng-Hall, 1991, p. 267.
49. Interview with Bowlby, January 14, 1989.
50. Bowlby, 1988b, pp. 38.
51. Interview with Ainsworth.
52. Macdonald, 1965, p. 143.
 BIBLIOGRAPHY

Abramis, M. (1990). Regulation of mother-infant affective engagement. Unpub-

lished doctoral dissertation, Yeshiva University, New York.

Adler, T. (1991, January). Seeing double? American Psychological Association Monitor,

pp. 1, 8.

Ainsworth, M. D. S. (1962). The effects of maternal deprivation: A review of find-

ings and controversy in the context of research strategy. In Deprivation of Maternal

Care: A Reassessment of its Effects. Public Health papers no. 14. Geneva: World
Health Organization.
Ainsworth, M. D. S. (1967). Infancy in Uganda: Infant care and the growth of love.

Baltimore:The Johns Hopkins University Press.

Ainsworth, M. D. S. (1969). Object relations, dependency and attachment: A theo-

retical review of the infant-mother relationship. Child Development, 40,

969-1025.
Ainsworth, M. D. S. (1983). A sketch of a career. In A. N. O’Connoll &. N. F. Russo
(Eds.), Models of achievement: Reflections of eminent women in psychology (pp.

200-219). New York: Columbia University Press.

Ainsworth, M. D. S. (1984). Attachment. In N. S. Endler &. J. McV. Hunt (Eds.),

Personality and the behavioral disorders (Vol. 1, pp. 559-602). New York: Wiley.

Ainsworth, M. D. S. (1985). Attachments across the life span. Bulletin of the New
York Academy of Medicine, 61, 792-812.
470 Bibliography

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of

Attachment: A psychological study of the strange situation. Hillsdale, NJ: Erlbaum.

Ainsworth, M. D. S. & Eichberg, C. (1991). Effects on infant-mother attachment of

mother’s unresolved loss of an attachment figure, or other traumatic experience.

In C. M. Parkes, J.
Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life

cycle. New York: Tavistock/Routledge.

Anisfeld, E., Casper, V., Nozyce, M., & Cunningham, N. (1990). Does infant carry-

ing promote attachment? An experimental study of the effects of increased physi-

cal contact on the development of attachment. Child Development, 61,

1617-1627.
Arend, R., Gove, E, & Sroufe, L. A. (1979). Continuity of individual adaptation

from infancy to kindergarten: A predictive study of ego-resiliency and curiosity in

preschoolers. Child Development, 50, 950-959.
Aron, L. (1991). John Bowlby, In memoriam. Psychoanalytic Dialogues, 1, p. 241.
Arsenian, J. M. (1943). Young children in an insecure situation. Journal of Abnormal
and Social Psychology, 38, 225-229.
Arvey, R. D., Bouchard, T. ]., Segal, N. L, & Abraham, L. M. (1989). Job satisfac-

tion: Environmental and genetic components, journal of Applied Psychology, 74

(2), 187-192.
Babies and day care. (1987, June 7). The Boston Globe, pp. A9, A12.
Bakwin, H. (1942). Loneliness in infants. American Journal of Diseases of Children,

63, 30-40.
Barglow, P, & Hoffman, M. (1985, May). Mother’s effect upon oral phase ego devel-
opment and attachment. Paper presented at the meeting ot the American
Psychoanalytic Society, Denver.
Barglow, P, Vaughn, B. E., & Molitor, N. (1987). Effects of maternal absence due to
employment on the quality of infant-mother attachment in a low-risk sample.

Child Development, 58: 945-954.

Barglow, P, Kavesh, L., Contreras, J. (1994). Offspring of working mothers: Peer
competence follow-up. Paper presented at the Thirteenth International Congress

of the International Association of Child and Adolescent Psychiatry and Allied

Professions, July, 1994.

Basch, M. F. (1988). Understanding psychotherapy : The science behind the art. New
York: Basic.
Basch, M. F. (1992). Practicing psychotherapy : A casebook. New York: Basic.

Bates, J., Masling, C., & Frankel, K. (1985). Attachment security, mother-child
interaction, and temperament as predictors ot behavior problem ratings at age
three years. In I. Bretherton and E. Waters (Eds.), Growing points in attachment

theory and research. Monographs of the Society for Research in Child Development, 50

(Serial No. 209).

Bates, J. E. & Marvinney, D. (1994). Child care history and kindergarten adjust-
ment. Developmental Psychology.
Bell, R. (1968). A reinterpretation of the direction of effects in studies of socializa-

tion. Psychological Review, 75, 81-95.

Bell, R., & Ainsworth, M. D. S. (1972). Infant crying and maternal responsiveness.
Child Development, 43, 1171-1 190.
 , 1

Bibliogra{)hy 47

Belsky, J. (1984). The determinants of parenting: A process model. Child

Development, 55, 83-96.

Belsky, J. (1986, September). Infant day care: A cause for concern. ZERO TO
THREE, pp. 1-7.
Belsky, J. (1987, February). Risks remain. ZERO TO THREE, pp. 22-24.

Belsky, & Cassidy, (in press). Attachment: Theory and evidence. In M. Rutter,
J. J.

D. Hay, & S. Baron-Cohen (Eds.), Developmental principles and clinical issues in

psychology and psychiatry. Oxford: Blackwell.
Belsky, J., & Isabella, R. A. (1988). Maternal, infant, and social-contextual determi-
nants of attachment security. In J. Belsky & T. Nezworski (Eds.), Clinical implica'

tions of attachment. Hillsdale, NJ: Erlbaum.

Belsky, J., & Rovine, M. (in press). Temperament and attachment in the strange sit-

uation: An empirical rapprochement. Child Development.

Belsky, J., & Steinberg, L. D. (1978). The effects of day care: A critical review. Child
Development, 49, 929-949.
Belsky, J., Youngblood, L, & Pensky, E. (1990). Childrearing history, marital quality,
and maternal affect: Intergenerational transmission in a low-risk sample. Develop-
ment and Psychopathology 1, 291-304.
Bender, L, & Yamell, H. (1941). An observation nursery: A study of 250 children

on the psychiatric division of Bellevue Hospital. American Journal of Psychiatry,

97, 1158-1174.
Benn, R. K. (1985, April). Factors associated luith security of attachment in dual career

families. Paper presented at the biennial meetings of the Society for Research in
Child Development, Toronto.
Bloch, D. (1978). So the witch wont eat me: Fantasy and the child’s fear of infanticide.

New York: Grove.

Bootzin, R., Loftus, E., Zajonc, R., & Hall, E. (1983). Psychology today: An introduc-

tion (5th ed.). New York: Random House.

Bomstein, M., & Lamb, M. (1992). Development in infancy: An introduction. Third
Edition. New York: McGraw-Hill, Inc.
Bouchard, T. J., & McGue, M. (1990). Genetic and rearing environmental influ-
ences on adult personality: An analysis of adopted twins reared apart. Journal of

Personality, 58 (1), 263-292.

Bouchard, T. J., Jr., Lykken, D. T, McGue, M., Segal, N. L., &. Tellegen, A. (1990,
October 12). Sources of human psychological differences: The Minnesota study

of twins reared apart. Science, 250, 223-22S.

Bowlby, J. (1940). The influence of early environment in the development of
neurosis and neurotic character. International Journal of Psycho-Analysis, 1,

154-178.
Bowlby, j. (1944). Forty-four juvenile thieves: Their characters and home-life.
International Journal of Psycho-Analysis, 25, 19-52, 107-127. Reprinted (1946) as

monograph. London: Bailiere, Tindall and Cox.

Bowlby, J. (1951). Maternal care and mental health. Geneva: World Health
Organization Monograph Series (2).
Bowlby, j. (1958a). Can I leave my baby? Ltmdon: National Association tor Mental
Health.
 ,

472 Bibliography

Bowlby, J. (1958b). The nature of the child’s tie to his mother. International Journal

of Psycho-analysis 39, 350-373.

Bowlby, J. (1959). Separation anxiety. International Journal of Psycho-Analysis, 41,
89-113.
Bowlby, J. (1960). Grief and mourning in infancy. The Psychoanalytic Study of the
Child, 15,3-39.

Bowlby, J. (1967). Foreword. In M. D. S. Ainsworth, Infancy in Uganda: Infant care

and the growth of love. Baltimore: Johns Hopkins.
Bowlby, J. (1970). Child care and the growth of love (2nd ed.). Harmondsworth:
Pelican.
Bowlby, j. (1973). Attachment and loss. Vol. 2: Separation. New York: Basic.
Bowlby, J. (1979). The making and breaking of affectional bonds. New York: Routledge.

Bowlby, J. (1980). Attachment and loss. Vol. 3: Loss, sadness and depression. New

York: Basic.
Bowlby, J. (1982). Attachment and loss. Vol. I: Attachment. New York: Basic (rev.

ed.).

Bowlby, J. (1985). The role of the psychotherapist’s personal resources in the treat-

ment situation. Unpublished manuscript.

Bowlby, J. (1987). Early days at the London child guidance clinic and training cen-
tre. Unpublished manuscript.
Bowlby, j. (1988a). Developmental psychiatry comes of age. American Journal of
Psychiatry, 145, 1-10.

Bowlby, j. (1988b). A secure base: Clinical applications of attachment theory. London:

Routledge.
Bowlby, J. (1991). Charles Darwin: A new New York: Norton.
life.

Bowlby, ]., Figlio, K., & Young, R. (1990). An interview with John Bowlby on the
origins and reception of his work. Free Associations, 21, 36-64-
Brazelton, T. B. (1983). Infants and mothers: Differences in development. New York:

Delta, (rev. ed.).

Brazelton, T. B., Koslowski, B., & Main, M. (1974). The origins of reciprocity: The
early mother-input interaction. In M. Lewis &. L. Rosenblum (Eds.), The effect of

the infant on its caregiver. New York: Wiley.

Bretherton, 1. (1985). Attachment theory: Retrospect and prospect. In 1. Bretherton
& E. Waters (Eds.), Growing points in attachment theory and research. Monographs
of the Society for Research in Child Development, 50 (Serial No. 209).
Bretherton, 1. (1987). New perspectives on attachment relations: Security, commu-
nication, and internal working models. In J. Osofsky (Ed.), Haiidbook of infant
development (2nd ed.). New York: Wiley.
Bretherton, 1. (1991). The roots and growing points of attachment theory. In C. M.
Parkes, J. Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life cycle.

New York: Tavistock/Routledge.

«

Bretherton, 1. (1992). The origins of attachment theory: John Bowlhy and Mary
Ainsworth. Developmental Psychology, 28, 759-775.
Bronfenbrenner, U. (1986, Fall). A generation in jeopardy: America’s hidden family
policy. Developmental Psychology Newsletter, 47-54.
 ^

Bibliography 473

Bruch, H. (1954). Parent education or the illusion of omnipotence. American Journal

of Orthopsychiatry, 24, 723-732.
Burlingham, D., & Freud, A. (1942). Young children in wartime London. London:
Allen Unwin.
Burlingham, D., & Freud, A. (1944). Infants without families. London: Allen &
Unwin.
Burnham, D. L. (1965). Separation anxiety. In Archives of General Psychiatry, 13,
346-358.
Byng'Hall, J. (1991). [Address at John Bowlby Memorial Service]. Infant Mental
Health Journal, 12, 267-268.
Byrne, S. (1977, May). Nobody home: The erosion of the American family. (A con-
versation with Urie Bronfenbrenner). Psychology Today, pp. 41-48.
Campos,]. ]., Barrett, K. C., Lamb, M. E., Goldsmith, H. H., & Stenberg, C. (1983).
Socioemotional development. In M. M. Haith & J. j. Campos (Eds.), Handbook
of child psychology: Vol. 2. Infancy and psychobiology (pp. 783-915). New York:

Wiley.
Caper, R. (1988). Immaterial Facts. Northvale, NJ: Aronson.
Casler, L. (1961). Maternal deprivation: a critical review of the literature. Mono-
graphs of the Society for Research in Child Development, 26 (Serial No., 80).
Caspi, A., & Elder, G. H. (in press). Emergent family patterns: The intergen-

erational construction of problem behavior and relationships. In R. Hinde &.

J. Stevenson-Flinde (Eds.), Relations between relationships within families. Oxford:

Oxford University Press.

Cassidy, J. (1988). The self as related to child-mother attachment at six. Child

Development, 59, 121-134.

Cassidy, j. (1990). Theoretical and methodological considerations in the study of
attachment and the self in young children. In M. Greenberg, D. Cicchetti, & E.
M. Cummings (Eds.), Attachment in the preschool years: Theory, research, and inter

vention. Chicago: University of Chicago Press.

Cassidy, ]., <Sl Berlin, L. J. (1992). The insecure/ambivalent pattern of attachment:

Theory and research. Unpublished manuscript.
Chess, S. (1964). Mai de Mere. American Journal of Orthopsychiatry, 64, 613-614.
Cassidy, J. and Shaver, P. R., eds.. Handbook of Attachment Theory and Research. NY:
Guilford.
Chess, S. (1987, February). Comments: “Infant day care: A cause for concern.” Zero
to Three, pp. 24-25.
Chess S., &. Thomas, A. (1982). Infant bonding: Mystique and reality. American
Journal of Orthopsychiatry, 52, 213-222.
Chess, S. & Thomas, A. (1987). Origins and evolution of behavior disorders. Cam-
bridge, MA: Harvard University Press.

Chess, S., Thomas, A., Birch, H. (1959). Characteristics of the individual child’s

behavioral responses to the environment. American Journal of Orthopsychiatry, 29,

791-802.
Chess, S., Thomas, A., Birch, H. G. (1965). Your child is a person. New York:

Viking Press.
 .

474 Bibliography

Chira, S. (1994, April 4). Broad study says home'based day care, even if by relatives,

often fails children. New York Times, p. 20.

Ciccbetti, D., & Beegbly, M. (1990). Children with Down Syndrome: Developmental

perspective. New York: Cambridge University Press.

Clarke, A. M., & Clarke, A. D. B. (Eds.) ( 1976). Early experience: Myth and evidence.

New York: Free Press.

Clarke'Stewart, K. A. (1989, February). Infant day care: Maligned or malignant?
American Psychologist, 44, 266-273.
Collins, G. (1987, November 25). Day care for infants: Debate turns to long-term
effects. The New York Times.
Crittenden, P. (1988). Relationships at risk. In J. Belsky & T. Nezworski (Eds.),

Clinical implications of attachment. Hillsdale, NJ: Erlbaum.

Crockenberg, S. (1981). Infant irritability, mother responsiveness, and social support

influences on the security of infant-mother attachment. Child Development, 7,

169-176.
Crowel, J. A. & Waters, E. (1995). Is the parent-child relationship prototype of later

love relationships? Studies of attachment and working models of attachment.

Poster symposium presented at the Society for Research in Child Development,
Indianapolis, Indiana, March 30-April 2, 1995.

Dinnage, R. (1980, May). Understanding loss: The Bowlby canon. Psychology Today,

pp. 56—60.
Dollard, J., & Miller, N. E. (1950). Personality and psychotherapy New York:

McGraw-Hill.
Dunn, J., & Plomin, R. (1990). Separate lives: Why siblings are so different. New York:
Basic.

Eagle, M. (1987). Recent developments in psychoanalysis. Cambridge: Harvard Univer-

sity Press.

Eagle, M. (1992, August). The relationship between attachment theory and psycho-
analysis. Paper presented at the annual convention of the American Psycho-
logical Association, Washington, D.C.
Easterbrooks, M. A., & Goldberg, W. A. (1990). Security of toddler-parent attach-

ment: Relation to children’s sociopersonality functioning during kindergarten.

In M. Greenberg, D. Cicchetti, &
E. M. Cummings (Eds.), Attachment in the

preschool years: Theory, research, and intervention. Chicago: University of Chicago

Press.

Egeland, B., & Farber, E. (1984). Infant-mother attachment: Factors related to its

development and changes over time. Child Development, 55, 753-771.

Egeland, B., Jacohvitz, D., <Sr Sroufe, L. A. (1988). Breaking the cycle of abuse:
Relationship predictions. Child Development, 59, 1080-1088.
Flicker, J., England, M., & Sroufe, L. A. (1992). Predicting peer competence and
peer relationships in childhood from early parent-child relationships. In R. Parke
& G. Ladd (Eds.), Family^peer relationships: Modes of linkage. Hillsdale, NJ:
Erlbaum.
Ellenberger, H. (1970).The Discovery of the Unconscious. New York: Basic.
Emde, R. (1983). Editor’s commentary: The Pinneau-Spitz controversy. In R. N.
 Bibliography 475

Emde (Ed.), Rene Spitz: Dialogues from infancy. New York: International Univet'

sities Press.

Erickson, M. E, Sroufe, L. A., & Egeland, B. (1985). The relationship between qual-

ity of attachment and behavior problems in preschool in a high risk sample. In I.

Bretherton & E. Waters (Eds.), Grotcing points in attachment theory and research.

Monographs of the Society for Research in Child Development, 50 (Serial No. 209,

Nos. 1-2), 147- 186.

Erikson, E. (1950). Childhood and Society. New York: Norton.

Evans, S. (1987, November 7). Psychological risk of day care told. The Washington

Post.

Fairbairn, W. R. D. (1952). An object relations theory of the personality. New York:

Basic.

Fonagy, P. Psychoanalytic and empirical approaches to developmental psychopathol-

ogy: An object relations perspective. Unpublished manuscript.
Fonagy, R, Leigh, T, Kennedy, R., Matoon, G., Steele, H., Target, M., Steele, M., &
Higgitt,A. (1992, October). Attachment, borderline states and the representation of
emotions and cognitions in self and other. Paper presented at Symposium on Devel-
opmental Psychopathology, Rochester, New York.

Fonagy, P, Steele, H., & Steele, M. (1991a). Maternal representations of attach-

ment during pregnancy predict the organization of infant-mother attachment at

one year of age. Child Development, 62, 891-905.

Fonagy, R, Steele, M., Steele, H., Moran, G., & Higgitt, A. (1991b). The capacity
for understanding mental states: The reflective self in parent and child and its

significance for security of attachment. Infant Mental Health Journal, 13, 200-217.
Fonagy, P, Steele, M., Moran, G., Steele, H., &
Higgitt, A. (1991c). Measuring the

ghost in the nursery: An empirical study of the relation between parents mental
representations of childhood experiences and their infants’ security of attach-
ment. Unpublished manuscript.
Fraiberg, S. (1977). Every child’s birthright: In defense of mothering. NY: Basic.

Fraiberg, S., Adelson, E., & Shapiro, V. (1975). Ghosts in the nursery: A psychoana-
lytic approach to the problems of impaired infant-mother relationships. Journal of
the American Academy of Child Psychiatry, 14, 387-421.
Freud, A. (1960). Discussion of John Bowlby’s paper. The Psychoanalytic Study of
the

Child, 15,53-62.

Fromm, E. (1955). The sane society. New York: Holt, Rinehart, & Winston.
Furman, E. (1981). Treatment-via-the-parent: A case ot bereavement. Journal of

Child Psychotherapy, 1: 89-102.

Gabon, F. (1979). Hereditary genius: An inquiry into its laws and consequences.

London: Julian Friedmann. (Original work published 1869).

Galvin, R. M. (1992, March/April). The nature of shyness. Harvard Magazine, pp.

41-45.
Gewirtz, J. L. (1961). A learning analysis of the effects of normal stimulation, priva-

tion and deprivation on the acquisition of social motivation and attachment. In

B. M. Foss (Ed.), Determinants of infant behavior. London: Methuen; New York:

Wiley.
 ,

476 Bibliography

Gilmore, K. (1990). [Review of A Secure Base]. Psychoanalytic Quarterly, 59,

494-498.
Goldberg, S. (1991). Recent developments in attachment theory. Canadian Journal
of Psychiatry 36, 393-400.
Goleman, D. (1991, February 12). Feeling cheerful? Thank brain’s left lobe. The
New York Times, pp. Cl, CIO.
Goncharov. I. (1954). Oblomov. (D. Margarshack, Trans.). Middlesex, U.K.:

Penguin. (Original work published 1859)

Goodman, S. (1991, December-1992, January). Presumed innocents. Modem Ma^
turity, pp. 25-29.
Gottesman, 1., & Shields, J. (1972). Schizophrenia and genetics: A twin study vantage

point. New York: Academic Press.

Greenberg, ]. R., & Mitchell, S. A. (1983). Object relations in psychoanalytic theory.

Cambridge: Flarvard University Press.

Greenspan, E. (1983). Little winners: Inside the world of the child sports star. New York:
Little, Brown.
Grosskurth, P. (1987). Melanie Klein: Her World and Her Work. Cambridge: Flarvard
University Press.
Grosskurth, P. (1991, September 29). What women wanted. (Review of Mothers of
Psychoanalysis.) The New York Times Book Review.
Grossmann, Karin. (1989, September). Avoidance as a communicative strategy in

attachment relationships. Paper presented at the fourth world congress of the

World Association of Infant Psychiatry and Allied Disciplines, Lugano, Swit-

zerland.
Grossmann, K., & Grossmann, K. (1991). Attachment quality as an organizer of
emotional and behavioral responses in a longitudinal perspective. In C. M.
Parkes, j. Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life cycle.

New York: Tavistock/Routledge.

Grossmann, K., Grossmann, K. E., Spangler, G. Suess, G., & Unzer, J. (1985).

Maternal sensitivity and newborns’ orientation responses as related to quality of

attachment in northern Germany. In 1. Bretherton & E. Waters (Eds.), Growing

points in attachment theory and research. Monographs of the Society for Research in

Child Development, 50 (Serial No. 209) 233-256.

Haft, W, <St Slade, A. (1989). Affect attunement and maternal attachment: A Pilot

study. Infant Mental Health Journal, 10, 157-172.

HaineS'Stiles, G. (Executive producer). (1991). Childhood. [Television series.] New
York: Thirteen/WNET and The Childhood Project.

Hamilton, V. (1990, Autumn). Reflections on 23 years of learning and friendship

with John Bowlby. The Tavistock Gazette, 29, 20-23.
Harlow, H. ( 1958). The nature of love. The American Psychologist, 3, 673-685.
Harlow, H. E, & Harlow, M. K. (1966). Learning to love. American Scientist, 54,

244-272.
Hartup, W. W. (1963). Dependency and independence. In H. W. Stevenson (Ed.),
Child psychology: The sixty 'second yearbook of the National Society for the Study of
Education, Part 1. Chicago: University of Chicago Press.
 Bihlio^aphy 477

Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment

process. Journal of Personality and Social Psychology, 52, 511-524.
Hinde, R. (1991). Obituary: John Bowlby. Journal of Child Psychology and Psychiatry,
32,2,215-217.
Hopkins, j. (1992). Failure of the holding relationship: some effects of physical

rejection on the child’s attachment and inner experience. In C.M. Parkes, J.

Stevenson-Hinde, & P. Marris (Eds.), Attachment across the life cycle. New York:

Tavistock/Routledge.
Hughes, (1989). Reshaping the psychoanalytic domain: The work of Melanie Klein, W.
J.

R. D. Fairbaim, D. W. Winnicott. Berkeley and Los Angeles: University of Cali-

fornia Press.
Isabella, R. A., & Belsky, J. (1991). Interactional synchrony and the origins of
infant-mother attachment: A replication study. Child Development, 62, 373-384.
Isabella, R. A., Belsky, ]., & von Eye, A. (1989). Origins of mother-infant attach-

ment: An examination of interactional synchrony during the infant s first year.

Developmental Psychology, 25, 12-21.

Isherwood, C. (1965). Ramakrishna and his disciples. New York: Simon and Schuster.

Jackson, D. (1987, July 7). Is day care bad for infants? Newsday, Part II, p. 3.

Jackson, D. D. (1980, October). Reunion of identical twins raised apart reveals some
astonishing similarities. Smithsonian, pp. 48-56.
Jacobson, S. W., & Erye, K. E (1991). Effect of maternal social support on attach-

ment: Experimental evidence. Child Development, 62, 572—582.

Jensen, A. R. (1973). Educability and group differences. New York: Basic.

Kagan, J. (1979). Overview: Perspectives on human infancy. In J. Osofsky (Ed.),

Handbook of infant development. New York: Wiley.

Kagan, J. (1982). Psychological research on the human infant: An evaluative summary.

New York: W.T. Grant.

Kagan, J. (1984). The nature of the child. New York: Basic.

Kagan, (1989, August). The return of temperament in psychological theory. Paper pre-
J.

sented at the 97th annual convention of the American Psychological Asso-

ciation, New Orleans.
Kagan, J.,
Kearsley, R. B., & Zelazo, P. (1980). Infancy: Its place in human develop-

ment. Cambridge: Harvard University Press.

Kagan, &. Klein, R. (1973). Cross-cultural perspectives on early development.
J.,

American Psychologist, 28, 947-961.

Kaplan, N. (1987, May). Internal representations of attachment in six-year-olds. Paper
presented at the biennial meetings of the Society tor Research in Child Develop-
ment, Baltimore.
Karen, R. (1987). Top dog/Bottom dog. New York: Donald 1. Fine.

Karen, R. (1990, February). Becoming attached. The Atlantic, pp. 35-70.

Karen, R. (1992, February). Shame. The Atlantic, pp. 40-70.
Kiser, L., Bates, J., Maslin, C., & Bayles, K. (1986). Mother-infant play at six

months as a predictor of attachment security at thirteen months. Journal of the

American Academy of Child Psychiatry, 25, 168-175.

Klein, M. (1975). The psychoanalysis of children. New York: Free Press.
 .

478 Bibliography

Kobak, R. (1987, April). Attachment, affect regulation and defense. Paper presented
atthe biennial meetings of the Society for Research in Child Development.
Kohn, A. (1993, April). Back to nurture. American Health, pp. 29-31.
Kohut, H. (1977). The restoration of the self. New York: International Universities

Press.

Kohut, H. (1980). Summarizing reflections. In A. Goldberg (Ed.), Advances in self

psychology. New York: International Universities Press.

Konner, M. (1989, January 8). Where should baby sleep? The New York Times

Magazine, pp. 39-40.

Kraemer, S. (1990). Bowlhy speaking. The Tavistock Gazette, 29, 16-17.

LaBarba, R. (1981). Foundations of developmental psychology. New York: Academic

Press.

Lally, J. R. (Executive producer), & Mangione, P. L. (Director). (1979). Flexible,

fearful, feisty: The temperaments of infants and toddlers [Video]. Sausalito,

different

CA: Far West Laboratories Center for Child and Family Studies for California
State Department of Education.
Lamb, M. E., Thompson, R. A., Gardner, W. R, Charnov, E. L., & Estes, D. (1984).

Security of infantile attachment as assessed in the “Strange Situation.’’ Behavioral

and Brain Sciences, 7, 127-171.
Lasch, C. (1979). The culture of narcissism. New York: Norton.

Laslett, P. (1962). The world we have lost. In E. Josephson & M. josephson (Eds.),

Man alone: Alienation in modern society. New York: Dell.

Leach P. (1990). Your baby and child (rev. ed.). New York: Knopf.

Levy, D. (1937). Primary affect hunger. American Journal of Psychiatry, 94,

643-652.
Levy, D. (1943). Maternal overprotection. New York: Columbia University Press.

Lewis, H. B. (1987). The role of shame in depression over the life span. In H. B.
Lewis (Ed.), The role of shame in symptom formation Hillsdale, Nj: Erlhaum.

Lewis, M., Feiring, C., McGuffog, C., & jaskir j. (1984). Predicting psychopathology

in six-year-olds from early social relations. Child Development, 55, 123-136.

Lieherman, A. (1991). Attachment theory and infant-parent psychotherapy: Some

conceptual, clinical and research considerations. In D. Cicchetti & S. Toth
(Eds.), Models and integrations: Rochester Symposium on Developmental Psycho-
pathology. Rochester: University of Rochester Press.

Lieherman, A. F. (1992). Infant-parent psychotherapy with toddlers. Development

and Psychopathology, 4, 559-574.
Lieherman, A. F. (1993). The emotional life of the toddler. New York: Free Press.

Lieherman, A. E, & Pawl, j. H. (1990). Disorders of attachment and secure base

behavior in the second year of life: Conceptual issues and clinical intervention.

In M. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the

preschool years: Theory, research, and intervention. Chicago: University of Chicago

Press.

Lieherman, A. E, Weston, D., and Paul, j. H. (1991). Preventive intervention and

outcome with anxiously attached dyads. Child Development, 62, 199-209.
Liedloff, j. (1977). The continuum concept: Allowing human nature to work successfully.

Reading, MA: Addison-Wesley.

 Bibliography 479

Liotti, G. (1990, November). John Bowlby: In memoriam. Paper given at conference

in honor of John Bowlby, Naples.

Liotti, G. (1992). Disorganized/disoriented attachment in the etiology of dissocia-

tive disorders. Dissociation, 5, 196-204.

Lopez, T, &
Kliman, G. (1979). Memory, reconstruction, and mourning in the
analysis of a 4-year-old child: Maternal bereavement in the second year of
life.

Psychoanalytic Study of the Child, 34: 235-371.

Maccoby, E. E., & Martin, J. A. (1983). Socialization in the context of the family. In

E. M. Hetherington (Ed.), Handbook of child psychology: Socialization personality

and social development (Vol. 4). New York: Wiley.

Macdonald, D. (1965). Against the American grain. New York: Vintage.

Mackenzie, M. (1990, Autumn). Dr. Marion Mackenzie remembers Dr. John Bowl-
The Tavistock Gazette, 29, 13-16.
by.

Mahler, M. S., Pine, E, &

Bergman, A. (1975). The psychological birth of the human
infant. New York: Basic.

Main, M., & Goldwyn, R. (in press). Interview-based adult attachment classifica-

Related to infant-mother and infant-father attachment.

tions:

Main, M. (1977). Analysis of a peculiar form of behavior seen in some daycare chil-
dren: Its history and sequelae in children who are home-reared. In R. Webb
(Ed.), Social Development in Daycare. Baltimore: Johns Hopkins University Press.

Main, M. (1991). Metacognitive knowledge, metacognitive monitoring, and singu-

lar (coherent) vs. multiple (incoherent) model of attachment:
Findings and

directions for future research. In C. M. Parkes, J.

Stevenson-Hinde, & P. Marris

(Eds.), Attachment across the life cycle. New York: Tavistock/Routledge.

Main, M., & Cassidy, J. (1988). Categories of response with the parent at age six:

Predicted from infant attachment classifications and stable over a one month
period. Developmental Psychology, 24, 415-426.
Main, M., & Hesse, E. (1990). Parents’ unresolved traumatic experiences are related
to infant disorganized attachment status: Is frightened and/or frightening parental

behavior the linking mechanism? In M. Greenberg, D. Cicchetti, E. M. &

Cummings (Eds.), Attachment in the preschool years. Chicago: The University of
Chicago Press.

Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and
adulthood: A move to the level of representation. In I. Bretherton & E. Waters

(Eds.), Growing points in attachment theory and research. Monographs of the Society

for Research in Child Development, 50 (Serial No. 209), 66-104.

Main, M. & Solomon, (1990). Procedures for identifying infants as disorganized/
J.

disoriented during the Ainsworth Strange Situation. In M. Greenberg, D.

Cicchetti, &. E. M. Cummings (Eds.), Attachment in the preschool years: Theory,
research, and intervention. Chicago: University of Chicago Press.

Main, M., & Weston, D. (1981). The quality of the toddler’s relationship to mother

and to father as related to conflict behavior and readiness to establish new rela-

tionships. Child Development, 52, 932-940.

Main, M., &

Weston, D. (1982). Avoidance of the attachment figure in infancy:
Descriptions and interpretations. In C. M. Parkes and J. Stevenson-Hinde (Eds.),
The place of attachment in human behavior (pp. 31-59). New York: Basic.
 .

480 Bibliography

Malan, D. (1990). [Contribution to Reminiscences and reflections on John Bowlby].

The Tavistock Gazette, 29, 13-16.
Mangelsdorf, S., Gunnar, M., Kestenbaum, R., Lang, S., & Andreas, D. (1990)
Infant proneness-tO'distress temperament, maternal personality, and mother-
infant attachment: Associations and goodness of fit. Child Development, 61,
820-831.
Martin, F. E. (1991). John Bowlby, 1907-1991: A tribute and personal appreciation.
Melanie Klein and Object Relations, 9, 63-69.
Marvin, R. S. (1977). An ethological-cognitive model for the attenuation of
mother-child attachment behavior. In T. Alloway, R Pliner, &. L. Krames (Eds.),

Attachment behavior. New York: Plenum.

Marvin, R. S., & Stewart, R. B. (1991). A family system framework for the study of
attachment. In M. Greenberg, D. Cicchetti, & M. Cummings (Eds.), Attachment
beyond the preschool years (pp. 51-86). Chicago: University of Chicago Press.
Mason, E. (1991). Where’s mummy! The Robertson studies of parent-child separation
[Video]. University Park, PA: Penn State Audio Visual Services. [In references,

page numbers refer to transcript.]

Matas, L., Arend, R. A., & Sroufe, L. A. (1978). Continuity and adaptation in the
second year: The relationship between quality of attachment and later compe-
tence. Child Development, 49, 547-556.
Mead, M. (1962). A cultural anthropologists’s approach to maternal deprivation. In
Deprivation of maternal care: A reassessment of its effects. Public Health Papers no.

14, Geneva: World Health Organization.

Miller, A. ( 1981 ). The drama of the gifted child. New York: Basic.
Miller, A. (1990). Banished knovuledge: Facing childhood injuries. New York: Double-

day.

Mitchell, E. (1991, September 30). Look who’s listening too. Time, p. 76.

Miyake, K., Shing-jen, C., Campos, J. J. (1985). Infant temperament, mother’s mode
of interaction, and attachment in Japan: An interim report. In 1. Bretherton <St E.

Waters (Eds.), Growing points in attachment theory and research. Monographs of the

Society for Research in Child Development, 50 (Serial No. 209), 116-291

Moore, S., &
M. (Producers). (1993, April 19). Medicine at
Ereeth, the crossroads.

New York: Thirteen/WNET and BBC-TV.

Morris, D. (1983). Attachment and intimacy. In G. Strieker & M. Eisher (Eds.),
Intimacy. New York: Plenum.
Morrison, A. (1989). Shame: The underside of narcissism. Hillsdale, NJ: Analytic
Press.

Mussen, P, Conger, J., & Kagan, J. (1956). Child development and personality (4th
ed.). New York: Harper & Row.
Neubauer, P. B., & Neubauer, A. (1990). Nature’s thumbprint. Reading, MA:
Addison-Wesley.
Osofsky, J. (1988). Attachment theory and research in the psychoanalytic process.
Psychoanalytic Psychology, 5, 159-177.
Pastor, D. L. ( 1981 ). The quality of mother-infant attachment and its relationship to
toddler’s initial sociability with peers. Developmental Psychology, 17, 323-335.
Phillips, A. (1988). Winnicott. London: Fontana.
 ,

Bibliography 48 1

Phillips, D., McCartney, K., Scarr, S., & Howes, C. (1987, February). Selective
review of infant day care research; A cause for concern. ZERO TO THREE, pp.

18-25.
The infantile disorders of hospitalism and anaclitic depres-
Pinneau, S. R. (1955).
sion. Psychological Bulletin, 52, 429—452.
Plomin, R. (1987). Developmental behavioral genetics and infancy. In J. D. Osofsky

(Ed.),Handbook of infant development. New York: Wiley.

Rayner, E. (1991, October). John Bowlby’s contribution, a brief summary.
Paper read

at Bowlby Memorial Service, British Psycho- Analytic

Society, London.

Renken, B., Egeland, B., Marvinney, D., Mangelsdorf, S., & Sroufe, L. A. (1989).
in early ele-
Early childhood antecedents of aggression and passive-withdrawal
mentary school. Journal of Personality, 57, 257—281.

Ricks, M. (1985). The social transmission of parental behavior: Attachment across

generations. In I. Bretherton &. E. YC^aters (Eds.), GTOtt^ing points in attachment the'

ory and research. Monographs of the Society for Research in

Child Development, 50

(Serial No. 209), 21 1-227.

Ricks, T. E. (1987, March 3). Day care for infants is challenged by research on psy-

chological risks. The Wall Street Journal, p. 35.

Rioch, (1988). The transference phenomenon in psychoanalytic therapy. In B.

J.

Wolstein (Ed.), Essential papers on countertransference. New York: New York Uni-

versity Press. (Reprinted from Psychiatry, 1943, 6, 147-156.)

Riviere, j. (1927). (Contribution to Symposium on child analysis). In International

Journal of Psychoanalysis 8, 373-377.

Robertson, (1953). A two'year'old goes to hospital [Film]. University Park, PA: Penn
J.

State Audio Visual Services.

Robertson, (1962). Hospitals and children: A parent'S'eye vieiv. New York:

J.

Gollancz.
Robertson, (1967). Kate, aged two years five months, in foster care
Robertson, ]., ].

for twenty'Seven days [Film]. Young Children in Brief Separation Film Series.

University Park, PA: Penn State Audio Visual Services.

Robertson, J., & Robertson, j. (1968). Jane, aged seventeen months, in foster care for

ten days [Film]. Young Children in Brief Separation Film Series. University Park,

PA; Penn State Audio Visual Services.

Robertson, 1969). John, aged 7 months, for nine days in a residem
Robertson, j., ]. ( 1

tial nursery [Film]. Young Children in Brief Separation Film Series. University

Park, PA: Penn State Audio Visual Services.

Robertson, ]., Robertson, J. (1971). Thomas, aged two years four moriths, in foster

Young Children in Brief Separation Film Series. Uni-

care for ten days [Film].

versity Park, PA; Penn State Audio Visual Services.

Robertson, j., & Robertson, J. (1973). Lucy, aged twenty'one months, in foster care for

nineteen days [Film]. Young Children in Brief Separation Film Series. University

Park, PA: Penn State Audio Visual Services.

Robertson, ]., & Robertson, j. (1973). Substitute mothering for the unaccompanied

child. Nursing Times, November 29. Offprint.

Robertson, ]., Robertson, j. (1989). Separation and the very young. London: Free

Association Books.
 .

482 Bibliography

Rodell, S. (1993, July 25). Memo to Hillary: Please have a look at Australia’s system

lEditorial notebook]. The New York Times, p. 16.

Rosen, C. M. (1987, September). The eerie world of reunited twins. Discover, pp.
36-46.
Rutter, M. (1981). Maternal deprivation reassessed. New York: Penguin.
Rycroft, C. (1985). Psychoanalysis and beyond. London: Hogarth.
Rycroft, C. ( 1992). Rycroft on analysis and creativity. New York: New York University
Press.

Sagi, A., Lamb, M. E., Lewkowitcz, K. S., Shoham, R., Dvir, R., & Estes, D. (1985).
Security of infant-mother, -father, and -metapelet attachments among kibbutz-
reared Israeli children. In I. Bretherton & E. Waters (Eds.), Growing points in
attachment theory and research. Monographs of the Society for Research in Child
Development, 50 (Serial No. 209), 257-275.
Sander, L. (1964). Adaptive relationships in early mother-child interaction. Journal
of the American Academy of Child Psychiatry, 1, 141-166.
Sargent, S., & Stafford, K. (1965). Basic teachings of the great psychologists. Garden
City, NY: Dolphin.
Sayers, J. New York: Norton.
(1991). Mothers of psychoanalysis
Scarf, M. (1976). Body, mind, behavior. New York: Dell.
Scarr, S. (1984). Mother care/Other care. New York: Basic.

Scarr, S. (April, 1991). Developmental theories for the 1990s: Development and individ'

ual differences. Paper presented at the biennial meetings of the Society for
Research in Child Development, Seattle.
Scarr, S., Phillips, D., & McCartney, K. (1993). Facts, fantasies and the future of
child care in the United States. In R. H. Wozniak (Ed.), Worlds of childhood. New
York: HarperCollins. (Reprinted from Psychological Science, 1990, 1, 26-35.)
and Weinberg, R. A. (1993). The nature-nurture problem revisited: The
Scarr, S.

Minnesota adoption studies. In R. H. Wozniak (Ed.), Worlds of childhood. New

York: Harper Collins, pp. 92-105. (Reprinted from I. Sigel & G. Brody (Eds.),
Methods of family research: Biographies of research projects. Volume 1: Normal famP
lies. Hillsdale, NJ: Erlbaum, pp. 121-151.

Schaffer, H. R. & Emerson, P. E. (1964). The development of social attachments in

infancy. Monographs of the Society for Research in Child Development, 29, (3, Serial
No. 94).
Schoenewolf, G. (1990). Turning points in analytic therapy. Northvale, NJ: Aronson.
Schore, A. (1994). Affect regulation and the origin of the self: The neurobiology of
emotional development. Hillsdale, N.J.: Erlbaum.
Schur, M. (1960). Discussion of John Bowlby’s paper. The Psychoanalytic study of the

child, 1 5, 63-84.
Segal, H. (1981). The work of Hanna Segal: A Kleinian approach to clinical practice.
Northvale, NJ: Aronson.
Segal, N. L. (1984, July/August). The nature vs. nurture laboratory. Twins, pp.
56-57.
Senn, M. J. E. (1977a). Interview with John Bowlby. Unpublished manuscript.
National Library of Medicine, Washington, D.C.
 Bibliography 483

Senn, M. J. E. (1977b). Interview with James Robertson. Unpublished manuscript,

National Library of Medicine, Washington, D.C.
Sennett, R. (1977). The fall of public man. New York; Knopf.

Shaver, P. Hazan, C. (1992). Adult romantic attachment: Theory and evidence.

&
In D. Perlman W. Jones (Eds.), Advances in personal relationships. Bristol, PA:
&
Taylor & Francis.
Silverman, D. K. (1991). Attachment patterns and Freudian theory:
An integrative

proposal. Psychoanalytic Psychology, 8, 169-193.

The story unfolds. Paper pre-
Slade, A. (1992, September). Mothers and their mothers:
Psychiatry and Allied
sented at the meeting of the World Association of Infant
Disciplines, Chicago.

Slade, A. (1992, October). Processes of parenting:

Remembrance of things past. Julia
Howe YC^ard Lecture presented at Barnard College, New York.
Clinical and theoretical
Slade, A. (1993, March). Affect regulation and defense:
of the Society for
considerations. Paper presented at the biennial meetings
Research in Child Development, New Orleans.
Attachment theory and research: Implications for the
Slade, A. (in press, 1997a).
theory and practice of individual psychotherapy
with adults. In Cassidy, J. and
Research. NY; Guilford.
Shaver, P. R., eds., Handbook of Attachment Theory and
symbolization, and affect regulation in
Slade, A. (in press, 1997b). Representation,
the concomitant treatment of a mother and child:
Attachment theory and child
psychotherapy. Psychoanalytic Inquiry.
in children
Slade, A., & Aber, L. (1987, October). Representational processes
presented at the
and adults: Clinical and developmental perspectives. Paper
for Contemporary Psychotherapy, New
meeting of the Society of the Institute
York City.
Unpublished manuscript. Nation-
Smuts, A. (1977). Interview with Dr. John Bowlby.
al Library of Medicine, Washington,
D.C.
Sommerville, C. J. (1990). The rise and fall of childhood. New York: Vintage.

Spitz, R. (1945). Hospitalism: An inquiry into the genesis of psychiatric conditions

The psychoanalytic study of the child, 1, 53-74.

in early childhood.
A peril in infancy [Film]. University Park, PA: Penn State
Spitz, R. (1947). Grief:

Audio Visual Services.

Bowlby ’s paper. The Psychoanalytic study of the
Spitz, R. (1960). Discussion of Dr.

child, 15,85-94.
o ju
Osofsky (Ed.), Handbook
; /
of
Sroufe, L. A. (1979). Socioemotional development. In J.

infant development. New York: Wiley.

Sroufe, L. A. (1983). Infant-caregiver

attachment and patterns of adaptation in

preschool: The roots of maladaptation and

competence. In M. Perlmutter (Ed.),
41-81). Hillsdale, NJ;
Minnesota symposium in child psychology (Vol. 16, pp.
Erlbaum. r-LUL j
Sroufe, L. A. (1988). A developmental perspective on day care. Early Chimood
Research Quarterly, 3,283-291.
at City University of New York, Graduate
Sroufe, L. A. (1989, Fehruary 10). Talk
Center.
484 Bibliography

SroLife, L. A., Carlson, E., & Shulman, S. ( 1993). The development of individuals in
relationships: From infancy through adolescence. Unpublished manuscript.
SroLife, L. A., Cooper, R. G., & DeHart, G. B. (1992). Child development: Its nature
and course (2nd ed.). New York: McGraw-Hill.
Sroufe, L. A., Egeland, B., & Kreutzer, T. (1990). The fate of early experience fol-
lowing developmental change: Longitudinal approaches to individual adaptation
in childhood. Child Development, 61, 1363-1373.
Sroufe, L. A., Fox, N. E. Pancake, V. R. (1983). Attachment and dependency in
developmental perspective. Child Development, 54, 1615-1627.
Sroufe, L. A., & Waters, E. (1977a). Attachment as an organizational construct.
Child Development, 48, 1184-1 189.
Sroufe, L. A. & Waters, E. (1977b). Heart rate as a convergent measure in clinical
and developmental research. MerrilPPalmer Quarterly, 23, 3-27.
Stern, D. N. (1985). The interpersonal world of the infant. New York: Basic.
Suess, G. J., Grossman, K. E., & Sroufe, L. A. (1992). Effects of infant attachment to
mother and father on quality of adaptation in preschool: from dyadic to individ-
ual organisation of self. International Journal of Behavioral Development, 15 (1)
43-65.
Sulloway, F. J. (1991, October 1). Darwinian psychobiography. [Review of Charles
Darwin: A new life]. The New York Review of Books, pp. 29-32.
Suomi, S. J., & Harlow, H. F. (1972). Social rehabilitation of isolate-reared mon-
keys. Developmental Psychology, 6, 487-496.
Suomi, S. J., & Harlow, H. F. (1978). Early experience and social development in
rhesus monkeys. In M. Lamb (Ed.), Social and Personality Development. New York:
Holt, Rinehart, & Winston.
Sutherland, J. (1990, Autumn). John Bowlby: Some personal reminiscences. The
Tavistock Gazette, 29, 13-16.
Tanner, ]., & Inhelder, on Child Development. Proceedings of
B. (1971). Discussions the
World Health Organization Study Group on the Psychological Development of the
Child, Geneva, 1953—1956. New York: International Universities Press.
Tellegen, A., Lykken, D. T, Bouchard, T. ]., Jr., Wilcox, K. J., Segal, N. L., & Rich,
S. (1988). Personality similarity in twins reared apart and together. Journal of
Personality and Social Psychology, 54, 1031-1039.
Thomas, A., Chess, S., <Sr Birch, H. G. (1969). Temperament and behavior disorders in
children. New York: New York University Press.
Todd, J. (1985, September). Finding my twin brought happiness beyond belief.
Woman, pp. 40-41.
Torrey, E. F. (1988). Surviving schizophrenia: A family manual (rev. ed.). New York:
Harper & Row.
Traub, J. (1986, March). Goodbye Dr. Spock: The brave new world ot better babies.
Harper’s, pp. 44-47.
Tronick, E., Als, H., Adamson, L., Wise, S., & Brazelton, T. B. (1978). The infant’s
response to intrapment bewteen contradictory messages in face-to-face interac-
tion. Journnl o/C/iiki Psyc/i/atry, 17, 1-13.

Trowell, J. (1990, Autumn). Dr. John Bowlby. The Tavistock Gazette, 29, 13-16.
 Bibliography 485

Troy,M. & Sroufe, A. (1987). Victimization among preschoolers: Role of attach-

L.

ment relationship theory. Journal of American Academy of Child and Adolescent

Psychiatry, 26, 166-172.
Turner, P. j. (1991). Relations between attachment, gender, and behavior with peers
in preschool. Child Development, 62, 1475-1488.
Vaillant, G. E. (1977). Adaptation to life. Boston: Little, Brown.

Vandell, D. L. & Corasaniti, M. A. (1990). Variations in early child care: Do they

predict subsequent social, emotional, and cognitive differences? Early Childhood

Research Quarterly, 5: 555-572.
van den Boom, D. C. (1988). Neonatal irritability and the development of attachment:
of
Observation and intervention. Unpublished doctoral dissertation. University
Leiden, Leiden, The Netherlands.
van den Boom, D. C. (in press). The influence of temperament and mothering
on attachment and exploration: An experimental manipulation of sensitive
responsiveness among lower class mothers with irritable infants. Child DeveU
opment.
Vaughn, B., Bradley, C., Joffe, L., Seifer, R., & Barglow, P. (1987). Maternal char-

acteristics measured prenatally are predictive of ratings of temperamental

Psy-
“difficulty” on the Carey Infant Temperament Questionnaire. Developmental
chology, 23, 152-161.
Deane, & Waters, E. (1985). The impact of out-of-home care on
Vaughn, B., K.,
questions. In 1.
child-mother attachment quality: Another look at some enduring
theory and research.
Bretherton &
E. Waters (Eds.), Growing points in attachment

Monographs of the Society for Research in Child Development, 50, (Serial

No. 209,

Nos. 1-2).
Vaughn, B. E., Lefever, G., Seifer, R., & Barglow, P. (1989). Attachment behavior,

attachment security, and temperament during infancy. Child Development, 60,

728-737.
Gove, E & Egeland, B. (1980). The relationship between out-of-
Vaughn, B., L.,

care and the quality of infant-mother attachment in

an economically dis-
home
advantaged population. Child Development, 49, 483-494.
Wachtel, P. (1977). Psychoanalysis and behavior therapy. New
York: Basic.

Waller, N. G., Kojetin, B. A., Bouchard, T. J., Jr.,

Lykken, D. T, Tellegen, A. &
religious interests, attitudes,
(1990). Genetic and environmental influences on
Psychological Science, 1,
and values: A
study of twins reared apart and together.

(2), 138-142. .

stability of individual differences in

intant-mother attach-
Waters, E. (1978). The
ment. Child Development, 49, 483-494
and
Waters, E„ Wippman, ]. &
Sroufe, L. A. (1979). Attachment, positive
affect,

Two studies in construct validation. Child

Deveh
competence in the peer group:

Opment, 50, 821-829.

Watson, j. B. (1928). Psychological care of infant and child. New York: Norton.

The parent trap. Insight, pp. 6-11, 34-35.

Wattenberg, D. (1992, March 2).

Weiss, R. S. (1975). Marital separation. New York:

Basic.

C. M. Parkes and j. Stevenson-

Weiss, R. S. (1982). Attachment in adult life. In
 , .

486 Bibliography

Hinde (Eds.), The place of attachment in human behaviour (pp. 171-184). New
York: Basic.
Winnicott, D. W. (1958). Through paediatrics to psycho-analysis New York: Basic.

Winnicott, D. W. (1965). The maturational processes and the facilitating environment.

New York: International Universities Press.

Wittenberg, 1. (1990, Autumn). John Bowlby. The Tavistock Gazette, 29, 13-16.
Yarrow, L. J. (1967). The development of focused relationships during infancy. In
Exceptional Infant, 1 ed. j. Hellmuth. Seattle: Special Child Publications.
Zeanah, C. H., Keener, M. A., Thomas, E, & Viera-Baher, C. C. (1987). Adolescent
mothers’ perceptions of their infants before and after birth. American Journal of
Orthopsychiatry, 57, 351-360.
Zeanah, C. H. & Zeanah, P. D. (1989). Intergenerational transmission of maltreat-
ment: Insights from attachment theory and research. Psychiatry, 52, 177-196.
Zero to THREE/National Center for Clinical Infant Programs. Heart Start: The
Emotional Foundations of School Readiness. Arlington, VA. [Booklet in two parts.]
Zero to Three, April/May, 1994. [Report on The Study of Children in Family Child
Care and Relative Care: Highlights of Findings (1994) by Ellen Galinsky, Carollee
Howes, Susan Kontos, and Marybeth Shinn.]
 INDEX

AAI. See Adult Attachment Interview 235, 239, 241-42, 245, 273, 284, 286, 316,

Aaron, Lewis, 431 328, 335, 434, 437-348

Aher, Lawrence, 242, 407 avoidant, 459n. 5 (ch. 17)

Abused children. See Child abuse child abuse and, 234-36

definition of, 165
Acting out, 48
Activity level, as temperamental characteristic, Ainsworth, Len, 130, 132, 143
274 Ainsworth, Mary, 6, 8, 9, 123-24, 125, 167,
Adaptability, as temperamental characteristic, 181, 197, 200, 201, 203, 286, 294, 350,

274-75 364,373,418
Adolescence background of, 1 29-30
attachment assessment in, 383 Bowlby relationship, 131-32, 433-35
avoidant attachment and, 232-33 on child stimulation vs. experience, 416
classification scales, 259
importance of reaching insecurely attached
commitment to psychoanalytic view, 208—9
children by, 229
criticisms of, 172, 264-66, 286, 298-99, 314,
secure attachment and, 197
Adoption, opposition to early, 16, 21 320-21,435-36
day-care controversy, 339
Adult
foster parents, support for, 255
attachment status, 363-69, 379-93, 440, 445
on inborn 290
insecure attachment in, 7, 394-408
traits,

infant crying research, 169-70

mourning as, 384-85
on language use/attachment status, 369-70
secure and anxious attachment in, 372,
learning theory, 1 69
398-401,445
on loss of leisure time, 415
security and exploration in, 387
Adult Attachment Interview, 363-66, 368-69,
on maternal attunement, 346, 349, 350
mission, 440
383, 387, 430
personality of, 131, 433
Affectionless children, 50-57, 100
predictive powers of attachment classifica-
Agee, James, 441
tions, 252-53
Aggression, 37, 39, 42-44, 49, 57, 61, 82, 90,
Strange Situation development, 143-50
112, 172, 180, 185, 188, 194, 196,214,
 7

488 Index

students 210-11, 23^37, 264, 370-71,435

of, emotional disturbance and, 390-91
viewpoint/methodology, support for, 162, as predictive of adult behavior, 6-7

163, 164-65, 188-89, 429-36 typical patterns of, 443-45

Alhersheim, Leah, 179-80, 182 Anxiously attached children, 171, 356
Alford, John, 32-33, 34, 50 adult attachment status, 366, 367
Ambivalence, 47-48, 375 at age six, 218-19
Ambivalent attachment change in classification, 254
adult mourning and, 383 cognitive capabilities, 207
adult romantic experience and, 386-88 day care and, 316, 317-18, 322
in children. See Ambivalently attached chil- degrees of, 224n.
dren feelings for others, 185
exploratory behavior and, 387 future difficulties of, 252-53
loneliness and, 383 gender-related behavior differences,
personality disorders and, 389-93 195-97
shame and, 397, 402-3, 404 German 262-63
study,
typical patterns of, 443-45 internal models of, 205
Ambivalently attached children, 47, 161 irritable children study, 303-5
adult attachment status, 445 manipulatory behavior, 399
angry resistance of, 246, 459n.5 (ch.l7) in middle-classhomes, 7
at age six, 218, 445 Muhamidi, 136, 140
attempts to influence mother of, 220 negative emotions and, 243
as babies, 150, 153-54, 156 peer relationships, 193-94
behavior patterns, 172-73, 195-96 predicting attachment style of, 301-2
157-58
categories, replication of parents, 376-77
documentation of, 300-1 response to parental flaws, 405-6
Donnie, 151-53 shame and, 238-51
feelings for others of, 185 strategies/defenses of, 220-37
friendships 195-96, 445
of, see also Ambivalently attached children;
gender-related behavior differences, 196 Avoidantly attached children
245
inferiority feelings of, Approach or withdrawal, as temperamental
parental frustration and, 253 characteristic, 274
reaction to change, 227-28 Arend, Richard, 179
speech/intemal working model of, 212-13, Arsenian, Jean, 147
214 Assault on Truth, The (Masson), 431
Suzie, 150, 153 The
Atlantic, (publication), 251, 318
teachers’ attitudes toward, 188, 445 Attachment
types of, 185-86 in adulthood, 363-69, 379-93, 439-40
American Academy of Pediatrics, 322 assessment. See Strange Situation
American Journal of Orthopsychiatry, 271 behavioral view of, 169
American Journal of Psychiatry, 1 behavior patterns, 135-36, 145-46, 172-73
American Psychological Association, 119, 122 hreast feeding and, 121-22, 134, 140
American Psychologist, The, 328 classification changes, 253-56
Amersham General Hospital, 81 dependency 170, 460n.l4
vs.,

Amy (case study), 191-92 exploration complementarity with, 203

Andry, R. G., 1 17 five phases of, 1 37
Angel Guardian Home, 18, 19 inherited temperament and, 299-301
Anger Minnesota studies, 177-89, 201
of ambivalent children, 246, 459n.5 (ch. 17) parental quality of care and, 304
attachment types and, 242 patterns, tenacity of, 226-33
holding on to, 406 predictions, 183-84
maternal, 356 quality of, 165, 246-47, 304, 356-57
Anna (case study), 13-14 scholarly research interest in, 201
Anna Freud Center, 430 speech modes and, 369-71
Anorexia, 223-24 styles. See Ambivalent attachment; Anxious
Anxiety, maternal, 356 attachment; Avoidant attachment; Secure
Anxious attachment, 155-56, 179n., 182 attachment
adult mourning and, 384-85 style transmission, 361-69
in adult relationship, 398-401 theory. See Attachment theory
in children. See Anxiously attached children Attachment and Loss (Btiwlby), 159, 160
cultural roots of, 41 1-24 “Attachment as an Organizational Construct”
Darwin as example of, 427 (Sroufe and Waters), 163
 5

Index 489

Attachment theory, 3, 5-9 Behavior genetics, 291, 296-98

criticisms of, 248-66 Behaviorism, 4, 7, 120, 166, 171-72
formulation, 87-102 and babies’ stimulatory needs, 89-90
fundamental message, 295-96, 297 empirical emphasis, 167-68, 215
preschooler study, 183-89 reinforcement theory, 5, 157, 169-70, 270
reactions to, 103-18, 286-87 Bell, Richard, 290
support for, 1 19-25 Bellevue Hospital, 17-20, 27
Attention span, as temperamental characteris- Belsky, Jay, 8, 301-2, 350, 374
tic, 275 day-care controversy, 317-18, 320-32,
Attunement, mother-infant. See Maternal 334-37
attunement Bender, Loretta, 8, 17, 18, 20, 50, 53, 60, 271
Australian health care system, 421-22 Benedek, Therese, 93
Avoidant attachment, 151-54, 156, 439-40 Bereavement. See Mourning
adult mourning and, 406 Berkeley Adult Attachment Interview, 363-69
adult romantic experience and, 386 Better Baby Institute, 416
Bowlhy’s personality exemplifying, 428 Betty (case study), 53
in children. See Avoidantly attached children Bielefeld (Germany) attachment study, 259-64
exploratory behavior and, 387 Biological determinism
loneliness and, 383 temperament debate, 289-312
narcissistic personality disorder and, 390-91 see also Heredity
repression and, 430 Bion, Wilfred, 79, 430
typical patterns of, 443^5 Birch, Herbert, 274, 277, 285, 293
Avoidantly attached children, 161 Blatz, William, 129-30, 132, 136, 171, 434
adult attachment status, 366, 380, 445 Blehar, Mary, 316
anger of,459n.5 (ch. 17) Block, Jeanne, 179
at age six, 218, 445 Bonding, 89, 90, 106; see also Imprinting

behavior patterns, 172-73,310-11 Borderline personality, 392, 466n.22

categories,157-58 Bouchard, Thomas, 291-92, 295-96
Clare, 233-34 Bowlby, John (Edward John Mostyn), 5,25,
day care and, 322, 329 37-38, 39, 71, 72, 73, 78. 79, 86, 145, 154,
defenses of, 245 237,255, 261,286,350,356,417
denial of attachment needs by, 221-25 on adult mourning, 384
dependency and, 186-87 Ainsworth’s influence on, 143, 158-59, 160,
distress and, 232-33 434
gender-related behavior differences, 195-96 on ambivalence, 40-41
German 259-61, 262-64
study, on attachment/self-esteem link, 238
interpretations concerning, 258-61 attachment theory formulation, 87-102
Juan, 245-46 on avoidantly attached children, 353
low self-esteem of, 240 background and research, 27-34
maternal attunement and, 352-53, 354-55 caricature of philosophy of, 270-71
commitment to psychoanalysis, 1 13
negative emotions of, 241—42
parental frustration and, 253 concern for child’s well-being, 347—348
peer relationships, 193, 194, 195, 444-45 criticisms of, 107, 109, 114, 250-51,435-36

reaction to change, 227-28 Darwin biography, 426-27

shame and, 246 day-care controversy, 318, 319-20, 337-39

speech/intemal working model of, 212-13 death of, 426

teachers’ attitudes toward, 186-87, 188, 445 defenders, 1 1

temperament of, 300 on difficult vs. easy baby, 297-98

types of, 186 on father’s role, 105
on inborn traits, 289, 345
Babies. See Infant
on infant’s need for affection, 5
influence 89-90, 106, 112, 118, 121, 122,
of,
“Baby Jessica” custody case, 419
Bakwin, Harry, 19-20, 27, 60 419-20, 428-29, 431, 432-33, 434
internal working model, 203, 204-9, 249
Balint, Michael. 38, 93
Baltimore, mother-infant study, 143-61 Klein relationship, 39, 44-45, 46, 55, 56,
108, 109, 114, 158
Banished Knowledge (Miller), 406
341^2 learning theory and, 169
Barglow, Peter, 331,
loss of attachment 415figures,
Barnard College, 242, 270
misreadings of theories, 115-17
Basch, Michael, 181n., 241, 437
Beebe, Beatrice, 347
on parental behavior, 49—52, 103—5, 361—62,
371
Behavioral and Brain Sciences (journal), 265
 1 4

490 Index

Bowlby, John (com.) Child care. SeeDay care; Nannies

personality of, 29-30, 131, 427-28 Child Care and the Growth of Love (Bowlby),
on primitive defenses development, 48-49 66
on rejection sources, 263-64 Child custody cases, 255
on repetition compulsion, 395-96 Child Guidance Center (London), 50
on self-reliance and secure attachment, 177, Childhood (TV program), 344
381-83 Child psychiatry, 27, 71, 72
on source of school phobia, 221 Children
view of women, 3 1 attachment and relating styles, 182, 183-89;
WHO report, 59-66, 68, 414-15, 419-20, see also Ambivalently attached children;

432 Anxiously attached children; Avoidantly

Bowlby, Richard, 29 attached children; Securely attached chil-
Bowlby, Tony, 3 dren
Bowlby, Ursula, 30, 31, 33, 88, 426, 432, day-care. See Day care
454n.6 death of parent, 366n.
Boys. See Gender of depressed mother, 352
Brazelton, T. Berry, 248, 285, 303, 334, 343, difficult vs. easy, 270-88, 298

347, 422, 436 environmental influence. See Environment

Breast fantasies of, 36, 37, 41, 45, 46, 50-51, 56, 57,
affectional ties/nursing link, 120, 121-22, 91
143 hostility toward parents, 55
infant’s relationship with, 41^3, 94-95, 120, institutionalized, 15-25, 59-65
121-22, 140 internal models of young, 205
instinctive sucking, 97 irritable, study of, 303-5

Ugandan weaning, 133-35, 137-38 model of relating, 219

Bretherton, Inge, 162, 215, 265, 266, 433, parental suicide, reaction to, 206
435 resistance to recognizing suffering of, 22
British Journal of Psychiatric Social Work, The, sexuality of, 200, 437, 439
85 siblings, 293, 307
British Psycho-Analytic Society, 26, 33, 41, 45, temperament, 275
108, 109, 114,371 see also Adolescence; Infant

Brody, Sylvia, 144 Children of the Great Dep>ression (Elder), 422

Bronfenhrenner, Urie, 308, 321, 324, 327, 332, Children’s Hospital (Boston), 347
340, 422, 423 Chomsky, Noam, 169
Brown University, 146n. Cicchetti, Dante, 163, 188
Bruch, Hilde, 271 City University of New York, 333
Bruner, Jerome, 465n.6 Clare (case study), 233-34
Biihler, Charlotte, 23, 92, 132 Clarke-Stewart, Alison, 328, 329
Burlingham, Dorothy, 61, 62, 73, 91, 101 day-care controversy, 339, 340
Buss, Arnold, 293 Clinging, 90, 96-98, 124, 169-70
Byng-Hall, John, 200, 428 Clinton administration, 334
Colorado Adoption Project, 293, 462n.7
Cambridge University, 31, 196, 429 Comings, David, 306
Can I Leave My Bahy’ (Bowlby), 103-4 Committee on the Welfare of Children in
Casler, Lawrence, 1 18 Hospital, 82
Cassidy, Jude, 328, 370, 435 Commonwealth Fund, 33
day-care controversy, 326 “Companion in the Bird’s World, The”
Center for the Advanced Study in the (Lorenz), 87
Behavioral Sciences, 122 Compulsive personality, 391, 392, 392n.
Central Middlesex Hospital (London), 72, 74, Conditioning, 168-69
76, 78-79 Connecticut family support program, 421
Character, as mirror of parents, 1 Conthiuum Concept, The (Liedloff), 416
Charles Darwin (Bowlby), 427 Cooperation and Interference scale, 164
C.;hess, Stella, 293, 299, 312 Cornell University Medical Center, 347
day-care controversy, 325 Cortisol level assessment, 302
difficult children hypothesis, 269-88 Crittenden, Patricia, 234, 460n. 23
Child abuse, 22, 234-36 Crying
cycle of, 405 parent attachment and infant, 96, 144-45,
love and aggression in, 235-37 168, 169-70
prevalence of, 22 reinforcement theory, 7, 157, 169, 270
 1

Index 491

Culture Durenherger, David, 334

anxious attachment roots in, 41 1-24 Dyad, 217
attachment classifications and, 259-60, 261,
264 Eagle, Morris, 430, 437
influence on temperament, 284 East African Institute of Social Research,
Custody cases,419 133
Easy children, 276, 282, 297
Darwin, Charles, 88, 89, 426-27 Eddie (case study), 228-29
Bowlhy biography of, 426-27 Edelston, Harry, 68
Davidson, Richard, 293 Egeland, Byron, 183-84
Day care Ego, 39
controversy concerning, 8, 3 1 3-44 Elder, Glen, 422
lack of U.S. system for, 421 Elicker, James, 192, 195

maternal attunement and, 350 Eliot, T. S., 295

Death Emde, Robert, 117-18,422-23
infant asylum rates, 19 Emerson, Peggy, 133
maternal, 28 Emotional Life of the Toddler, The (Lieberman),
parental suicide, 206 439
see also Mourning Emotions, negative, 49-50, 241—43
Death in the Family, A (Agee), 441 Empathy, of secure children, 185, 195
Death 43-44
instinct, 41, Englund, Michelle, 192, 195
Delusional projection, 48 Environment
Denial of reality, 48 factors in early childhood, 3, 27-28
Dependency infant’s struggle to control, 203
anxious attachment and, 186 influence on hehavior/personality, 167, 290,

attachment vs., 170, 460n.I4 293-94

as irrelevant to attachment, 101-2 influence on intelligence, 21-22
secure attachment and, 251 interaction oftemperament and, 284-85
self-reliance and mutual, 381-82 interplay between heredity and, 7, 28, 258,

as undesirable trait, 413 273,274, 286,313

Depression Eriksen, C. W., 118

genetic basis for, 293 Erikson, Erik, 38, 92, 166, 179

maternal, 352 Escalona, Sihylle, 21

“Depressive position,” child’s, 44 Ethology, 87-90, 122, 132, 145, 259

Deprivation. See Maternal deprivation Bowlhy’s influence on, 89-90, 106, 112, 118,
Derek B., 52-54, 154 428-29
Eugenics movement, 4, 16, 273
Detached children. See Affectionless children
Every Child’s Birthright (Fraiberg), 333
Developmental psychology, 167-73, 291, 306
Ainsworth’s influence on, 436 Exploratory behavior, 203, 387

behavior genetics and, 291

biological determinism in, 306 Fairhaim, Nance, 34

interest in Piaget, 208 Fairhairn, Ronald, 39, 90, 93, 396

poomess'of'fit debate, 287 Faithfull, Theodore, 32

False self, 94, 352
and psychoanalysis, 166-67
Family life, historical changes in, 41 1-14; see
quality of attachment and, 247
also Parents
Difficult children, 269-88, 276-79, 284, 297,
303-5, 461n. 28 Family systems theorists, 38
Annie, 370 Fantasy, 112, 188
Dillard,
infants’ and children’s, 36, 37, 41, 42—44, 45,
Dillinger, John, 54
Dismissive adult, 365-66, 373-74, 445
46,51,56,57,91
reality as cornerstone of, 1 3
Disorganized/disoriented category, 214, 218-19,
406n.22 Father
temperamental characteristic, child’s attachment to, 198-200
Distractihility, as
Freud (Sigmund) fixation on, 91
275
jealousy of, 362-63
Donnie (case study), 151-53, 456n. 13
relationship with child, 8, 51
Double hind communication, 309-10
role of, 105,296-97, 315, 335
Drama of the Gifted Child, The (Miller), 406
Females. See Gender
Drive theory, 39, 90, 95
Feminism
Duckling imprinting, 87
criticism of attachment theory, 8
Dunn, Judy, 293
 1

492 Index

Feminism (com.) Grief: A Peril in Infancy (film), 23-35, 65

day-care controversy, 319 Grodin, Martin and Georgette (fictional char-
influence of,288 acters), 398-401, 403^

Fenichel, Emily, 324 Grosskurth, Phyllis, 44

Ferenczi, Sandor, 93 Grossmann, Karin, 242, 259, 261-62, 300, 302,
Field, Tiffany, 325 355,418-19, 466n.28
First Year of Life, The (Spitz), 25 Grossmann, Klaus, 259-60, 263—4, 300, 302,
Flexible, Fearful, or Feisty (Lieherman), 297 388,418
Fonagy, Peter, 366, 372-73, 431 Guatemala, birth study, 420
“Forty-four Juvenile Thieves: Their Characters
and Home-Life” (Bowlby), 57-58 Haft, Wendy, 354
Foster care, 60, 84, 85, 255, 423 Hamilton, Victoria, 29
Fraiherg, Selma, 235-36, 248, 333, 363, 405 Handbook of Child Development, 329
Freud, Anna, 41, 45, 99, 166 Hargreaves, Ronald, 59, 123
on mother-infant relationship, 61, 62-63, 86, Harlow, Harry, 119-22, 124-25, 147, 204
89, 91-92, 93 Harlow, Margaret, 124
personality of, 108 428
Harris, Mattie,

reaction to Bowlby ’s theories, 113, 114 Hartmann, Heinz, 38

Freud, Sigmund, 40, 101, 166, 414 Hatred, 37,40-49, 56, 112, 160rr. , 205, 222,
on conflicting wishes, 40 236, 239-41, 245, 361, 394, 402, 404, 41
criticism of, 432 Hazan, Cindy, 385-88
on death instinct, 41, 44, 112 Heimann, Paula, 108
drive theory, 39, 90, 95, 112 Heinicke, Christoph, 99, 101, 159
on hysteria, 101, 158, 432 Herbert, Barbara, 291
on infantile sexuality, 33-36 Heredity, 2-3
and infant-mother bond, 90 and destiny, 272, 274
psychoanalytic technique, 394, 414 eugenics movement, 273 4,

on repetition compulsion, 395 influence con adaptation, 254

seduction theory, 287, 432 interplay between environment and, 7, 27,

trauma theory, 26-27, 35-36 254, 273,274, 286-94,313

Fromm, Erich, 38 temperament debate, 289-312
Fry, Elizabeth, 66 Herman, Imre, 92
Fry, Margery, 66 Hinde, Robert, 122, 124, 260, 433
Hippcocrates, 273
Galinsky, Ellen, 334, 336 Holden, Hyla, 427
Gabon, Francis, 273 Homelessness study, 59-66
Gender Hopkins, Juliet, 108, 233-4, 239, 453n.28
attachment differences, 196-97 Homey, Karen, 38
emotional systems, 377 Hospices, 424
oedipal period, 36 Hospitalism, 19-20
personality disorder differences, 392n. Flospital stays, effects on children, 16, 20,
secure/insecure differences, 197 49-50, 52-53,67-82, 105, 120
Genetics. See Behavior genetics; Heredity Hotoph, Norman, 87
George, Carole, 211, 363 Howes, Carolee, 322
Georgette (fictional character), 398—401, 403-4 “Humiliated fury,” 246
Germany, attachment study, 261-64 Huxley, Julian, 88
Gesell, Arnold, 4, 5, 21, 168-9 Hysteria, Freudian, 101, 158, 391n., 431
Girls. See Gender Hysterical character, 35-36, 391n., 392, 393
Global ratings, 164
Goal-corrected partnership, 204 Id, 3Q
Going to Hospital with Mother (film), 81-82 Identical twins. See Twins
Goldfarb, William, 20, 59-60 Identification, 63, 109, 111, 115, 199, 208, 224,
Goldwyn, Ruth, 211, 219, 363 226, 377,382, 385,388, 407,437
“Good-enough” mother, 356 Imprinting, 87-88
Goodship, Daphne, 291 Infancy in Uganda (Ainsworth), 141-2, 143
Gosling imprinting, 88 Infant
Grandiosity, 225, 246, 356, 390, 393 affection needs, 3—4, 18-23
Great Britain, psychoanalysis in, 108-9 breast relationship, 41-43, 95, 120, 121-22,
Greenberg, jay, 287-8 140
Greenhurg, Mark, 264 crying and parent attachment, 5, 96, 144-45,
Greenspan, Stanley, 422-3 157, 169-70, 270
 1

Index 493

emotional life of, 345-57 on temperament formation, 294-96, 300,

familiarity with mother’s scent, 40-41, 345 301
fantasy life, 41-44 Kaplan, Nancy, 363, 431
as genetic construction, 4 Karen, Raphael, 342
-mother bond, 89, 90, 91, 106; see also Karen, Thaleia, 342
Attachment theory Kate (case study), 83
-mother relationship study, 132-42 Kennell, John, 453n.8
mourning capabilities, 100 Kenya, difficult babies in, 284
search for mother’s face, 121 Kernberg, Otto, 1 10, 430
separation. See Maternal deprivation; Keynes, John Maynard, 295
Separation; Strange Situation Kibbutz, children in, 319n., 462n.l 1

sexuality, 35 Kibuka, Katie, 134, 135, 137

struggle to control environment, 203 Kierkegaard, S0ren, 345
temperament, 296, 297-98 Klaus, Marshall, 453n.8
“Infant Day-Care: Maligned or Malignant?” Klein, Melanie, 33, 38, 57-58, 92-93, 94, 100,
(Clarke-Stewart), 328 158, 166
Inferiority feelings. See Self-esteem Bowlby relationship, 39, 44, 46, 55-56, 57,
Inherited qualities. See Heredity 107-8, 109, 114, 158
Insecure attachment, 140, 155n. personality of, 44
in children. See Anxiously attached theories, 40-46
children Klopfer, Bruno, 130
passing on, 361-78 Kobak, Roger, 241,393,435,436
working through, 394-408 Kohut, Heinz, 160, 346, 390, 414, 465n.5,
see also Anxious attachment 466n.20
Instincts, 88-91, 95-96 Konner, Melvin, 418
Institute for Child Guidance (New York City), Kiibler-Ross, Elisabeth, 384

13, 17
Institute of Child Development (Minnesota), Lamb, Michael, 198, 264-65, 320
166 Lancet, The (journal), 68, 85, 431
Institutions, effect on children, 16-25, 60 T Ppfpr 41 1

Intelligence, environmental effects on, 21, 22 Laura (case study), 76-78, 154, 239, 452n.25
Intensity of reaction, as temperamental charac- Leach, Penelope, 248, 261
275
teristic, League of Nations survey, 55
Internal working model, 202-9, 249, 370 Learning theory, 167-72, 203
speech manifestation of, 215 “Learning to Love” (Harlow), 124-25
International Journal of Psycho-Analysis, 57, 108 Leisure time, loss of, 415
Intrapsychic processes, 107-8 Lester, Barry, 146n.

Iowa Child Research Welfare Station, 2 Levy, David, 13-14, 16-17, 28, 50, 59-60, 64
IQ. See Intelligence Lewis, Helen Block, 246
Irritable children, study of, 303-5 Lewis, Jim, 291-92
Isaacs,Susanna, 107, 109 Libido, 39, 90
Isherwood, Christopher, 250n. Lieberman, Alicia, 142, 319, 404, 435, 439
Liedloff, Jean, 416-19
James, William, 346 Little Albert (case study), 168
Jane (case studies), 23-25, 83, 277-78 Locke, John, 5
London Child Guidance Clinic, 33
Japan, attachment studies, 300-1, 418
362
Jealousy, father’s, Loneliness, lack of secure base and, 383

Jensen, Arthur, 307 “Loneliness in Infants” (Bakwin), 19

Jerry (case study), 281 Longitudinal studies, 132-33, 144, 256, 459n.6,
Jewish Family Services, 20 462n.7
developmental psychology tool, 167
Jewish Foundling Homes, 17 as

John (film), 84-85 Main’s, 21 1-19

Johns Hopkins University, 6, 143, 148, 162 New York, 274-84, 290

Jones, Ernest, 45 of personality, 294

Joseph P. Kennedy Award, 22 Lorenz, Konrad, 70, 87, 88, 132

Journal of Child Psychotherapy The, 85 ,

Loss (Bowlby), 110, 209

Juan (case study), 245-46 Louisville Twin Study, 462n.7

Love, 1-3, 7, 9, 14, 15, 36, 41-49, 55-58, 76,
Kagan, Jerome, 326 84, 89, 90, 91, 93, 98, 100, 102, 105-6,

day-care controversy, 315-16, 325-26, 112-13, 120-22, 144, 146, 170-71, 189,
328-29, 339-40 200, 205-9, 214, 216, 222-32, 239, 243-46,
 3 ,

494 Index

Love (cone.) Medicine at the Crossroads (TV program), 306

250-51, 257, 264, 289, 295, 308-9, 312, Memory barrier, 2

340, 343, 361-66, 374-82, 385-87, Men. See Gender

396-98, 404-5, 408, 41 1-12, 419, 427, 441 Mental hygiene movement, 27
Love, maternal. See Mother-love Mercer, Edith, 1 30
“Love quiz,” 385-86 Metapsychology, 1 1

Lowden, Molly, 34 Meyer, Adolph, 27

Lucy (case study), 83-84 Miller, Alice, 406
Milner, Marion, 108
MacCarthy, Dermod, 81 Minnesota studies, 177-89, 197-98, 253, 300,
Macdonald, Dwight, 441 321
Magic Years, The (Fraiberg), 235 Minuchin, Salvatore, 458n.30
Mahler, Margaret, 180-81, 204, 438 Mischel, Walter, 307-9
Main, Mary, 163, 198, 395, 435 Mitchell, Stephen, 287-88
on adult attachments, 387, 401-2 Molly (case study), 351-52
on attachment style transmission, 363-69, Monkey research, 1 19-25, 252, 255, 298
371,372 Monotropy, 99
background and research, 210-19 Mciod quality, as temperamental characteristic,
classification system, 388 275
on speech modes and attachment, 369-71 Morris, Ivy, 81
Makerere College (Uganda), 132 Mortality rates, infant asylums, 19-20
Malan, David, 115 Mother
“Mai de Mere” (Chess), 271 of ambivalently attached child, 374-76
Males. See Gender of anxiously attached child, 155-56, 182
Manipulative behavior, 399 attunement to child. See Maternal attune-
Marriage ment
effect of, on children, 200 of avoidantly attached child, 156, 213, 374
as secure base, 404-5 as blame target, 296-97
Marris, Peter, 99, 433 -child relationship, 2, 3, 6, 97-98, 132—42,

Martin (fictional character), 398-402 180-83, 354-55

Martin, Freda, 338, 428 day-care controversy, 3 1 3-44
Marvin, Cheri, 433 death of, 27
Marvin, Robert, 200, 204, 241, 264, 433, deep-seated psychology of, 304
458n.30 depressed, 352
Mary (case study), 75 emotional attitude of, 28-29, 356-57
Mason, Edward, 68, 73 “good-enough,” 356
Masson, Jeffrey, 431 guilt about difficult child, 271

Masturbation, 34, 45 hatred for one’s, 56

Maternal attunement, 190, 345-57 -infant bond, 89, 90, 106, 453n.8; see also

Maternal Care and Mental Health (Bowlby), 60 Attachment theory

Maternal deprivation, 14-25, 34, 60-66, 155 influence on child’s temperament, 279
Anna, 1 3-14 overprotect ive. See Maternal overprotection
attachment predictions and, 252 personality and child’s attachment status,
Bobby, 61 302
criticism of studies, 116-18 response to crying infant, 7, 96, 144-45, 170

Derek B., 52-53 role of, 105-6, 199

dimensions of, 123 as secure base, 1 16

effects of, 49-58 of securely attached children, 141, 155, 156,

partial forms, assessment of, 155 157, 182
Patrick, 61-62 separation from, 23-25, 28, 29, 50, 53-58
psychoanalysis and, 34 social support for, 420
Reggie, 61 working, 3 1 3-15, 319, 330, 333, 341
separation and, 1 34 46 3n. 12
studies, 13-25 Maternal deprivation; Mother-love
see also

WHO report, 59-66 Mother Care/Other Care (Scarr), 314

see also Hospital stays; Separation “Mother killers,” 276
Maternal overprotection, 16, 272-73 Mother-love, 42, 66, 18, 120-22 1

Max Planck Institute for Psychiatry, 347 icon status of, 250
McClintock, Barbara, 158 “primary maternal preoccupation," 356
McIntosh, Millicent, 270-71 and promotion of well-being, 14-15
Mead, Margaret, 123, 141 see also Maternal deprivation
 Index 495

Mourning Parents
in adulthood, 384-85, 406 and anxiously attached children, 229-32,
bereavement counseling, 424 253, 378
impediments to, 406 attachment style transmission, 361-78
in infancy/childhood, 99-100 causes of nonattachment, 1 39
Muhamidi (case study), 139-41, 467n.25 character as mirror of, 1

Multi-Site Day Care Consortium, 331 death of, 366n.

of difficult children, 277-79
Nannies, 104, 319-20, 331, 341-42 effects on children, 26-28, 49-52, 177-89
Narcissism, 160, 258 emotional problems of, 35, 242
Narcissistic disturbances, 390-91, 392-93, flaws of, 405-6
465n.5, 466n. 20 hatred for one’s, 56
Nash, Ogden, 107 impact of children on, 285
National Association tor the Welfare of poorness-of-fit, 183, 287, 297
Children in Hospital, 85 reaction to slow-to-warm-up child, 281
National Association of Mental Health, 103 secure attachment to both, 198
National Institute of Child Health and Human social support for, 421
Development, 264, 325, 331 sources of advice 269-71
for,

Nature of the Child, The (Kagan), 248, 260 suicide of, 206—7
“Nature of the Child’s Tie to His Mother, The” unmet needs of, 243
(Bowlby),95, 114, 208 unreflective, 373-74
Nature-nurture controversy, 7, 313; see also unresolved conflicts of, 34
Environment; Heredity see also Father; Mother
Negative attention-seeking, 142 Parkes, Colin, 384, 433
Negative emotions, securely attached children Parry, L. A., 68
and, 241 Passive aggression, 48, 399
Neglect, 239 Patterns of Attachment (Ainsworth), 161

Neill,A. S., 32,34 Pavlov, Ivan, 167

Neonatal Behavioral Assessment Scale, 303 Pawl, jeree, 142
Neubauer, Peter, 290, 292, 309-10 Peer relations, 192-98
Neurosis, etiology of, 35, 37 Pennsylvania State University, 8, 301, 321
New York Academy of Medicine, 23 Permissiveness, 270
New York Longitudinal Study, 274-85 Persistence, as temperamental characteristic,
New York State Psychological Association, 275
117 Personality
New York Times, 325, 336, 421 development/parenting style, 177-89
NIMH. See National Institute for Mental genetic component, 293
Health inborn traits, 272, 273, 462n.28
Nonattachment, 138 influence of environment on, 290, 293-94
Nursing (lactation). See Breast shaping and real-life events, 171
Nursing Times, The, 68 and temperament, 284
NYLS. See New York Longitudinal Study Personality disorders, 389-93
Phillips, Deborah, 326
Object relations theory, 38-39, 46 Piaget, jean, 132, 171, 203-4, 205, 208
Obbmov (Goncharov), 412 Pine, Fred, 430
Obsessive-compulsive disorder, 39 In. Pinneau, Samuel, 1 17

Oedipus complex, 35-39, 40-41 Plato, 273

On Death and Dying (Kubler-Ross), 384 Play, symbolic, 180
Operant conditioning, 168 Plomin, Robert, 293, 312
Oral stage, 38 Poorness-of-fit concept, 283, 285-88, 297

Chigin of the Species, The (Darwin), 88 Poverty, attachment and, 183-89, 197-98, 199,

Osofsky, Joy, 422-23 221, 316, 422

Overprotection. See Maternal overprotection Preindustrial society, attachments in, 41 1-12
Preoccupied adult, 445
Paddington Green Children’s Hospital, 94 Preschoolers, attachment study, 183-89

Pammy (case study), 282 Primary affect hunger, 17

Pancake, Van, 193 Primary maternal preoccupation, 354-55
Papousek, Hanus and Mechthild, 347 Primitive defenses, 48-49

Papua New Guinea, 418 Prison nurseries, 24-25

Parental leave law, 333, 421 Progressive education, 31-33

Parentified, 221 Prostitutes, affectionless childhoods of, 55

 1 1

496 Index

Prugh, Dane, 123 Rycroft, Charles, 35-36, 94, 109, 431, 450n.31
Psychiatry, biological determinism in, 306; see
also Child psychiatry Sagi, A., 467n. 1 1

Psychoanalysis Sally (case study), 81-82

Bowlhy’s commitment to, 1 1 3 Salter, Mary Dinsworth. See Ainsworth, Mary
contributing to guilt of parents of difficult Sam 354
(case study), 353,
children, 279 Sameroff, Arnold, 422-23
and developmental psychology, 166-67 Sander, Louis, 144, 348
fantasy life of children, 37 San Francisco General Hospital, Infant-Parent
Freudian technique, 394, 414 Program, 236
inGreat Britain, 108-9 Sarah Lawrence College, 21
maternal deprivation and, 34 Scaffolding, 265n.6
parental neglect and, 428-29 Scarr, Sandra, 306-7, 325, 335
“Psychoanalysis and Child Care” (Bovvlhy), day-care controversy, 339, 356
371-72 Schaffer, Rudolph, 101, 133, 159
Psychoanalytic Dialogues (journal), 431 Schemata, 203
Psychoanalytic Quarterly, The, 428-30 Schenk-Danziger, Lotte, 23
Psycho- Analytic Society (British). See British Schizoid personality, 309, 391, 463n.61
Psycho-Analytic Society Schmideberg, Melitta, 44
Psychoanalytic Study of the Child (annual), Schoenewolf, Gerald, 288
113 School phobia, source of, 221
Psychopathetic personality disorder, 53 Schur, Max, 1 13-14
Psychopathology, twin studies, 309-10 Secure attachment, 6, 160, 350
Psychotherapy, 402-3 adult attachment status, 381-82, 385-84,
401-2,445
Ralph (case study), 280-81 in children. See Securely attached children
Ramakrishna and His Disciples (Isherwood), future well-being and, 188
250rv. modern harriers to, 414
Rank, Otto, 100 self-reliance and, 177
Rapprochement phase, 181, 438 typical patterns of, 444-45
Rayner, Eric, 432 Secure base, 136, 141, 146, 171, 383
Reality lack of, 383
as cornerstone of fantasy, 107 Secure Base, A (Bowlby), 434
Reich, Wilhelm, 34 Securely attached children, 172, 179n., 407
Reinforcement theory, 5, 157, 169-70, 270 Amy, 191-92, 195
Rejection, 239 at age six, 218, 445
Relatedness assessment. See Strange Situation behavior patterns, 172-73
Repetition compulsion, 395-96 change in classification, 253-54
Repression, 48, 1 1 cognitive capabilities, 207
Resistant babies, 1 50 dependency and, 251
Rhesus monkeys. See Monkeys follow-up study, 180, 182, 183
Rhythmicity, as temperamental characteristic, gender-related behavior differences, 197
274 German study, 262
Rihhle, Margaret, 93 internal models of, 205
Ricciuti, Henry, 320 negative emotions and, 241
Rioch, Janet McKenzie, 207 peer relations, 193, 194, 195-96, 444, 445
Riviere, Joan, 33,41, 108, 114, 450n.29. problems of, 405
450n.3i self-worth feelings, 240
Robertson, James, 72-86, 99, 101, 113, 159, social skills, 185

214, 384,433 speech/internal working model, 212

background and research, 72-80 teachers’ treatment of, 445
films, 82-83 Security theory, 129-30, 136
Robertson, Joyce, 72n., 73, 82-86 Seduction theory, 287, 431
Rocky Mountain News, “love quiz,” 385 Segal, Hanna, 107, 109
Romantic love, 385-86 ’Self-esteem, attachment and, 238, 240, 245
Rorschach ink blot test, 130, 433 Self-reliance
Rfisenhlatt, Jay S., 453n. /4 mutual dependence and, 381-82
Royal Hospital for Sick Children, 80 secure attachment and, 177
Royal Society of Medicine, 78 “Self-Reliance and Some Conditions That
Rubenstein, Judith, 322 Promote It” (Bowlhy), 381
Rutter, Michael, 265, 318 Senn, Milton, 21, 59
 1

Index 497

Separation on anxiously attached children, 316

anxiety. See Separation anxiety attachment follow-up studies, 178-79, 198,
assessment. See Strange Situation 253-54, 256
infant response to, 1 39 on attachment style transmission, 363
maternal deprivation and, 134 background and research, 163-65
mother-child, 27, 28 on behavior-problem children, 286
stages of emotional reaction to, 74-75 on custody cases, 237
see also Maternal deprivation day-care controversy, 339-40
Separation (Bowlby), 86, 208, 289 on difficult infants, 298
Separation anxiety empirical data, 201
Bowlby theory on, 100-1, 114 on influence of Ainsworth, 435, 436
day-care children, 315-16 on loving parents, 378
onset, 261 preschoolers study, 183-89, 193, 194, 195
“Separation Anxiety" (Bowlby), 115 and spawning of attachment industry, 210
Sexual attachments, earlier attachment figures on teachers’ importance, 228, 423
and, 396 temperament vs. attachment debate,
Sexuality 299-300
of children and infants, 36, 199, 436-37, Stanley (case study), 280
439 Steele, Howard and Miriam, 366-67
drive theory, 39, 90, 95 Steinberg, Larry, 317, 332
Shame Stern, Daniel, 347-56
ambivalent attachment and, 240-41, 245 Stevenson-Hinde, Joan, 196
anxious attachment and, 239 Stevie (case study), 351
formation, 239-43 Stone, Fred, 80-81
Juan, 245—46 Storr, Anthony, 43
Laura, 239 Strange Situation, 4, 148-61, 173, 201, 237,
Shaver, Phillip, 385-88 301,368
Sheldon, William H., 273 for adults, 364
Sheppard 144
Pratt, ambivalent 246
child’s resistance in,

Shirley, Mary, 132 avoidant child’s calm during, 300

Siblings, 293, 307 classification predictions, 367

Skeels, Harold, 21-22 criticism of, 249, 264-65

Skinner, B. E, 168 day-care children and, 32-34

Slade, Arietta, 354, 407 development of, 143-61
Slow-to-warm-up children, 276, 279-81, as measurement of temperament, 299

284 modifications, 317, 458n.i6

Smiling, parent attachment and, 96 patterns of secure and anxious attachment,

Smuts, Alice, 80 444

Snugli, use of, 418 replication of, 163

Society, and child-care issues, 420-24 six-year-olds in, 216-19

Society for Research in Child Development, temperament factor, 285
306,313,327 value to researchers, 172, 173
Solitude, importance of healthy, 382 Stress, attachment and, 305, 390-91, 422

Solnit, Albert, 430 Sublimation, 48

Sommerville, John, 414 Sucking instinct, 97
Speech Suicide, reaction to parental, 206-7

acquisition, 169 Sullivan, Harry Stack, 38, 192, 395

as manifestation of internal working model, Sulloway, Frank, 426

215 Summerhill, 31-32
modes/attachment status, 369-71 Suomi, Stephen, 252, 256, 298
Spence, James, 75-76, 79, 452n.22 Superego, 55,91

Spitz, Rene, 93, 102, 120, 450n.i7 Susie (case studies), 283

criticism of, 117-18 Suttie, Ian, 453n. 14

support 24
for, 1
Suzie (case study), 150, 456n. 13
Symbolic play, 180
Spock, Benjamin, 270
291-92 Syracuse, New York, family support program,
Springer, Jim,
SRCD. See Society for Research in Child 421
Development Szanton, Eleanor, 324, 333-34

Sroufe, Alan, 210, 219, 237, 238, 251, 286,

Tavistock Clinic, 29, 71, 75, 1 14, 233, 239,
294, 363, 375, 423, 457n. 25
337-38, 430
on ambivalently attached children, 375
 1

498 Index

Teachers University of Virginia, 200, 264, 306, 325, 433

ability to reach anxious children, 228 University of Wisconsin, 119, 252, 293
attitudes toward specific attachment types,
186-88, 445 Van den Boom, Dymphna, 303-5, 420
influence on temperament, 284 Vaughn, Brian, 302, 317, 326, 330
training for dealing with anxious children, Vygotsky, Len, 465n.6
423
Teenagers. See Adolescents Wages, child care workers, 334
Temperament Wall Street Journal, 324, 325
characteristics,274-84 Waters, Everett, 162-63, 166, 172, 177, 178,
inborn, 289-312, 462n.i8 257, 258
interaction of environment and, 285 Watson, John B., 5, 167-68, 269
poor fit between parent and child, 282-83, Weaning, in Uganda, 134, 137
287-88, 297 Weiss, Robert, 384
Thieves, Bowlby’s study of, 50-58 Weston, Donna, 198, 211
Thomas (case study), 83, 214 Where’s Mommy: The Robertson Studies of
Thomas, Alexander, 274, 277, 285, 293, 299 Mother'Child Separation (film), 73, 452n.i6
Thomas: Ten Days in Foster Care (film), 214 WHO. See World Health Organization
Thorndike, Edward, 167 Widows and Their Families (Marris), 99, 433
Threshold of responsiveness, as temperamental Winnicott, Clare, 108
characteristic, 275 Winnicott, Donald, 38, 39, 93-95, 351, 373
Tienari, Pekka, 309 false self theory, 39, 94
Tinbergen, Niko, 87, 88, 132 on “good-enough mother,” 356
Toilet training, 26, 276, 285 relationship with Bowlhy, 93-95
Tommy (case study), 278 relationship with Klein, 108
Top Dog/Bottom Dog (Karen), 398 Wolf, Katherine, 144
Transference, 394-96 Women. See Gender
Transitional object, 356 Working Mother magazine, 318
Trauma theory, 26-27, 35-36 Working mothers, 105-6, 313-14, 463n.l
Trohriand Island (Papua New Guinea), 418 World Health Organization, 59-66, 79, 123,
Tronick, Edward, 347^8, 349 419-20
Troy, Michael, 194 Bowlhy ’s report, 59-66, 68, 415, 419-20,
Turner, Patricia, 196, 458n./6 432
Twins follow'up report, 123
studies of identical, 307-1 W. T. Grant Foundation, 251
studies of separated, 290-93, 462n.7
TwO'Year^Old Goes to Hospital, A (film), 76-80 Yale University, 264, 371
Yequana (South America), 416-17, 419
tribe

Uganda, Ainsworth’s work in, 132-42 Yeshiva University (N.Y.C.), 347

Ugliness, anxious attachment and feelings of, “Young Children in an Insecure Situation”
239 (Arsenian), 147
Unconscious distinction, between memory and Young Children in Brief Separation (film), 82-84,
fantasy, 35 85-86
United Nations, 58, 59 Young Children in Hospital (Robertson), 82
University of California (Berkeley), 21 1-19
University of Denver, 385 Zero TO Three (organization), 324, 325, 333,
University of Fraiherg, 259 334, 422
University of Leiden, 303 child-care recommendations, 422
University of Minnesota, 163, 166, 291, 322 Zero to Three (publication), 323, 324, 325
University of Regensburg, 263 Edward, 323, 325, 422-23
Zigler,

University of Toronto, 129 Zone of proximal development, 465n.6

I ^ ;
I

3-
 N« iongef ^ «* **
Boeto*) LMy.
Sale •< tiH« li^wii* ** ^***^
 I’SYCHOI ()C;Y

OXFORD

“A tour dc forcc....BeautifulIy written and a joy to read, it is a must for every

parent and parent-to-be.” — Michael Franz Basch, M.D.,
Professor of Psychiatry, Rush Medical College, Chicago

The struggle to understand the parent-child bond ranks as one of the great quests of modern
psychology, one that touches us deeply because it holds so many clues to how we become who
we are. How are our personalities formed? How do our early struggles with our parents re-

appear in the way we relate to others as adults?

In Becoming Attached, Robert Karen offers fresh insight into some of the most fundamen-
tal issues of emotional life. He explores such questions as; What do children need to feel that

the world is a positive place and that they have value? What are the risks of day care for chil-
dren under one year of age, and what can parents do to manage those risks? What experiences

in infancy will enable a person to develop healthy relationships as an adult?

Becoming Attached is not just a voyage of discovery in child emotional development and its

pertinence to adult life but a voyage of personal discovery as well, for it is impossible to read
this book without reflecting on one’s own life as a child, a parent, and an intimate partner in

love or marriage.

Fascinating reading. ..a considerable achievement. — Contemporary Psychology'

“Robert Karen... is one of our smartest and most accessible guides to the arcane world of psy-

choanalytic theory and research.” •

Elle

“Eloquent, dramatic, and inspiring. .. .A masterful job.” Mary Ainsworth, Ph.D.,

Professor Emeritus, University ofVirginia

Robert Karen is a clinical psychologist in private practice and an award-winning author. He

is Assistant Clinical Profes.sor at the Derner Institute of Advanced l^sychological Studies, Adel-

phi University.

Cover design hy Kathleen i1. Lynch

00
Cover photofjraph © 1996 Danny liriijht / Photonica

Oxford Paperbacks
Oxford University Press
9 780195 1150 17
U.S. S17.95 ISBN 0-19-511501-5