Digitized by the Internet Archive

in 2012

http://archive.org/details/foundationsoffa100hoff
Foundations of

Family Therapy
FOUNDATIONS
OF
FAMILY THERAPY

A Conceptual Framework
for Systems Change

LYNN HOFFMAN

BasicBooks
A Division ofHzrperCoU'msPublisbers
The author is grateful for permission to reprint material from the following sources:

W. Robert Beavers, Psycotherapy and Growth. New York: Brunner/Mazel, 1977. By permission
of the author and the publisher.

E. Wertheim, "Family Unit Therapy and the Science of Typology of Family Systems," Family
Process 12 (1973). By permission of the publisher.

Lynn Hoffman, "Deviation-Amplifying Processes in Natural Groups," in Haley, J., (ed.),

Changing Families. New York: Grune and Stratton, 1971. By permission of the publisher.

Lynn Hoffman, "'Enmeshment' and the Too Richly Cross-Joined System," Family Process 14
(1975). By permission of the publisher.

Lynn Hoffman, "Breaking the Homeostatic Cycle," in Guerin, P., (ed.), Family Therapy: Theory

andPractice. New York: Garner Press, 1976. By permission of the publisher.

Lynn Hoffman, "The Family Life Cycle and Discontinuous Change," in Carter, E., and M.
Orfanides (eds.), The Family Life Cycle. New York: Gardner Press, 1980. By permission of the
publisher.

Library of Congress Cataloging in Publication Data

Hoffman, Lynn.
Foundations of family therapy.

Includes bibliographical references and index.

1. Family psychotherapy. I. Title .

RC488.5.H6 6i6.89'i56 80-6&Q56

ISBN: 0-465-02498-X AACR2

Copyright © 1981 by Basic Books, Inc.

Printed in the United States of America

Designed by Vincent Torre

20 19 18 17 16 15
 To My Family

Martha, Joanna, Livia, and Ted

Contents

Acknowledgments ix

Foreword, by Donald A. Block, m.d. xi

Prologue Behind the Looking Glass 3

Chapter 1 Early Research on Family Groups 16

Chapter 2 The Dynamics of Social Fields 37

Chapter 3 The Second Cybernetics 50

Chapter 4 Typologies of Family Structure 67

Chapter 5 The Concept of Family Paradigms 86

Chapter 6 The Pathological Triad 10

Chapter 7 The Rules of Congruence for Triads 126

Chapter 8 Triads and the Management of Conflict 140

Chapter 9 The Simple Bind and Discontinuous Change 156

vi t
 Contents
Chapter 10 The Thing in the Bushes 176

Chapter 11 Breaking the Symptomatic Cycle 197

Chapter 12 Family Therapy and the Great Originals 219

Chapter 13 Historically Oriented Approaches

to Family Therapy 242

Chapter 14 Ecological Structural and Strategic Approaches 256

Chapter 15 The Systemic Model 284

Chapter 16 Theories About Therapeutic Binds 305

Chapter 17 Issues on the Cutting Edge 323

Epilogue Toward a New Epistemology 339

Notes 351

Index 36s

via
Acknowledgments

A great many people were part of the long journey that resulted in
this book. want to express my appreciation to Don Bloch, Director
I

of the Ackerman who very generously bestowed on me an

Institute,

unofficial grant, by giving me the space and time to complete the

manuscript. At the Ackerman Institute, an atmosphere of collegial
and creative enterprise was provided by Olga Silverstein and Peggy
Papp, of the Brief Therapy Project, and by members of my current
research team, Gillian Walker, Peggy Penn, John Patten, Joel Berg-
man, and Jeffrey Ross. I owe many of my ideas to the stimulating
discussions have had with these important colleagues.
I

Peggy Penn and Carl Bryant read early drafts of the manuscript
and offered sound advice and enormous encouragement, for which
I am most At a later stage, the manuscript was carefully
grateful.
read by Paul Dell and Carlos Sluzki, whose excellent suggestions are
incorporated into the text.
As for colleagues less involved in the final process but whose
intellectual energy and personal support constituted an invaluable
contribution, I thank Mara Selvini Palazzoli and her
would like to
associates in Milan, Giuliana Prata, Luigi Boscolo, and Gianfranco
Cecchin, who have given me unfailing encouragement. I am also
most grateful to other colleagues, who have taught me, inspired me,
and believed in me: among these are the late Don Jackson, Virginia
Satir, Jay Haley, Dick Auerswald, Salvador Minuchin, Harry

Aponte, Carl Whitaker, Monica McGoldrick, Carrell Damman, and

Harry Goolishian.
For understanding and appreciating my work and helping make
it known to colleagues in England and Europe, I am indebted to John

IX
 ACKNOWLEDGMENTS
Byng-Hall, Philippe Caille, Rosalind Draper, and Mia Andersson.
At Basic Books, I want to thank my editor, Jo Ann Miller, who
believed in the project from the beginning and who helped me push
and pull the manuscript into final shape. It is rare to find an editor
who will involve herself so unstintingly and creatively in all the
and polish to a book. Project editor Julia Strand
details that give spit
worked patiently and diligently during every stage, transforming
the manuscript into a finished book. I appreciate, too, the efforts of
Leo Goldberger, who led me to Basic Books.
Finally, I must express my gratitude to the staff and faculty of the
Ackerman me with a respectful yet challeng-
Institute for providing
ing environment in which to work, and for the many helping hands
they held out to me in matters both large and small. To these people,
and to my family, who have put up with me during the whole long
process, and to many others to whom owe much but do not have
I

the space to mention, my thanks.

Lynn Hoffman
February, 1981
Foreword

In thisvolume Lynn Hoffman undertakes an eagerly awaited dis-

play and integration of the theory and technique of family therapy.
Her vision is panoramic; she possesses that relevant and encyclope-
dic fund of information that could have come only from long and
thoughtful observation of the best of her colleagues at work, from
having struggled with teaching and doing family therapy herself,
and finally from having encompassed the now voluminous perti-
nent literature. The attempt is daring; it is aptly titled "founda-
tions."
It is safe to say that this is book
the earliest time at which this
could have been written; equally so, that it has been written none
too soon. We are at the end of the second great cycle of growth in
the field. It is necessary to take stock, pull together the loose threads,
and consolidate the gains that have been made. This volume does
that superbly; it will provide a solid base for the future growth that
is to come.
Attention to family as a clinical entity and as a fruitful area of
theoretical concern developed in a tiny but portentous fashion in
the third decade of this century. Clinical psychiatry during that
period,and the more important years following World War II, was
dominated by psychoanalysis, itself already struggling with revi-
sionistmovements. Psychoanalysts like Sullivan, Horney, Thomp-
son, and Fromm-Reichmann, among others, were enlarging the per-
spectives of their science to include insights from field theory,
linguistics, and cultural anthropology. Thus, as psychoanalytic the-
ory constructed ever more intricate models of intrapsychic se-
quences and structure-functions, "news of a difference" was begin-

XI
 FOREWORD
ning to be heard more insistently. The difference was context first, —
that context, in linear terms, had an effect; later, that context was the
effect. Clinical concern with family followed most naturally, and
the first growth cycle began.
great
As Hoffman makes clear, the systemic evolution in psychother-
apy was contemporaneous with profound shifts in the perspectives
of the natural and the other behavioral sciences. As a single in-
stance, the Society for General Systems Research was organized in

1954 and began publication of its estimable yearbook series two

years later. I have just opened the second volume of that series at
random and come upon Anatol Rappaport's critique of Lewis F.
Richardson's "Mathematical Theory of War." Under the heading
"Etiology of Deadly Quarrels," we find, "The temptation to look for
lineal cause-effect relations, especially where events vital to our
survival are involved, always persists." Indeed it does, and Hoffman
takes precisely this issue as her departure and termination points.
Hoffman begins her integrative effort by jumping right into the
epistemological struggle: "The central concept of the new episte-
mology —both the homeostatic and the evolutionary paradigms— is

the idea of circularity." She lingers over linearity but a moment and
moves along quickly with a description of her personal intellectual
journey. The essence of that journey, it seems to me, lies in the
effort to understand creation, genesis: "I was intrigued by the
. . .

idea that deviation, or deviance, per se, was not the negative thing
it was thought to be, once one abandoned the point of view of those

who wanted it corrected." The associated epistemological riddle of

morphostasis-morphogenesis provides the central thread of Founda-
tions of Family Therapy on which the works of theorists and clinical

innovators of the family therapy field are hung like so many spar-
kling jewels. Bateson is the centerpiece: Hoffman's consideration of
his early work, Naven, and the concept of schismogenesis provides
the foundation for her discussion of "these curious self-reinforcing
cycles" that ultimately are the concern of the family psychother-
apist.

imagine some of the scholars, investigators, and clinical innova-

I

tors whose work is described in this volume may be miffed at the

inevitable abridgement of their work. But look at the extraordinary
array of talent: All the stars of the first magnitude are included, and,

xii
FOREWORD
with rare exceptions, all of the lesser luminaries. Hoffman has paid
attention, is graceful and inclusive. I am impressed with the payoff
of her years of careful observation; she knew who was good and,
even more, who was going to be good.
I want to endorse her choices wholeheartedly and unequivocally.
The work of the authors cited here is the basic reading list in this
field. Serious students might begin right now to educate themselves
by systematically following the path Hoffman has traced out: Bate-
son, Haley, Buckley, Wertheim, Watzlawick, Jackson, Ackerman,
Minuchin, Rabkin, Selvini Palazzoli, Auerswald, Wynne, Whitaker,
Satir, Weakland, Paul —
Hoffman has absolute pitch for the good
—
ones Prigogine, Elkaim, the list goes on and on.
It is important to remember that family psychotherapy is a clini-

cal science. The test of the goodness of its theories is that they
generate (or rationalize) actions leading to change in a direction
taken to be desirable (by someone who will pay the bill). We may
wish as well for the theories to be elegant, parsimonious, and iso-
morphic with other good theories; we should like for them to be
empirically verifiable (witness Bateson's impatience with efforts to
empirically verify the double bind hypothesis). But the clinician
will always buy bad theory that works, will willingly be effective
rather than rigorous if that works. The miracle of family therapy has
been that the shift to a systemic perspective
— "I need to meet with
—
your family in order to assist you" is an effective intervention in
and of itself. Conscientiously adhered to, from an open and non-
blaming stance, the assembling of a family in order to study and
modify the pain or malfunction of one of its members is enormously
facilitating. I have commented elsewhere that all clinicians should

now be obligated to explain why they have chosen not to begin their
involvement with a problem by this route.
There are two rivers of ideas flowing through the volume. The
first, as I have noted, builds on Bateson and is concerned with

assembling the elements of a unified theory of family function and

family therapy. Hoffman might well have limited herself to that
monumental task and to those authors whose works are most rele-
vant. But the pioneers, the iconoclasts, and the great originals also
have much to say to us. Where, in fact, a unified theory does not
—
yet fully exist surely this volume is helping to birth it a strategic —
xiii
 FOREWORD
choice for Hoffman is how to deal both with the one truth and the
many truths.
Of course all the authors cited are both foxes and hedgehogs,
knowing both many truths and only one (but knowing that one
well).Yet the great naturals and the pure truth seekers need differ-
ent treatment. Compare, for example, Ackerman and Minuchin,
Satir and Jackson. Both pairs originated and worked in the same
general setting: Satir and Jackson at the cybernetics-oriented Mental
Research Institute in Palo Alto, California; Ackerman and Minuchin
having their roots in the east coast world of psychoanalytic child
psychiatry. All four have powerful personas, and there no doubt is

that they make a strong impression; they are distinctive and indi-
vidual, no one would mistake one for the other. Yet Ackerman and
Satir need to be preserved as people if their genius and special
contributions are to be properly noted. Hoffman is to be credited for
her awareness that this is so and for her ability to accomplish the

task. She has a curious and beneficent eye, a mode of inquiry that
asks not only, " What is going on here?" but "What is going on here
that is good?" The results of her inquiry are set forth lucidly and
felicitously. It has been my good fortune to know all the folks with
whom Hoffman deals in this volume: I can attest that she is highly
accurate in her selection of their essential ideas and, where personal
description is called for, a word painter with a complex palette that
she uses well.
Consider her sketch of Virginia Satir. Satir has been enormously
influential; she was one of the original group at the Mental Research
Institute in Palo Alto, the founding mother of the field of family
therapy. Over the years she has energized multitudes of patients
and colleagues. Yet it is fair, I think, to say that Satir is sui generis,

as Milton Erickson, in Hoffman's view and mine. Magnificent

is

exemplars, they teach by being. Their genius sharply contrasts with

the cookbook techniques of Jay Haley, master dramatist that he is,

who eschews all "use of self" and forces his students to be circum-
scribed and explicit about what is to be fixed and how to do it.
But it is Hoffman's anecdote about Satir to which I wish to direct
the reader. It takes place in a The identified patient
first interview.
is a young man who has impregnated two young girls. The family
is consumed with shame; the boy sits isolated in a corner of the

xiv
FOREWORD
treatment room. And there is Satir's marvelous opening line,
". . we know one thing for sure: we know you have good seed."
.

In one sentence we have positive connotation, reframing, a master-

ful —
power rebalancing and likely enough a "deep" interpretation
of the history and dynamics of the event. We must be grateful to
Hoffman: She was there (it took place in 1963), she knew it was
significant, she remembered it and has told us about it.
This volume is crowded with such anecdotes, drawn from direct
observation and from a sensible reading of the literature. Above all
there are Hoffman's original ideas, worked into and interpreted with
the other material she presents. As a single example, her linking of
Ashby and Minuchin in her classic paper, "'Enmeshment' and the
Too Richly Cross-Joined System," will provide nourishment for
clinicians and theorists for many years to come. I would have liked
the book to be longer, and mean by that to praise it with a faint
damn. The constraints of space limitation should be noted by the
reader, particularly the tyro, who should take this volume as a
prodigious homework assignment. Readers will need to return to it

again and again —as good a reason for buying a book as any I can
think of.

Donald A. Bloch, m.d.

February, 1981

xv
Foundations of

Family Therapy
Prologue

Behind the
Looking Glass

This book is newly discovered kingdom, the world

a journey to a
behind the looking glass. For me, the advent of the one-way screen,
which clinicians and researchers have used since the 1950s to ob-
serve live family interviews, was analogous to the discovery of the
telescope. Seeing differently made it possible to think differently.
And new ways of thinking have led to an epistemological revolu-
tion, one that touches all the sciences and that challenges many
traditional concepts, from the belief in linear causality to theories
of individual motivation.
Family therapy, although not a behavioral science per se, is in the
odd position of being one of the few areas of behavioral research
and practice to be influenced by this epistemological shift. It is
therefore more than just a novel therapeutic technique; it is based
on new assumptions about human behavior and human interaction
that have far-reaching implications. To really understand it, we will
have to go back several decades and explore the diverse themes and
concepts around which the family movement has evolved.
 Foundations of Family Therapy

A Bicameral Model

Let us start with the technological invention just described: the

screen. The late anthropologist Gregory Bateson speaks in Mind and
Nature of the advantages of a bicameral format —the jump to a new
perspective or emergence of new possibilities that follows the plac-
ing together of two eyes, two hands, two chambers of the brain. 1
This format applies also to the one-way screen. The screen turned
psychotherapy into a bicameral interaction that offered a similar
chance to explore a new dimension. One had two places to sit. One
could take a position, and have somebody else take a position com-
menting on or reviewing that position.
It is not strange, then, that the screen became a stake-out place
from which to view the fauna of a realm that had always been
before us yet never truly seen. One of the early discoveries made
by those who first viewed families with schizophrenics was that
what were thought to be mental illnesses belonging to individuals
might not be illnesses in the medical sense. In fact, they might not
be disorders at all. Rather, they could be seen as orderly manifesta-
tions that had meaning in the families or other social settings in
which they occurred.
Not only the process of assessment but the process of therapy
profited from the two-chamber framework. The use of the two
—
rooms to divide the tasks of therapy regardless of how this divi-
—
sion was described led to a new and more powerful way of organ-
izing systems change. With this format it became possible to aban-
don what was becoming for many an outmoded concept: the
concept of the therapist as a free-standing agent acting upon a
free-standing subject, the client or family.
Why was this concept becoming outmoded? To explain, I will

have to enlarge the field of vision and describe a cluster of ideas that
has been rocking our Aristotelian universe for a long time. The shift
to these ideas is linked very closely, first, to developments in such
fields as physics, biology, mathematics, and second, to the cognitive
sciences that have emerged from computer technology. The figures
who seem to have had the most impact on the family field in its
BEHIND THE LOOKING GLASS
infancy were, oddly enough, not so much psychotherapists but such
scientists as information theorist Claude Shannon, cyberneticist
Norbert Wiener, and general systems theorist Ludwig von Ber-
talanffy. One must add to this list Gregory Bateson, whose synthe-
sizing genius showed how ideas from such divergent sources could
be useful to the understanding of communication processes, includ-
ing those associated with psychopathology. Bateson was also one of
the first to introduce the idea that a family might be analogous to
a homeostatic or cybernetic system.
Unfortunately, however, for those who fam-
like simplicity, the

ily field did not develop in a straightforward fashion from the

ideas of these early thinkers. There are now two distinct genera-
tions of thought in family therapy. Building on the cybernetic
model, theorists like the late psychiatrist Don Jackson, at the
Mental Research Institute in Palo Alto, California,emphasized the
equilibrium-maintaining qualities of symptomatic behaviors in
families, as if they were literally analogous to homeostatic ele-
2
ments. Recently, theorists like University of Texas psychologist
Paul Dell have developed an evolutionary rather than a homeo-
3
static model. Deriving his evolutionary epistemology from the
work of a group of scientists who came to prominence during the
1970s, such as physicist Ilya Prigogine, or biologist Humberto
Maturana, Dell conceptualizes families, as well as all living sys-
tems, as evolving, nonequilibrium entities capable of sudden
transformations. Dell applies this evolutionary paradigm to a con-
sideration of family systems, in contrast to the homeostatic para-
digm of the earlier family thinkers.

Circular versus Linear Thinking

The central concept of the new epistemology— both the homeo-

static and the evolutionary paradigms — is the idea of circularity. In
the field of mental health there has been a growing disenchantment
with the linear causality of Western thought. Mental illness has
traditionally been thought of in linear terms, with historical, causal

5
 Foundations of Family Therapy
explanations for the distress. Efforts to explain symptomatic behav-
ior have usually been based on either a medical or a psychodynamic
model. The former compares emotional or mental distress to a bio-
logical malfunction or illness. Treatment consists of finding an "eti-
ology" of the so-called illness (a typically linear construct) and then
instituting a treatment, such as administering drugs or devising
other means of altering or blocking those bodily processes which are
considered responsible for the patient's state. The people in charge
of this treatment would of course be doctors, and the settings would
often be hospitals.
The psychodynamic model is influenced by nineteenth-century
discoveries about such forms of energy as electricity and steam. As
with the medical model, etiology is conceived of in linear terms.
Symptoms are said to arise from a trauma or conflict that originated
in the patient's past and that has for a variety of reasons been
relegated to the unconscious. Treatment consists of helping the
patient to recover the memory of the repressed event, which could
an unacceptable wish, and to reexperience the
also be a fantasy or
emotions that were buried within it. Once the secret material
becomes known to the patient and the buried emotions are "worked
through" within the safe confines of the therapeutic relationship,
the patient will presumably no longer need the symptom.
Thus these two models typically see symptomatic distress as a
malfunction arising either from biological or physiological causes,
or from a repressed event in the past. In both models the individual
is the locus of the malfunction, and the etiology is connected with
an imperfection in his genes, biochemistry, or intrapsychic develop-
ment.
After decades of strict adherence to these models, a new concep-
tualization began to emerge. Evidence provided by the watchers
behind the screen supported the growing disenchantment with the
linear, historical view. If one saw a person with a psychiatric afflic-
tion in a clinician's office, it would be easy to assume that he or she
suffered from an intrapsychic disorder arising from the past. But if
one saw the same person with his or her family, in the context of
current relationships, one began to see something quite different.
One would see communications and behaviors from everybody
present, composing many circular causal loops that played back and

6
BEHIND THE LOOKING GLASS
forth, with the behavior of the afflicted person only part of a larger,
recursive dance.
Of all those writing about the shift to a circular epistemology, it

is Gregory Bateson who most persistently tried to capture this elu-

sive beast. In Mind and Nature, he makes a distinction between the
world of physical objects and the world of living forms. 4 The physi-
cal world, the world of Newton, assumes a billiard ball model in
which causality is linear and forces act unidirectionally upon things.
Bateson objects that the world of living forms is poorly explained
by comparing it to a billiard table. In the world of living forms not
just force but information and relationship become important.
The classic example of this viewpoint is the difference between
kicking a stone and kicking a dog. In the case of the stone, the
energy transmitted by the kick will make the stone move a certain
distance, which can be predicted by the heaviness of the stone, the
force of the kick, and so forth. But if a man kicks a dog, the reaction
of the dog does not depend wholly on the energy of the man,
because the dog has its own source of energy, and the outcome is
unpredictable. What is transmitted is news about a relationship
the relationship between the man and the dog. The dog will respond
in one of a number of ways, depending on the relationship and how
it interprets the kick. It may cringe, run away, or try to bite the man.

But the behavior of the dog in turn becomes news for the man,
which may modify his own subsequent behavior. If, for instance,
the man is bitten, he may think twice before kicking that particular
dog again.
Therefore, Bateson would argue, we need a new grammar, a new
descriptive language, to depict what is going on in the living world.
What grammar? First, as we might expect, it ob-
characterizes this
jects to "thing" language, which grows out of linear notions of cause

and effect, in preference to a recursive language, in which all ele-

ments of a given process move together. "The man used a scythe to
mow a field" is thing language and is linear. It suggests that one
marked-off segment (a man) took another marked-off segment (a
scythe) and used it to chop up another segment (a field). One gets
the linear progression: A, using B, acted upon C, to effect D. Here
is a recursive, circular description of the same process by Mary

Catherine Bateson, the anthropologist's daughter:

7
 Foundations of Family Therapy
A man with a scythe is constrained by the form of the scythe; indeed his
own body motion is informed by the curves of his tool, a concrete proposi-

tion about the interlocked movement of man and tool through deep grow-
ing fields across the generations; as time passes, his own musculature will
become a record of the scythe's teaching, first in stiffness, then in emerging
grace and skill. We need time to understand this system, to get beyond
seeing it as simply instrumental. 5

In the case of living systems, it is not possible to assign one part a

causal influence vis a vis another, or to put in any linear markers at
all. As Bateson says, a brain does not "think." What "thinks" is a
brain inside a man who is part of larger systems residing in balance
within their environment. One cannot draw a line indicating one
part that thinks and another that is profiting by the thinking. "What
6
thinks is a total circuit."
Similarly, in describing the evolution of the horse, Bateson talks
about the relationship between horse and grass in which each reacts
back upon the changes of the other. To speak of the horse "evolv-
ing" and the brain "thinking" as if they were not part of an ongoing,
self-reflexive process that includes other elements would be to ig-
nore the laws of relativity for living forms. Newtonian descriptions
classify an item according to inherent attributes and characteristics.
Recursive descriptions define an item in terms of its relationship
with other items. To quote Bateson again:

I was utterly fascinated, and still am, with the discovery that when you
use language rightly to describe a flowering plant you will say that a leaf
is a lateral organ on a stem which is characterized by having a bud, namely

a baby stem, So the definitions became: a stem is that which

in its axil.
bears leaves, and a leaf that which has a stem in its angle; and that which
is

is in the angle of the leaf is a baby stem, and so on. 7

Ideas like these have extraordinary implications, not least when

applied to the field of psychotherapy. The therapist can no longer
be seen as "impacting" on the client or family through personality,
craft, or technique. The therapist is not an agent and the client is

not a subject. Both are part of a larger field in which therapist,

family, and any number of other elementsand react upon each act
other in unpredictable ways, because each action and reaction con-
tinually changes the nature of the field in which the elements of this

8
BEHIND THE LOOKING GLASS
new therapeutic system reside. A circular epistemology forces the
therapist to take account of the fact that he or she is inevitably part
of this larger field, an inextricable element of that which he attempts
to change.

The Beginning of the Journey

At the time this book was conceived, what we can now call the
homeostatic model of the early family therapy researchers was in
its infancy, and the evolutionary model based on recent ideas from

physics and other scientific fields hardly born. In 1970 I wrote a

paper on deviation-amplifying processes, originally entitled "Be-
yond Homeostasis/' 8 which contained ideas that now seem to me
an attempt to bridge the two positions, or move the first one for-
ward. It was published simultaneously with a similar essay, Albert
7
Speer's "Family Systems: Morphostasis and Morphogenesis/
which also broke the confines of the homeostatic model. 9
This book arose out of a compelling need to build a framework
that would explain where the concepts flowing into both models
came from and how these models fit together with the many other
ideas and models that are still bubbling up out of the family therapy
field. (Family therapy was, and still is, a wondrous Tower of Babel;

people in it speak many different tongues.) In addition, I have tried

to integrate other research in the social and behavioral sciences that
can back up the observations of clinicians working with families.
My fascination with social fields and with describing them sys-
tematically also played a part in the explorations that resulted in
this book. My experience resembles that of those early meteorolo-
gists who weather systems could not be understood
realized that
one man's downpour could be another man's drought.
locally, that
In studying weather systems one might encounter complex redun-
dancies in the way various elements intersected: wind, vector cur-
rents, clouds, moisture, coldand warm fronts, time zones, latitudes
and longitudes, the pulls of lunar gravity, or the flares on the sun.
Above all, there would have to be some way to account for the
 Foundations of Family Therapy
changing differences among these variables. To grow and evolve, the
science of meteorology had to find a kind of crow's nest
from which
it was possible to observe patterns and sequences moving the same

way or differently through time, rather than seeing only particular

pieces of weather that happened to occur in this or that place. In
short, what had to be discerned were the larger configurations that
make up our modern weather charts. Satellite photographs, taken
from literal sky-borne crow's nests, now show the spirals of these
weather systems, graphically portrayed in cloud formations circling
the earth.
The study of human behavior has undergone similar changes. As
long as one stood on earth, so to speak, and experienced rain one
day and sun the next, one had to invent a demonology that con-
trolled these different manifestations. Similarly with unusual
behaviors. A
demonology explaining irrational behaviors was in-
vented time and again by human groups to explain the variable
weather of the soul. In some periods, powerful spirits were believed
to act upon a person from without; at other times, powerful im-
pulses were seen to control the person from within.
Only recently has it been recognized that just as weather can be
seen as large, moving systems, so perhaps human behaviors may
derive from large relationship configurations moving through time.
To say, "This is a schizophrenic," that is, a person with a supposed
mental disorder, makes as much sense as to state, "This is a rainy
day." The so-called schizophrenic can just as well be described as
a manifestation of a larger weather system in human affairs. The
next step is to find an imaginary from which to view the
satellite

patterns and sequences that will give us weather charts for such
behaviors, at least within small, reasonably stable groups.
That, of course, is the problem. Behaviors cannot be studied apart
from the fields in which they occur, but the fields must be integrated
enough to allow study. How much easier it is to understand the
movements of ants or the dances of bees. If only large human
structures —nations, societies, cultures —were as homogeneous or
predictable. The family one system that transcends the limits of
is

the single person yet is bounded enough to serve

small and clearly
as a researchable unit. In the family, as in weather, once one leaves
the individual and surveys the family as a systemic entity residing

lo
BEHIND THE LOOKING GLASS
within even larger fields, one begins to see clear redundancies and
distinct patterns.
Thus it is not hard to see the powerful lure family research held
forone eager to explore social fields from a systemic view. I first
stumbled upon family research in Palo Alto in 1963. There, at the
Mental Research Institute, I began to see from the studies that had
come out of Gregory Bateson's 1952-1962 research project on com-
munication that a change in a family depended very much on the
interplay between deviation and the way deviation was kept within
bounds.
I was also intrigued by the idea that deviation, or deviance, per
se, was not the negative thing it was thought to be, once one aban-

doned the point of view of those who wanted it corrected. Devia-

tion (including symptomatic and irrational behaviors of all kinds)
could be highly important for a group. Although homeostasis was
a central concern of the Palo Alto family researchers, when I read
their writings I found myself more interested in what worked against
homeostasis; what introduced variety, strangeness, novelty. It

seemed to me paradoxical that families with symptomatic members

were thought to be pathogenic since I began to suspect that only
when someone or something deviated from the family norms could
the family derive new information and evolve new structures.
Without some chink through which variety might enter, there
seemed no possibility for a system to achieve basic structural
change. Most families must reorganize as the generations are born,
grow old, and die. If a family could not achieve that kind of change,
it would most likely not survive.

Families with symptomatic members thus became illustrious ma-

terial for study, since in those families the issues of change would

be most intensely highlighted. Therefore I began by asking what

were the properties of family fields in which new information, and
hence change, entered with all. Were there explana-
difficulty if at
tions for the stability of these fields? They seemed to remain rela-
tively the same despite the need for periodic reorganizations that
every family must face. Was there research in other fields that might
throw light on these mysteries? Was there even a language in which
to express these concerns, since our old language seemed designed
by its very structure to obscure them? These were the kinds of
 Foundations of Family Therapy
questions to which I sought, if not answers, at least indications that
would tell me where to look.

Organization of the Book

This story unfolds with the detective work of those early family
researchers who
gazed into the murky depths of families with
first

schizophrenic members and recorded what they saw. Much early

research on schizophrenia and the family, as Dell points out in a
recent essay, merely sought to provide a new theory for its cause,
whether that was the family, the "schizophrenogenic mother," or
some other agent. 10 I will focus on the research that does not pri-
marily offer a new etiology but moves us along the epistemological
path I am tracing.
In Chapter 2 I move model suggested by Bate-
to the cybernetic
son's seminal ideas on schismogenesis. This term, though ponder-
ous, nevertheless contains a blueprint for the way social groups
cohere or split, stay viable or reorganize. It is also a concept that can
be applied to many escalation processes, especially those found in
social interaction.
Early clinicians had been fascinated by the tendency they per-
ceived in families to maintain the status quo, and posited that some-
thing akin to homeostatic mechanisms were responsible. What in-
terested me was the contrary process whereby antihomeostatic
processes might take over. The implications of these processes for
systems change is the subject of Chapter 3. Some small deviation
could get out of hand and create a "runaway" or positive feedback
chain. was anyone's guess whether the original organization of
It

the system would change, be destroyed, or stay the same. But it

seemed to me that in the grand scheme of things, deviation was a
source of new information essential to the survival and evolution
of social groups, and that the early family theorists had not empha-
sized this aspect sufficiently.
In Chapters 4 and 5 I begin to investigate in more detail the issue
of family typology. At first, family researchers tried to link symp-

12
BEHIND THE LOOKING GLASS
tomatology to family types: the schizophrenic family, the alcoholic
family, and so on. But this kind of typology is hard to establish,
especially since a family may contain persons displaying a number
of different symptoms. I also explore other attempts to create
typologies: bipolar ones, in which families are ranged along a con-
tinuum with each end representing an opposite form of organiza-
tion; process models, with families organized according to different
types of sequences; grid models, representing more than one dimen-
sion;and developmental models, showing a continuum from "pa-
thology" toward "normalcy." However the typology is arranged,
different categories of families are usually linked to different classes
of disorders. At the same time there is the possibility that the whole
question of typology may be either premature or a dead end, and
that a focus on family "paradigms," or system-wide formulas for
processing information and change, may be more useful.
In Chapters 6, 7, and 8, I try to put under the microscope the
tissue of one particular kind of family: the family that produces
severe psychiatric disorders. Researchers had found in this one type
of family, at least, distinct patterns of organization either different
from, or more intense than, similar forms in seemingly normal fami-
lies. The application of coalition theory by members of the Bateson
group to the structures typically found in "disturbed" families led
me to a broader look at coalition theory and its first cousin, struc-
tural balance theory. Of all the areas of social psychology that I

explored, the theory of structural balance (though originally in-

tended to explain cognitive, not social, fields) was the only theory
that was in any way predictive for the formal interaction sequences
one could see operating in families with symptomatic members.
Consequently, in this group of chapters the focus is directly on
the characteristics of triangles one can expect to see in "disturbed"
families. These triangles obliterate generation lines, confuse appro-
priate boundaries between family subgroups, and subvert the fam-
ily hierarchy as prescribed by a given culture. At the same time, we
find that they are associated with families so rigidly organized as to
make any change in organization problematic, especially changes
associated with the growing up of the children. Research, both in
families and in organizations, suggests that one possible reason for
the persistence of these inappropriate triangles is that the child (or
 Foundations of Family Therapy
other third party) presents a problem that keeps covert difficulties
or conflicts in important executive pairs from surfacing.
At this point we move from an emphasis on family theory to an
emphasis on theory of change. The book becomes far more clinically
oriented and process variables rather than structural variables come
into view. The idea that living systems often make sudden evolu-
tionary shifts at natural transition points in the family life cycle is

the subject of Chapter 9. The appearance of symptoms at these

times may indicate that a particular transition is perceived by the
family as problematic, even dangerous. Not all families are able to
negotiate these transitions on their own. When a symptom devel-
ops, it can be seen as a constant reminder, in symbolic form, of the
need for change, while same time apparently blocking it.
at the
7
Chapter 10, 'The Thing in the Bushes/ explores the target most
therapists seem to be gunning for in terms of repetitive cycles or
sequences. These sequences are presented not as dysfunctional but
as having a logic and a meaning at the family-system level, even
though they may be experienced as painful or stressful by individ-
ual family members. Chapter 11 examines how this type of se-
quence is broken or disrupted by therapists working in four differ-
ent models.
Although this book is not intended to be a historical overview,
it seems to me that the practice of therapy form of live research,
is a
a notion I illustrate in the subsequent chapters on family therapy
pioneers and major schools. The clinical work of successful pioneer-
ing therapists is a prime source of information on families and
family therapy. The experienced clinician intuitively recognizes the
shape of common symptomatic configurations and knows how to go
about changing them. The schools I describe are singled out because
they represent consensus positions: a consolidation of practical and
theoretical issues following the trails blazed by the first explorers.
I thenmove to an examination of an important new development:
the systemic approach of Mara Selvini Palazzoli and her colleagues
in Milan. Originally influenced by the early formulations of the
Bateson group, the Milan Associates continue to work more and
more framework of circular causality.
closely within a Batesonian
In both their theory and their therapy they have taken a leap toward
an idiosyncratic and original model that is very different from that

H
BEHIND THE LOOKING GLASS
of their current Palo Alto colleagues. Chapter 15 describes their
present work.
The last two chapters are speculative and raise more questions
than they answer. Chapter 16 is a discussion of the therapeutic
double bind, and of various theories about why it works. Chapter
17 describes issues that are now coming to the fore and that lead
to a consideration in the closing chapter of the implications of the
new evolutionary epistemology to which we have been alluding
throughout the book.
On a more personal level, this evolutionary epistemology can be
applied to my own journey. Looking back on the thought and study
that led to this book, as well as to other work in the field, what may
seem like blind spots and dead ends also represent stages in a neces-
The very logic
sary process. of an evolutionary model prohibits
throwing away trials that are unsuccessful. The only prohibition is

against continuing to repeat such trials.

With this proviso, let me turn back to the 1960s, when I first
became acquainted with the work and writing of the early systems
thinkers, clinicians, and other pioneers of the family therapy move-
ment, and try to convey the enormous impact their discoveries made
upon my own thinking and writing and upon the development of
the family field.

15
Chapter 1

Early Research
on Family Groups

Live Observation

It was when people with symptomatic behaviors were first ob-

served in their natural habitat — the family, rather than a clinician's
office —that the family movement began. One can say that there is

such a thing as an epidemiology of ideas. Paralleling its use in the

growing field of animal ethology, live observation began to be used
with human families in formal or informal research during the
1950s. At the same time lone clinicians were stumbling onto family
therapy and then stumbling onto each other as they sought to
validate the information they were getting.
However, the rules of the psychoanalytic establishment banned
the contamination of therapy by the inclusion of relatives. As a
result, "treatment" occurred mostly in the guise of research. An
at-home anthropology developed in which clinicians took the lead,
with the family context of psychiatric disorders becoming visible
for the first time. As the Chinese say, ''Only the fish do not know

16
EARLY RESEARCH ON FAMILY GROUPS
that it is water in which they swim/' Humans also have an inability

to see the relationship systems that sustain them.

There was no father or mother of family therapy, and no first
family therapy interview. Like Topsy, the movement "just
growed." The major impetus for its growth came out of the work
of researchers like Nathan Ackerman in New York; Murray Bowen
in Topeka and Washington, D.C.; Lyman Wynne and Margaret
Singer at the National Institutes of Mental Health in Bethesda; Carl
Whitaker in Atlanta; Salvador Minuchin and E. H. Auerswald at the
Wiltwyck School in New York State; Ivan Boszormenyi-Nagy,
James Framo, and Gerald Zuk in Philadelphia; Theodore Lidz and
Stephen Fleck at Yale; and Gregory Bateson, Don Jackson, Jay
Haley, John Weakland, Paul Watzlawick, John Bell, and Virginia
Satir in Palo Alto, to mention only a few. These people, their core-
searchers, and many others in cities across the country became the
backbone of a new and engrossing movement of practice and ideas. 1
Most of these researchers focused on investigating the properties
of the family as a "system." By "system" they usually meant any
entity whose parts co-vary with each other and which maintains
equilibrium in an error-activated way. Their emphasis was on the
role played by symptomatic behaviors in helping to balance or
unbalance that system.
However, the family movement represents more than a different
approach to therapy. It is a different way of looking at behavior, and
it can be described as a kind of communications research focusing
on the face-to-face relationships of people in ongoing groups. This
position has been summarized in Watzlawick, Jackson, and Beavin's
classic Pragmatics of Human Communication, a book that was also the first

attempt to popularize the seminal ideas of Bateson and his group. 2

The Bateson Project and "Learning to Learn"

In the 1950s, Gregory Bateson was leading a remarkable research

project that attempted to classify communication in terms of levels:
levels of meaning, levels of logical type, and levels of learning.

17
 Foundations of Family Therapy
Among other areas of interest — such as animal behavior, paradoxes,
hypnosis, play — the group looked at patterns of schizophrenic
transaction. They wondered whether these patterns might arise
from an inability to discriminate between levels of logical type
between the literal and the metaphoric, for example. People who are
thought of as crazy will use figures of speech literally, or talk in
metaphors without acknowledging it. The Bateson group hypothe-
sized that a person with this kind of difficulty might, in Bateson's
words, have "learned to learn" in a context in which this difficulty
was in some way adaptive. If this learning context could be under-
stood, the mysteries of schizophrenic speech and behavior might be
understood. Because the family is the primary learning context for
humans, they reasoned that the family of the schizophrenic might
have shaped his peculiar ways via the peculiar communication re-
quirements that were imposed upon him.
At the same time, psychotherapists who were independently
watching schizophrenics in the context of the family observed that
if the patient got better, someone else in the family usually got

worse. It was almost as if the family required the presence of a

person with a symptom. Not only did the Bateson group find evi-
dence for this assumption, but it was impressed by the extent to
which the family encouraged, even demanded, that the patient
show irrational behavior.
Noting the obstinacy with which change was resisted, even when
it meant the improvement of a loved one, Jackson coined the term
"family homeostasis." He described family interaction as "a closed
information system in which variations in output or behavior are
fed back in order to correct the system's response." 3 Haley elabo-
rated on this, comparing the family to a servomechanism with a
"governor":

Granting that people in on-going relationships function as "governors" in

relation to one another, and granting that it is the function of a governor
to diminish change, then the first law of human relationships follows:
When one person indicates a change in relation to another, the other will act upon the first
4
so as to diminish and modify that change.

The Bateson group became identified with an idea of the family

as an equilibrium-maintaining entity partly because so much of the

18
EARLY RESEARCH ON FAMILY GROUPS
group's research was done with families that had an extremely
restricted range of behavior. A major question was whether a family
could be said to behave as a "system" at all —whether all families
had communications than one would
a greater patterning in their
expect if these communications were ruled by chance. This question
seemed to be answered affirmatively in experiments showing a
greater rigidity of communication patterns (for example, in order of
speaking) in families in which somebody had a symptom than in
which nobody did. 5
families in
However, the Bateson group showed in their clinical work a clear
awareness of the importance of deviation in leading to a new setting
for the family system. Jackson often observed what he called a
"runaway" while working with families. This referred to any ampli-
fying feedback process that escalated rapidly, leading to a break-
down, a blowup, or some violent result. Jackson often said that he
preferred to work with a family in which this kind of movement
was taking place. With a very immovable family, especially one that
included a chronic schizophrenic, he would sometimes try to start
a runaway as a therapeutic gambit. 6 This could be done by "pre-
scribing the symptom" —
that is, increasing the angle of deviance of
the patient from the rest of the family. Alternatively, the therapist
could reinforce any family member's behavior, pushing it to con-
tinue in the same direction in a kind of redurtio ad absurdum. Such
interventions would presumably threaten the homeostasis of the
family and thus lead family members to grasp the therapist's
suggestions more readily, in the hopes of establishing a new balance
or to come up with a new balance of their own.

The Double Bind

What is strange is that during the early phases of the Bateson

group's study of schizophrenic communication, nobody thought of
observing schizophrenics with their families at all. Instead, inter-
views were held with hospitalized patients at the Palo Alto Veterans
Administration Hospital, where Bateson was a consultant. As a

19
 Foundations of Family Therapy
result of these interviews, and many conversations between mem-
bers of the group, the double bind hypothesis began to take shape.
In 1956, thenow famous paper, 'Toward a Theory of Schizophre-
7
nia/ was published and the double bind concept was finally born. 7
"Double bind" describes a context of habitual communication
impasses imposed on one another by persons in a relationship sys-
tem. Under some circumstances these impasses seemed to elicit the
responses known in their aggregate as schizophrenia. A double bind
was in essence a multilevel communication in which an overt de-
mand at one level was covertly nullified or contradicted at another
level."Dominate me" is a relatively nontoxic example of the double
bind. Here the person addressed can only "dominate" by obeying
— which is the opposite of domination. Such a request is therefore
impossible to respond to. Like any paradox, it has to be teased apart
into its two levels: (1) the reported wish to be submissive, con-
tradicted by (2) the implicit or explicit command that the listener
obey the speaker. The "command" message is of a higher logical
type than the "report" message, because it states who sets the rules
for the subclass of allowable behaviors. The only way one can
respond to such a request is to point out how impossible it is, to
make a joke of it, or to leave the field. But when none of these
courses is possible —and when the confusion between report and
command levels imposes itself as a kind of confusion in the re-
ceiver's mind — then serious trouble can arise.

In the original double bind article, examples of this kind of im-

passe were given and formal conditions were set as prerequisites for
its appearance in a toxic or pathogenic form. Here are the basic
ingredients:

1. A
primary negative injunction, ''Don't do that."
2. A
secondary negative injunction at another level which conflicts
with the first: "Don't listen to anything I say" (perhaps conveyed by tone
of voice or manner).
3. An injunction forbidding comment (usually nonverbal cues rein-
forcing rules that no longer need to be made explicit) and another forbid-
ding the person to leave the field (often delivered by context, as when the
person is a child).
4. A situation that seems to be of survival significance, so that it is
vitally important for the person to discriminate correctly among the mes-
sages.

20
EARLY RESEARCH ON FAMILY GROUPS
5. After a pattern of communication containing these elements has
become established, only a small reminder of the original sequence is
needed to produce a reaction of panic or rage.

As an illustration the article cites the example of a mother who

is feeling bothered by a child, but instead of saying "Go away, I'm

sick of you" says "Go to bed; you're very tired and I want you to
get your sleep." If the child accepts this loving concern at face value
and attempts to draw closer, the mother will probably move away.
If he challenges the loving behavior or reacts negatively, she will

probably get angry. If he comments on her anger, she may get even
angrier. Thus he will be punished for discriminating accurately. He
will most likely be too confused to comment on his predicament,
and he will be unable, being a child, to leave the field. This would
be a formal example of a double bind.
It is amazing to think that the double bind article was written
without first-hand observation of the transactions described. But
this omission is understandable if one considers the influence of
traditional psychoanalytic thought. Most of the early family re-
searchers had an analytic orientation, which not only held that a
symptom was a sign of internal dysfunction originating in the past,
but also forbade the therapist to see relatives of a patient for fear
of contaminating the intense relationship with the therapist, or
transference, that was seen as an essential ingredient of the thera-
peutic process.
Fortunately, a chance occurrence alerted Bateson and his col-
leagues to the importance of what was going on in the family in the
present. Jay Haley, who met with patients at the VA hospital and
audiotaped his conversations with them, found that one young man
had severe anxiety attacks every time his parents visited him. To try
to find out why the patient reacted as he did, Haley asked the
parents tocome to the next interview. Out of subsequent meetings
with the young man, his parents, and his therapist came a serendipi-
tous development: an audiotape in which an obviously adoring
mother, in the space of a few minutes, turned a fairly rational son
into a person who displayed confused thinking, contradictory state-
ments, irrelevant remarks, and other communication patterns as-
sociated with the "thought disorders" of the condition known as
schizophrenia.

21
 Foundations of Family Therapy
The incident occurred shortly after Mother's Day. The young
man's mother showed the therapist a card that her son had sent her
from the hospital. The therapist read the legend: 'To One Who Has
Been Just Like a Mother to Me." The mother declared that she felt
very hurt. The son defended himself; saying, "Look, Ma, I only
wanted to sting you a little." The mother then put on a dazzling
display of contradictory statements, defining herself as willing to
accept any hurt would help him, placing herself in the cate-
if that
gory of the Virgin Mary who would do anything for her son, and
at the same time declaring that all she and his father wanted was

for him to stop picking them to pieces, because they were ordinary
people who did not deserve that kind of treatment. The son backed
down under he claimed that he didn't even
this onslaught. First
remember the card, then he blamed the commissary for selling cards
like that, then he said that he hadn't been particular enough about
the wording on cards since his illness, and finally he insisted that
he felt that she had been a "good enough mother." When his father
added helpfully, "A real mother," he repeated, "Yeah, a real mother
—so that's all." 8

We seem to be dealing here with a rather eerie cat-and-mouse

game. Such games probably justify the initial reason Bateson be-
came interested in schizophrenic communication: the apparent ina-
bility of the schizophrenic to distinguish between the literal and the
metaphoric. This seemed to be translated into a virtual allergy to
any "report" message that secretly encloses a "command" message
indicating that the sender controls the relationship.
Seen in some "thought disorders" may be redefined as
this light,
maneuvers in a desperate struggle. The vague, amorphous, or dis-
qualifying communications presented by the other parties in this
struggle (always the mothers of schizophrenics in the early litera-
ture) may be equally well defined as maneuvers. One could see the
logic of a resort to confused thinking if accepting the definition of
the relationship implied by another person's statement would
amount to giving the other person the upper hand.
For instance, a mother might say to a hospitalized son: "The way
you're staring at me, you look ready to hit the ceiling." Son might
counter this apparently innocent remark (implying hostility on his
part, of course, and not so innocent) by saying, "I'm not tall enough

22
EARLY RESEARCH ON FAMILY GROUPS
to hit the ceiling/' and follow this statement with silly jumps.
Mother might then ask, "Have you taken your medication today?"
(translated: You're sick, not mad at me —
but I still control the defi-
nition of the relationship).
Using metaphor in a "crazy" way (without indicating that it is a
metaphor) is another way to escape. An example of this comes from

a filmed interview with the family of a schizophrenic girl.The father

tells the girl that she is to say anything she likes to the doctor. The
girl replies with a seemingly irrelevant remark: "It's a free country,
but the prices are fixed." Translated from schizophrenese, this
might mean: "Dad says I'm free to talk, but we all know that
whatever I say will be used against me."
At any rate, once researchers and clinicians began to observe live
family interaction, there was no going back to ivory-tower specula-
tion. A torrent of articles describing the marvels of communication
in families with schizophrenics began to pour out. Most of the
articles written by researchers associated with the Mental Research
Institute in the 1960s contained brilliant examples of strategies
designed to establish or escape definitions of relationships by mem-
bers of these families. These communications could be disguised as
irrational, confused, amorphous, or irrelevant, but they were seen
to be deadly nonetheless.
Many of these articles are reprinted in Jackson's two-volume
Human Communication Series, Therapy, Communication and Change,
and Communication, Family and Marriage. 9
Some of my favorites are
Weakland and Jackson's "Patient and Therapist Observations on
the Circumstances of a Schizophrenic Episode," Jackson and
Yalom's "Conjoint Family Therapy as an Aid to Intensive Ther-
apy," and Weakland and Mothers of Schizo-
Frye's "Letters of
phrenics." A thoughtful and entertaining compendium of "dou-
10

ble binding" communication has also been compiled by Sluzki,

Beavin, Tarnopolosky, and Veron in their article "Transactional
Disqualification: Research on the Double Bind. 11 For an exhaus-
tive compilation and critique of the double bind literature, see
Sluzki and Ransom's Double Bind: The Foundation of the Communicational
Approach to the Family. 12

23
 Foundations of Family Therapy

The Infinite Dance of Shifting Coalitions

work of the Bateson group furnished

In spite of the fact that the
a focal point for students of nonpsychodynamic transactional
thought, the communications approach had an inherent drawback:
it tended to be essentially dyadic. Since conversations between
three or more people were too complex to be analyzed on a mi-
crolevel, the unit of attention was usually the two-person exchange.
The double bind theory was itself originally formulated in dyadic
terms. There was a binder and there was one who was bound,
although the reciprocal nature of the bind was acknowledged. The
theory implicitly isolated a unit comprising two communicators,
with the focus of interest the characteristic type of exchange be-
tween them.
As a result, a number of articles qualifying the original double
bind theory began to emerge. Weakland was the first to break out
7
of the dyadic mold, with a i960 essay, 'The 'Double Bind Hypoth-
esis of Schizophrenia and Three-Party Interaction/' a brilliant prefi-
13
guring of later thinking in the field. 1962 the authors of the
In
original double bind article offered a critique that downplayed the
focus on individual behaviors or single sequences in favor of the
theory's emphasis on circular systems in interpersonal relations. 14
The following year Watzlawick put up a stouter defense of the
original paper than the authors had done, even while conceding that
they should have made the mutual nature of the bind clearer, in-
stead of presenting it as a one-way street. 15 Essays in the 1970s by
both Weakland and Bateson suggest that this concept does not have
to do with the etiology of schizophrenia at all, but is part of a larger
attempt to establish an epistemology within which terms like "eti-
ology" and "schizophrenia" would have no meaning. 16
The first step in this chain of revisions, as Weakland's i960 article
makes clear, was to look at schizophrenic behavior in terms of triads
instead of dyads. Members of the Bateson group were beginning to
show an interest in coalitions, and although they did not specifically
acknowledge the sources of this interest, it is instructive to note
Bateson's contribution. He offered an analogy from game theory for

24
EARLY RESEARCH ON FAMILY GROUPS
a type of behavior that the group had noticed over and over again
in families with a schizophrenic: the fact that no two persons
seemed to be able to get together, whether to be close or to disagree,
without a third person taking part. For this phenomenon Bateson
used the phrase "the infinite dance of shifting coalitions/' 17
Bateson's argument was was formally similar
that this behavior
to the instability of a five-person game described by Von Neumann
and Morgenstern in Theory of Games. 18 Von Neumann assumed that
intelligent, gain-oriented players could be expected to form coali-
tions to maximize their profits. However, the situation changed
when the number of players became five. Von Neumann describes
a possible five-person game (equally applicable to a three-person
game, according to Bateson) in which coalition possibilities became
inherently unstable. Every time a winning arrangement formed,
reasons of self-interest would force a new one to take its place.
Thus, as Bateson put it, 'There will always be a circular list of
alternative solutions so that the system will never cease from pass-
ing on from solution to solution, always selecting another solution
which is which preceded it." 19
preferable to that
Bateson thought that Von Neumann's five-person game offered
a rough analogy to what went on in families with a schizophrenic.
He added the proviso that in these families, three persons seemed
to be a sufficient number to get the same result. In schizophrenic
families, no two members ever seemed able to get together in a
stable alignment. Either another family member would intervene, or
the two who paired off would feel so uncomfortable about exclud-
ing the other person that they would dissolve the coalition them-
selves.
Though the original double bind described a two-person arrange-
ment, Bateson saw a way, through the game metaphor, to translate
the concept into a particular kind of family organization. He argued
that the untenable predicament of the schizophrenic could arise
from having to participate in the interactional equivalent of Von
Neumann's game. A robot would be insensitive to the fact that
every reasonable solution he arrived at was immediately proven
wrong. But human beings are not this insensitive. In fact, they have
an inflexibility bestowed upon them by their greatest asset, their
ability to learn —that is, their ability to acquire automatic responses

*5
 Foundations of Family Therapy
to habitual problems. Without this capacity, a person would be
forever inventing solutions to each problem as if he were encounter-
ing it for the first time. This is why human beings have a commit-
ment to the process of adaptation at the deeper level of habit.
Bateson argued that in a system where adaptations are not allowed
to persist at deeper levels, as in Von Neumann's unstable game, it
is logical to assume that the individual involved will experience
extreme disruption and pain. He will be caught in a perpetual se-
quence of double binds, situations in which there is always a pen-
alty for being right.

The "Control" Theory of Schizophrenic Transaction

In a later paper discussing the evolution of ideas during the ten-

year course of the Bateson project, Haley compares Bateson's inter-
pretation of the double bind to his own. 20 Bateson had suggested
this model for the behaviors in the family of a schizophrenic: ''the
schizophrenic communicates as if he expected to be punished every
time he indicates that he is view of the context of his
right in his
721
own message/ Haley argues that there is an implicit assumption
about motivation in this kind of thinking, similar to the traditional
idea that people are driven by inner needs and desires such as fear
of punishment, wish for love, and avoidance of pain.*
In Strategies of Psychotherapy, trying to move from an individual to
a systemic level, Haley examined the effect of the double bind tactic
on the larger interpersonal field of the family. 24 He started with the
idea, shared by the entire Bateson group, that the disqualification
of meanings is a recurrent feature of communication in a family
with a schizophrenic. A disqualified message is a statement affirmed
at one level and disconhrmed at another: "Of course I love you,"

*Paul Dell has pointed out that Haley is nevertheless in favor of ascribing something like
motivation to the system. 22 In the paper cited above, Haley states that a power struggle may
be expressed as a "need" of the system when individuals within it must struggle for control
because hierarchical levels are not clearly defined. 23

26
EARLY RESEARCH ON FAMILY GROUPS
said in an angry tone of voice. Note that if one person in a family
is disqualifying his own and others' messages, it will be hard for
everybody else not to reciprocate. The only response to messages
that conflict on different levels, Haley remarks, is more messages

that conflict on different levels. Thus we are left with a vicious

circle; once established, it keeps on going.
Now what, Haley asks, might one expect of persons caught in a
situation like this? He hypothesizes that they will become inordi-
nately sensitive to having their behavior governed by others. Dis-
qualification of meanings is, after all, a tactic a person can use to
control someone else's behavior. But it is a sword that cuts two
ways. It can be used as a counterstrategy to prevent one's own
behavior from being controlled. Thus a picture emerges of a type of
family dominated by issues of control. One example of a covert
battle for control of the relationship goes as follows (here a mother
is talking to her grown son):

patient: Did you bring my laundry?

mother: How do you feel?
patient: Do you have my laundry?
mother: You look sad.
patient: I'm O.K.
mother: Are you angry with me?
patient: Yes. 25

Moving to the larger framework of the family as a cybernetic

system, Haley remarks that just as the "governor" in a servomech-
anism controls the range of movements within it, so people in a
family act to control the range of one another's behavior. The trag-
edy of a family that uses this tactic is that the struggle for control
exists not only at the level of the particular rule, but also at the
metalevel of "Who is to set the rules?" The Russellian Theory of
Logical Types proposes that all messages consist of (1) a statement
and 26
(2) a statement about that statement. The theory establishes
a hierarchy of types or levels of abstractionand prohibits the joining
of a higher level with a lower level. Therefore there is no such thing
as a simple communication; every message is qualified by another
message on a higher level. When these two levels of message are

27
 Foundations of Family Therapy
treated as one, as in the example of a signboard reading "All State-
ments on This Signboard are Untrue/' we have a self-contradictory
situation or paradox.
Basing his argument on Russell's Theory of Logical Types, Haley
states that in the family control struggle, if rules on two levels of
abstraction are jammed together, a similar communicational diffi-
culty results, and there no way to end the struggle. At Level One,
is

each reports a statement. At Level Two, each attempts to define the

relationship that acts as the context for that statement. But since
any decision about behaviors (Level One) cannot be made without
an agreement as to who is to decide what behaviors are to be
allowed (Level Two), there is a constant denial and confusion about
almost anything anyone in such a family tries to say or do.
In the dialogue quoted above, the son was trying to focus on
concrete issues and the mother insisted on moving to issues that had
to do with how he was feeling. If the son agreed with the mother
about what was appropriate to talk about, he not only placed him-
self in a somewhat infantile position but gave her the right to decide
what should be talked about. This becomes a good illustration of the
"control" theory that Haley proposes as an explanation for the
constant mode of disqualification observed in families with a
schizophrenic member.

The Undifferentiated Family Ego Mass

At the same time the Palo Alto group was studying schizophrenic
communication, researchers from a clinical rather than a communi-
cations background were working the same territory. As this book
will not attempt to cover much of the important work in this area,
readers are referred to Riskin and Faunces's exhaustive study, "An
27
Evaluative Review of Family Interaction and Research." However,
one or two early figures deserve to be singled out for the boldness
of their ideas and their willingness to break with the traditional
language of psychodynamic theory.
Murray Bowen was one of the first psychiatrists to hospitalize

28
EARLY RESEARCH ON FAMILY GROUPS
whole families for observation and treatment. He had begun in the
1950s with the idea that schizophrenia was the result of an un-
resolved symbiotic tie with the mother. After working for about a
year with mothers and their children in a milieu treatment setting
at the Menninger Clinic, he began to feel that schizophrenia was a
sign of a larger pathology in the whole family and tried to have as
many family members as possible live on the hospital ward during
treatment. Subsequently, he developed a three-generation hypothe-
sis for schizophrenia. According to this theory, the grandparents of

the schizophrenic child were relatively mature, but one child, very
attached to the mother, remained extremely immature. Later, this
child chose an equally immature spouse. The result of the combined
immaturities in this marriage was a child who was so symbiotically
tied to the mother as to be schizophrenic.
From this research came many of Bowen's ideas about the charac-
teristics of emotional disturbance, applied both to families and to

the individuals who lived in them. The ideas that are especially
important for our discussion include the multigenerational trans-
mission of emotional illness; the importance of working with the

family of origin; and the concept of "differentiation." These areas

and other aspects of "Bowen Theory" are fully described in
Bowen's Family Therapy in Clinical Practice. 28
One of Bowen's major contributions to family theory is his think-
ing on the part played by triangles in family interaction. Triangula-
tion is a process that occurs in all families, all social groups, as
twosomes form to the exclusion of, or against, a third. The triangle
is an essential building block for Haley's theory of pathological
systems and Minuchin's structural approach to family theory. Un-
like Haley's and Minuchin's relatively static forms, however,
Bowen's sense of triangles is a fluid one.
For Bowen, a two-person emotional system will form a three-
person system under stress. For instance, tension might arise be-
tween the two, and the one who was more uncomfortable would
relieve tension by "triangling in" a third person, perhaps by telling
a story about that person. Then the tension would shift to the new
twosome, relieving the tension between the original pair. But the
outsider, once he becomes drawn in, may respond to the tension by
accepting an alliance with one of the others, so that the outsider of

29
 Foundations of Family Therapy
one moment becomes the insider of the next. For instance, a struggle
between a son and a mother might erupt over his going to school,
but become deflected when the father enters the scene and attacks
the mother and defends the son.
In addition, the action may not remain localized within the origi-
nal triangle but may activate other triangles, involving more and
more people. Bowen describes how a family in distress can draw in
more and more outsiders.

In periods of stress, the process can involve the entire nuclear family, a
whole spectrum of more peripheral family members, and even nonrelatives
and representatives of social agencies, clinics, schools and courts. In peri-
ods of calm, the process can remain relatively contained within a small
segment of the family, such as the symbiotic relationship in which the
emotional process plays back and forth between mother and child with the
father isolated from the intense twosome. 29

One may ask how the process Bowen describes can be seen to
obey lawful movements. If no alliance remains static from one mo-
ment to the next, isn't this a recipe for chaos? This problem has
dogged family observers for years, since the apparent confusion of
behaviors in families with a high degree of pathology nevertheless
seems to add up to a highly restricted set of choices. Bowen attempts
an explanation by saying that no matter how chaotic triadic behav-
iors in a family may look, these behaviors are nevertheless passing
along very limited and almost preordained pathways. He believes
that when been together for a long time, the process
a family has
of triangling goes through such a fixed chain reaction that a skilled
observer can often predict its stages and, by inserting himself into
the sequence, even control it.
One problem Bowen talks about is the distinction between triadic
processes in families with disturbed members and families that are
presumably "normal." Bowen associates pathology with rigidity
and suggests that, although all families create triadic patterns, these

patterns will become more rigid when the family is facing a change
or undergoing stress, and will be more flexible in periods of calm.
In addition, Bowen continually moved away from a linear defini-
tion of pathology as a condition transmitted through the nuclear
family from one generation to the next. Instead, he became more

30
EARLY RESEARCH ON FAMILY GROUPS
and more fascinated by the evolutionary processes of the larger
kinship group. When working with a family member, he will help
that person trace back the lines of his singular destiny to relation-
ship configurations that may have existed before he was born. A key
to Bowenian thinking is the idea that if someone can achieve a more
flexible position in one family triangle, even a distant one, this can
have positive repercussions on other, closer ones and may even
nullify injunctions from the past that constrict relationships in the
present. Bo wen sees networks of triangles as deeply linked and
reactive to one another. As in a spider's web, a touch at any spot
will vibrate right across the web. Thus, in a family, a change in one
corner may activate unpredictable responses in another corner, and
may help to free persons long caught in static and inhibiting posi-
tions, including the person initiating the change. This emphasis on
the family of origin has had an incalculable influence on the devel-
opment of the family therapy movement.
Bowen presents one particular attribute of families as an index of
pathology: the concept of ''differentiation." He noticed that families
with a psychotic member exhibited an intense, clinging inter-
dependence which he called the "undifferentiated family ego
mass." This is such a ponderous phrase and calls up so many psy-
chodynamic meanings that it can easily be dismissed. But that
would be a mistake. Struggling to refine his concept, Bowen tried
again: "a pre-existing emotional 'stuck-togetherness.' " Later on we
shall find something similar in the idea of "pseudomutuality" that
Wynne uses to describe the gluey quality of the family of the
schizophrenic, and Minuchin's notion of the "enmeshed" family.
What all these observers are describing is a tightly coalesced
triadic structure, and the problemone of wording. A noun like
is

"mass" suggests a lump of stuff all composed of one substance; the

adjective "undifferentiated" suggests that this mass has no pieces or
sub-parts. But "stuck-togetherness" comes a bit closer to the idea
that what we are getting at is a set of interconnecting, mutually
repercussive relationships. If they are "undifferentiated," they are
so in the sense that none of the parts or pieces can move indepen-
dently of the others or the whole.
For Bowen, at any rate, lack of differentiation or "fusion" was a
sign of trouble in a family. By contrast, he postulated that when

31
 Foundations of Family Therapy
individual members maintain a high degree of differentiation, the
family will do well and its members will do well. Differentiation,
7
however, is not to be confused with "emotional cutoff/ which is

a defense against too intense fusion that may take the form of a
family member leaving physically and staying out of contact but
never really freeing himself psychologically. In that sense, emo-
tional cutoff is really lack of differentiation in disguise.
Bowen has a superior feel for triadic processes in families and has
put his finger on an essential aspect of the way they work. Part of
this aspect has to do with the permutations of relationships as they
shift from moment to moment, and the larger schema represented
by the sequences they take. For example, one may find an unaltera-
ble pattern around the "bad" behavior of a child. Stage one: Mother
coaxes, child refuses to obey, mother threatens to tell father (father-
mother against child). Stage two: When father comes home, mother
tells him how bad child has been, and father sends child to his room

without supper. Mother sneaks up after father has left the table and
brings child a little food on a plate (mother-child against father).
Stage three: When child comes down later, father, trying to make
up, offers to play a game with him that mother has expressly forbid-
den because it gets him too excited before bedtime (father-child
against mother). Stage four: Mother scolds father for this; the child,
overexcited indeed, has a tantrum and is sent to bed; and the origi-
nal triangle comes round again (mother-father against child).

Although Bowen himself does not describe triadic sequences in

detail, his view contributes to a "process taxonomy" of
such precise
family interaction. According to his writings, triangles in families
shimmer through their preappointed changes like a light show.
There is an inner, Euclidean logic to this view, and later researchers
and therapists in the family field have justified Bowen's preoccupa-
tion with the triangle by continuing to expand and deepen it.

32
EARLY RESEARCH ON FAMILY GROUPS

Pseudomutuality and the Rubber Fence

Lyman Wynne is another psychiatrist-researcher who started out

with an interest in the thought disorders of schizophrenia and the
influence of family communication style on these disorders. He, like
the Bateson group, took a systems view of the family and not only
noted the redundancies that seemed to be characteristic of families
with a schizophrenic, but backed up Bateson's observations on
shifting coalitions and Haley's on unstable dyads: "In the family of
a schizophrenic, the structure of alignments and splits seems to shift
in a bewildering rapid fashion but with one feature of great con-
stancy: the meaning of any particular alignment does not clearly
30
emerge/' Instead, he noted, these splits and alignments came
across to the observer as unformed, fragmented, and separated from
each other in a "psychological apartheid." The result was that the
alignments did not seem like true closeness, but what Wynne called
"pseudomutuality," and the splits did not seem like real hostility or
distance, but "pseudohostility." Feeling that there were invisible
laws governing the appearance of these displays, Wynne remarked,
"When an alignment has developed within a given family therapy group, look for
an emerging split at another level or in another part of the group; if a split emerges,
" 31
expect an associated alignment to come into view.
Wynne felt that these changes from splits to alignments, back and
forth, had something do with homeostatic maintenance processes
to
in the family, although he did not explain how. He also believed
that these processes took place in any family, but that in a family
with a schizophrenic member they were particularly vivid and pro-
nounced.
As an illustration, Wynne describes a sequence that took place
with stereotyped regularity in a family with two daughters, one of
whom had been diagnosed a catatonic schizophrenic and had been
hospitalized. Betty, the patient, was eighteen; Susan was three years
younger. Wynne and his cotherapist noted two features in the fam-
seemed to be closely related:
ily structure that first, an alignment
between Betty and her father, and second, a split between the
parents. The repetitious sequence that illustrated both observations

33
 Foundations of Family Therapy
was this: During a therapy session the parents would start to quar-
rel, usually over one of Betty's symptoms, like her compulsive
cleaning of the house. The father would take Betty's side, and Susan
would characteristically come in on the side of the mother. As the
argument heated up, the father would start to intensify his atten-
tions to Betty. At these times he might move close to her, even
touching her. If he did, she would often jump abruptly away, in a
move of rejection. Instead of reacting back toward Betty, the father
would turn to the mother and accuse her of being "vicious" and
"mean." Oddly, the mother would refuse to rise to the challenge but
would cut off this exchange with a flat remark like, "Well, that's all
right." The father would lapse into silence and the sequence would
usually be over.
Wynne does not detail Betty's symptoms during these sequences
but says that in general Betty's most psychotic behavior followed
her intense rejections of her father, and he remarks on the cyclical
nature of these interactions. Any rejection of a parent by a daughter
would set the parents bickering, especially a rejection by Betty. But
no matter how loud the arguments became, they seemed to be more
formal than real. The parents never got violent or threatened to
divorce. More important, this pattern seemed to be kept going by
everybody. For instance, even though the mother objected to the
father-Betty intimacy, if it failed to materialize she might remind
them that a father and daughter ought to be fond of one another.
We can see this cycle as a repetition of several interaction states.
One form that polarizes the family, with Susan and mother
is a
aligned against Betty and father. Another would be the father-Betty
coalition against mother, with father protecting and mother attack-
ing Betty. Wynne's example thus gives a graphic picture of how the
daughters' behaviors may have worked as a kind of counterpoint,
mitigating the periodic escalation of hostilities between husband
and wife. Evidence for this idea can be found in the sudden shift
in the triad, as Betty violently rejects her father, shows psychotic
behavior, and apparently triggers off a temporary truce between the
parents.
A striking feature of this family cycle was the apparent agreement
of all concerned to accept a "cutoff point." Animal ethologists use
that term to describe the cessation of ritualized hostilities between

34
EARLY RESEARCH ON FAMILY GROUPS
two males engaged The bickering of the
in fighting or aggression.
parents in the sequence described by Wynne seemed to have the
same ritualized quality, as did their mutual decision to break it off.
Wynne makes the point that the closeness of the mother to peo-
ple outside the nuclear family, notably her own mother and her
employer, frequently replaced Betty as the focus of argument be-
tween the couple. Thus, in describing a cyclical repertory for this
family, one would have to place a mother-grandmother alignment
next to the father-Betty tie. One would expect the cycle to involve
a periodic drawing together of the wife with her mother (employer,
other relatives, and so on). This closeness would presumably coin-
cide with periods of distance between husband and wife. It is logical
to think that this distance would start off the father-Betty coalition,
or perhaps one of the daughters would trigger parental bickering
with a rejecting remark. Either of these moves would bring mother
back into the picture, and some version of the cycle described would
follow.
Wynne's descriptive papers on families with schizophrenics show
that two features struck him with particular force. One was the
strangely unreal quality of both positive and negative emotions, for
which, as we have said, he used the terms "pseudomutuality" and
32
"pseudohostility." The other was the matter of the boundary
around the family. Wynne was the first to comment on its peculiar
nature: an apparently yielding, but actually impervious, barrier
against outsiders (especially therapists). Wynne called it the "rub-
33
ber fence/'
Wynne's explanation of how these features operate in such fami-
lies is that there is an intense wish by family members for mutual

relatedness which excludes the ability to tolerate difference or dis-

sent. The illusion of "pseudomutuality" reinforces the party line
that all are linked together. The "rubber fence" forms a boundary
around this illusion which protects the family from the perils of new
information or potential change. Children in these families are thus
caught in the dilemma of never being able to differentiate or disen-
gage, because any attempt brings on expectations of disaster.
Wynne went on to speculate how the behaviors that produce and
perpetuate these family features might, in a child, create the kind
of thinking disorders displayed in schizophrenia and other affective

35
 Foundations of Family Therapy
disturbances. Denials of thoughts or feelings, inability to judge
objective ''out there" reality, blurring of differences of opinion,
cryptic and fragmented utterances would all be logical ways to
communicate in a family in which one had no basis for what was
7
"rear except validation by other family members, and where the
only "reality" validated would be intense loyalty and closeness. An
attempt by one member at individuation might cause the family to
focus all its fear and dislike of non-mutuality onto him or her,
making that person into a scapegoat. This would reinforce the fam-
ily's primary value of closeness and enslave the negatively per-

ceived person just as effectively as if he had never tried to leave the

fold.
No treatment of the work on schizophrenic communication
should ignore the contribution of British psychiatrists R. D. Laing
and A. Esterson, whose concept of mystification (closely related to
the double bind) was illustrated by a series of studies of schizophre-
nicyoung women carefully observed in the context of their families.
The evidence that the apparently distorted perceptions of these
women were fed and supported by disguised communications from
other family members is forcefully documented in the book based
on these studies, Sanity, Madness, and the Family. 34 The behavior of the
schizophrenic was adaptive; it was a logical response to an illogical
situation.
To describe this behavior, however, most of the researchers we
have discussed used a cybernetic analogy, since families with schiz-
ophrenics (or maybe all families) seemed to possess something like
a homeostatic element that resisted change. Symptomatic behavior
was considered an integral part of this resistance.
The writer who drew most brilliantly upon the cybernetic anal-
ogy was Bateson. Bateson's ideas about the patterning of social
fields,and the cybernetic paradigm he developed to support those
ideas, were to have a unique influence on the family researchers
who were his disciples and those who followed after.

36
Chapter 2

The Dynamics
of Social Fields

Bateson and the Grand Design

One of Bateson's central concerns was what he called "the pattern

that connects/' He believed that at some level of structure there is
a congruence among the laws governing different types of events.
Speaking of his father, a noted British geneticist, he said,

In this early —and as I think he knew —his best work, he posed problems
of animal symmetry, segmentation, serial repetition of parts, patterns, etc.
... I picked up a vague mystical feeling that we must look for the same
sort of processes in all fields of natural phenomena —that we might expect
to find the same laws at work in the structure of a crystal as in the
sort of
structure of society, or that the segmentation of an earthworm might be
comparable to the process by which basalt pillars are formed. 1

One of the elder Bateson's areas of study was the way parts of
organisms differentiate. Some do so serially, down a hierarchical
ladder, like the legs of a lobster; this is metameric differentiation.
Others differentiate symmetrically, with each part exactly like the
other, like the radial tentacles of a jellyfish.

37
 Foundations of Family Therapy
What, one might have lobsters and jellyfish to do with the
ask,
structure of human society? The younger Bateson, looking for a
grand design, thought they might have a lot to do with it. He needed
what he called a "visual diagram," and the analogy of differentia-
tion in biological structures provided him with the form thatcame
closest to explaining one of the problems that fascinated him most
when he started his career: the patterning controlling social segmen-
tation.
It was around the time he was studying the Iatmul culture in New

Guinea in the early 1930s that the problem began to haunt him. He
had gone out like a good cultural anthropologist, equipped with the
proper background (he had been a student of Franz Boas), using the
right tools (notebooks and informants), following the usual proce-
dures (immersing himself in the life of the culture), and generating
appropriate themes through which to interpret his findings (con-
cepts like "eidos" and "ethos").
However, he did one improper thing: He did not stay within the
confines of the universe as defined by anthropologists of that day.
A ceremony caught his eye that did not lend itself either to orthodox
interpretation or to his ingenious categories. This was the naven
ceremony, and the processes it expressed seemed to deal with social
instability: that is, how conflicts and divisions within the group
were handled. The main findings of Bateson's resulting book, Naven,
contributed greatly to his subsequent thinking about social pro-
cesses. 2
The Iatmul of New Guinea were headhunting people living in
villages of two hundred to a thousand inhabitants. Placement in the
various phratries, moieties, and clans was determined by patrilineal
descent, but the lines connecting each family with the mother's kin
were emphasized in less formal ways. What struck Bateson most
forcefully was the absence of hierarchy in this society. There was
no governing body, no chief, no status structure of importance in
any of the villages. If someone wronged someone else, there was no
higher authority to whom disputes could be brought. Clan feuds
might and retaliations would be exacted, either through sor-
result,

cery or through more direct means, like killing someone. At the

same time there was an intense emphasis on rivalry and display. The
clans or initiatory moieties were constantly competing. What

38
THE DYNAMICS OF SOCIAL FIELDS
seemed to happen when rivalries became too intense or quarrels
insoluble was that one of the feuding clans would split off and form
a new village.

Here is where the visual diagram afforded by biological differen-

tiation processes came in. In a later article discussing Iatmul society,
Bateson explains:

Impressed by the phenomenon of metameric differentiation, I made the

point that in our society withits hierarchical systems (comparable to the

earthworm or the lobster), when a group secedes from the parent society,
it is usual to find that the line of fission, the division between the new
group and the old, marks a differentiation of mores. The Pilgrim Fathers
wander off in order to But among the Iatmul, when two groups
be different.

in a village quarrel, and one half goes off and founds a new community,
the mores of the two groups remain identical. 3

Bateson never explored further his idea that hierarchical societies

tend to produce heretical groups when they divide, whereas sym-
metrical groups produce carbon copies. What he looked at more
thoroughly (and what is more germane to this study) is the mech-
anisms countering fission in the Iatmul, chief among which seemed
to be the naven.

The naven was a ceremony, or group of gestures deriving from this

ceremony, which affirmed the attachment between a child, a law,
and his mother's brothers, his waus. Navens might be performed at
important moments in the child's life. They would take place when
he changed from one status to another, or when he achieved certain
culturally approved "firsts," (his first fish spearing or his first homi-
cide). Naven gestures were also elicited by inordinate displays of
pride or boasting by laua to wau. In these cases a naven behavior by
wau would be in the nature of a chastisement, reminding laua that
he was stepping out of line.
A really grand naven could draw in relatives from both sides of the
family. The children's uncles on the mother's side would dress up
like silly old women and treat the children like mock husbands. On
the father's side, the female relatives would don male ceremonial
attire and strut about caricaturing male warlike displays. Iatmul
women enjoyed this all the more because they were normally sub-
ordinate to the men.

39
 Foundations of Family Therapy
At Bateson assumed that the naven operated like a sort of
first

social glue, strengthening the ties to the mother's family of origin

in this very patrilineal culture. But the visual pattern that began to
dominate his thinking was of planes of cleavage crossing each other.
The main split line was between brothers and brothers-in-law and
the clans they represented. Thus any custom that strengthened the
link between them, or counteracted its tendency to break, would be
important.
Along with the effort to see the structure of Iatmul society stati-
cally, in theform of a design, came another effort to see it dynami-
cally, in a state of movement. It was at this point that Bateson

coined the term "schismogenesis." This word described the type of

escalation found throughout the natural world and exemplified by
the vicious circle and called by other researchers "mutual reaction
7
processes/ "deviation-amplifying mutual causal processes," "posi-
tive feedback chains," and the like.

Schismogenesis

Bateson applied his new term primarily to relationships between

people, as opposed to mutual causal processes in general. He defined
it as a "process of differentiation in the norms of individual behavior resulting from
^ "4
These processes are distin-
cumulative interaction between individuals.
guished by the fact that they develop by mutual reaction, exponen-
tially. In Iatmul society Bateson observed the presence of self-rein-

forcing cycles in which the actions of A would trigger B's responses,

which would then trigger an even more intense reaction from A, and
so forth. These cycles could be broken down into two types. One
of these types Bateson called "symmetrical" meaning that the es-
calating behaviors of A and B would be essentially similar, as in
cases of rivalry or competition. The other type he called "comple-
mentary" because the self-generating actions would be different, as
in cycles of dominance-submission or succoring-dependence.
Bateson cites several examples of schismogenic cycles. One is the
self-reinforcing-relationship sequence that results in some types of

40
THE DYNAMICS OF SOCIAL FIELDS
mental illness. Bateson specifically mentions paranoia, in which the
patient, being distrustful, triggers responses in others that have the
result of justifying his fearsand making him even more distrustful.
Another example is the type of marital maladjustment that results
when one partner is extremely assertive and the other very compli-
ant, and these characteristics become progressively accentuated, the
one partner becoming more and more compliant as the other
becomes more and more assertive.
Bateson notes that these processes can occur in arenas other than
interpersonal ones. He mentions the way culture contacts between
two societies can lead to special arrangements that can be either
symmetrical or complementary, the symmetrical spirals represented
by arms races and the complementary ones represented by tensions
between social classes.
Bateson also suggests that there are two aspects to these self-
reinforcing processes. There is the exponential process that is self-

stopping or gets stopped, and there is the exponential process that

does not get stopped but does not destroy the system. Of particular interest
is the escalation that passes beyond the limits of the previous ar-
rangement, apparently rushing toward ineluctable doom, and then
emerges with a transcendent synthesis nobody had foreseen.
In this chapter we will be addressing ourselves to the two types
of exponential process described by Bateson and to the principles
that govern not only self-stabilizing escalations but the ones that
threaten to get out of hand. In addition, we will examine the ideas
of other researchers who have puzzled over these curious self-rein-
forcing cycles and devised some useful theories about them.

Mutual Reaction Processes

Kenneth Boulding, in his Conflict and Defense, investigates the prop-

erties of self-reinforcing cycles using the term "mutual reaction
processes/' 5
He devotes a good part of his treatise (published in
1962) to an analysis of formulas presented by the British political
scientistand mathematician L. F. Richardson in a 1939 study of

4*
 Foundations of Family Therapy
armament races and international hostilities with the unusual title
7
'The Statistics of Deadly Quarrels/ Richardson had devised math-
ematical equations to express escalating hostilities between nations,
dominance-submission patterns, and the like. Boulding translates
these formulas into graphs, which he calls "Richardson Process
Models."
Boulding is same thing Bateson is talking
clearly talking about the
about: processes in which a movement by one party changes the
field of the second, forcing a compensatory move by the second

party, and so on. In economics, Boulding says, this process is exem-

plified by the price war; the political scientist knows it as an arma-
ments race; students of relationships find it in competitive clashes
between husbands and wives; labor movement people see it in the
struggles between union and management; and it is even found in
the animal kingdom, in the relationship between predator and prey.
Boulding is mixing interpersonal and noninterpersonal processes
here, and does not make use of Bateson's distinction between sym-
metrical and complementary differentiation, but he does show rep-
resentative cases of both. One graph depicts a symmetrical escala-
tion in which two parties intensify hostilities until they reach what
7
Boulding calls an "equilibrium point/ which place the activity
at

stops. According to Boulding, who has taken the idea from Richard-
son, this point represents an intersection where aggression is can-
celed by some increasing factor like war weariness or fatigue. In
another graph depicting this situation, the escalation mounts indefi-
nitely, but Boulding says this could not happen in real life because
there would be some boundary which could not be passed without
the system breaking down or the activity shifting to a new form:
arms races to war, marital quarrels to divorce, and so on.
In the different case of a dominance-submission cycle, Boulding
follows Richardson in graphing a situation analogous to the preda-
tor-prey relationships in sparsely inhabited terrain. In the Arctic, for
instance,where wolves subsist on rabbits, an increase in the rabbit
population means more wolves. But the increase in the wolf popula-
tion begins to erode the rabbit population. This in turn reduces the
wolf population until the rabbit population builds up again.
The same situation often holds true in human relationships,
where one person is apparently dominated, the other the one who

42
THE DYNAMICS OF SOCIAL FIELDS
dominates. The dominant person will increase his power over the
submissive one until the submission reaction becomes so extreme
that it no longer stimulates a dominance reaction. The submissive
person will then become increasingly assertive until the dominance
reaction is triggered off again and the cycle repeats.
The terms "dominant" and "submissive" are unfortunate because
they suggest a power struggle rather than a systemic sequence that
neither person has the power to resist. The beauty of the concept
of complementary escalation is that it avoids the tendency to see
such struggles in terms of individual motivation. In fact, as Boulding
points out, such arrangements tend to have a pattern of circularity,
moving about a point of equilibrium. The graph which depicts them
form of a spiral. Although Boulding does not take
fittingly takes the
up the question of complementary mutual reaction processes get-
ting out of hand, examples can be imagined: parents disciplining
their children to the point of injuring them or worse; "master races"
exterminating "inferior" ones in the name of racial purity.
In a 1949 essay on Balinese culture, Bateson, too, contemplated
the implications of Richardson's equations, which he saw were in
some sense mathematical expressions of his concept of schismo-
genesis. 6 For him, as with Boulding, the mysterious question was:
What puts a stop to such processes, since their very nature is to
escalate? We are back to an examination of contrary movements: a
sequence with a rise and fall which is self-stabilizing, versus a
sequence with an escalation that mounts until some breaking point
is reached or some other event occurs that checks the rising curve.
In other words: deviation-counteracting sequences versus devia-
tion-amplifying sequences.
Pondering the Richardson equations, Bateson was dissatisfied
with Richardson's assumption that if a symmetrical escalation
reached an equilibrium point, or reversed itself, this would be due
to some such factor as fatigue. Instead he posited another possible
answer: the dual need for tension arousal followed by tension re-
duction, which is characteristic of many organisms and is expressed
in activities such as fighting or making love.
Bateson investigated other kinds of stops to schismogenesis, stops
that were not physiological but based on external social constraints.
He pointed out in Naven that one factor that prevents a "runaway"

43
 Foundations of Family Therapy
is found in the nature of the relationship between two parties.
to be
If there is enough mutual dependence between two complementary

parties, one of whom is stronger than the other, differentiation will

never proceed beyond the point where schismogenic tendencies are
countered by mutual dependency needs. Similarly, a symmetrical
escalation may be held in check by reciprocal arrangements based
on the interests of both parties, like a barter agreement or my own —
—
1980s examples the holding of hostages or nuclear warheads.
Bateson also discusses a variant in which one complementary
process might counteract another. He takes as an example a mar-
riage in which the tensions arising from a relationship based on
assertion-submission are relieved by a shift to a relationship based
on caretaking dependency. 7 The relationship is still complementary,
but the values it represents shift from negative to positive. Symp-
toms like depression or a psychosomatic illness in a spouse may be
a response of this kind, or may signal the cutoff point for an asser-
tion-submission sequence that has gone too far.

There are larger, societal checks to schismogenesis. Bateson notes

that participating in a national celebration may lessen interclass
tensions; in the same way, a war may unite a country divided
against itself. He also felt that the custom of headhunting among
the Iatmul, which pitted the society against other groups, may have
lightened the interior stresses due to rivalry and competition.
The anthropologist Fredrik Barth describes a different example of
a social checking mechanism in his study of the way the fiercely
warring Pathan clans of Afghanistan are prevented from totally
exterminating one another. 8 He notes the deterrent effect of a minor
chieftain whose support can always tip the balance between two
more powerful chiefs. If who needs a lesser man's support
a chief
to win against a rival chief puts too much pressure on his rival, the
rival can offer the lesser man enough concessions, including his own
chieftainship, to get him to defect. Knowing this, the first chief will
never threaten to wreak as much vengeance on his vanquished
enemy as he might if the third party were not there.
These examples, however, do not suggest what laws or guiding
principles may be governing these mysterious action/reaction pro-
cesses, or the checks which arise to stop them. Bateson's contribu-
tion was a growing suspicion that there might be some inbuilt,

44
THE DYNAMICS OF SOCIAL FIELDS
self-equilibrating arrangement in social groupswhich kept schis-
mogenic movements under control. In his study of the Iatmul he
was already mulling over the idea that symmetrical and comple-
mentary types of schismogenesis might operate in a mutually coun-
teracting way. For instance, he posited that a small dose of symmet-
rical behavior in a complementary relationship might act to check

the tendency toward progressive differentiation. He took as an ex-

ample the relationship of an English squire to his villagers, which
is a complementary and not always comfortable one. If the squire

played cricket with the villagers once a year, Bateson noted, this
small action might be enough to ease the strain.
Later, discussing Richardson's formula for rivalry, which states
that the intensity of A's actions is proportional to the degree to
which B is ahead of A (B — A), Bateson observed that the appropri-
ate formula for a complementary progressive change would be the
opposite, since A's actions would be proportional to the degree to
which B is behind A (A — B). Thus he says,

Notably this formulation is itself a negative of the formulation for rivalry,

the stimulus term being the opposite. It had been observed that symmetri-

cal sequences of actions tend sharply to reduce the strain of excessively

complementary persons or groups. It would be tempting to ascribe this
effect to some hypothesis which would make the two types of schismo-
genesis in some degree psychologically incompatible, as is done by the
above formulation. 9

But what made the greatest impact on Bateson's efforts to clarify

his thinking about mutual causal processes was his discovery of
cybernetics or the science of self-correcting systems like ser-
vomechanisms. In the 1958 Epilogue to Naven, he describes the
influence upon his thinking of the Josiah Macy, Jr., Conferences
which were held during the 1950s and at one of which, in Princeton
in 1955, he was asked to speak. 10 It was during these meetings that
he became seriously intrigued with the ideas of Claude Shannon,
Norbert Wiener, and other cybernetic theorists and began to rethink
his concept of schismogenesis in terms of the error-activated feed-
back cycles found in self-governing systems.
The analogy Bateson found most useful was that of a steam
engine with a governor. The whole arrangement depends on a loop

45
 Foundations of Family Therapy
in which the more there is of something, the less there is of some-
thing else. In contrast, when the situation is such that the more there
is of something the more there is of every other element, you have
what systems engineers call a positive feedback chain or a "runa-
way." In a steam engine, if the governor were constructed so that
the more the arms diverged the more the supply of steam increased,
this would cause the engine to operate faster and faster, until it
exhausted the available amount of steam or its flywheel would
break. A third possibility would be some form of external restraint
— for instance, representatives of the next system up (the engineers)
might come in and prevent a breakdown by stopping the machine.
Bateson saw the potential of this model in explaining the behav-
iors he had observed while he was studying Iatmul culture. In par-
ticular he was now able to analyze the baffling naven ceremony as
if it were a loop of behaviors keeping certain variables of the social
system within bounds.

The Naven Ceremony as a Cybernetic Mechanism

Noting the forces promoted vicious cycles

in Iatmul society that
leading to either schism or war, Bateson asked what checks existed
to prevent these outcomes. With self -corrective circuits as a model,
Bateson believed an answer was now possible. Might there not exist
in this society a system by which appropriate complementary
behaviors would bring about a corrective decrease in symmetrical
behaviors? Could not the system be self-corrective in a circular
fashion? Bateson reviewed his findings on Iatmul culture in this
light.

was the overriding structural fact that the weak-

First of all, there

est links in Iatmul society were those among maternal relatives.

These were broken when a group seceded from its community, since
women went to live in the clan of the man they married.
Bateson quite naturally asked why, if the naven ceremony was to
be understood as providing integration in this area, the emphasis
was on strengthening uncle-nephew ties rather than the brother-

46
THE DYNAMICS OF SOCIAL FIELDS
brother-in-law ones. He surmised that it was due to the importance
of stressing complementary links as a corrective to symmetrical
escalations threatening the stability of the group. Clan rivalry made
the possibility of fighting or fission a constant threat, and the naven
seemed to occur when the delicate balance between competing clans
was about to tip. Thus Bateson observed that the naven ceremonial,
which is a caricature of a complementary sexual relationship, was
set off by overweening (symmetrical) behavior. When a laua boasts
But the full
in the presence of wau, the latter exhibits naven behavior.
naven occasion takes place in the context of a step toward vertical
mobility on the part of the laua, as when he becomes a successful
warrior or hunter. This would be even more of a symmetrical escala-
tion, not only as regards wau, but in relation to wau's entire clan.
However, Bateson was not satisfied with a vague hypothesis of
one schismogenic process intensifying until a corrective contrary
process was set in motion. He wished to find a more specific way
to explain the mechanics of this arrangement. This led him to cyber-
netics, and to the use of a concept that seemed to be crucial to an
understanding of the way cybernetic systems worked.

First and Second Order Change

Up to now, Bateson and other researchers had seen two ways in

which deviation processes might operate: a self-stabilizing se-
quence, which the naven typifies, and, alternatively, an escalation
leading to destruction of the system. But there is a third possibility,
which is that a runaway or escalation can set a leap in motion that
may transform an entire system.
According to systems theorist W. Ross Ashby, this third possibil-
ity is explained by a two-level model for change. Living systems,
Ashby noted, are not only able to vary their behaviors in response
to minor variations body keeps within an opti-
in the field (as the
mum range of temperature by perspiring when it is hot and shiver-
ing when it is cold), but are often able to change the "setting" for
the range of behaviors whenever the field presents an unusually

47
 Foundations of Family Therapy
serious disruption (as in animal species that developed the capacity
to grow thicker fur when winters became colder, or worked out a
pattern of migrating to warmer climates until spring). 11
This type of "bimodal" feedback is Ashby, because
useful, says
it enables the entity or organism to survive both day-to-day fluc-
tuations and drastic changes. He called the corrective responses to
minor fluctuations "first order change" and the responses to drastic
differences in the environment "second order change." The analogy
most often used to illustrate this distinction is the homely house
thermostat. The automatic shifts it makes to keep the room within
a certain temperature range are first order changes. To make a sec-
ond order change, however, as when the outside temperature falls

precipitously, the householder must alter the setting of the thermo-

stat.

Bateson was looking for factors that would control the potential
runaways of schismogenesis, but he also realized that schismogenic
processes could be useful in breaking up an inappropriate, out-
moded, or unhealthy stability. In his Epilogue to Naven, he paid
tribute to Ashby's formal analysis regarding change in steady-state
systems.
Let us follow Bateson's lead in applying these ideas in detail to
the workings of Iatmul society. For instance, one variable of critical

importance to the society was intense rivalry. In the absence of any

hierarchical structure for resolving conflict, the balance of power
between clans had to remain relatively even. If one or another clan
gained even a small advantage, it had to be counteracted before an
escalation started that might get out of hand. The all-out naven
ceremony may be seen as a first order change, one which would
substitute symmetrical escalations for complementary ones. As the
latter is incompatible with the former, this would effectively ham-

per the development of symmetrical runaways, and maintain the

status quo.*
But what if were set off which the usual
a positive feedback chain
ceremonies were unable to control? I have surmised that it is possi-
ble that splitting is a second order change that comes about when

*Bateson, unfailingly circular, allowed also for the possibility that too much complemen-
tary behavior might trigger off symmetrical displays.

48
THE DYNAMICS OF SOCIAL FIELDS
complementary themes are unable to check the escalation of mutual
belligerence. Among the Iatmul, a group will go off to form its own
village whenever tensions above a certain point. Bateson notes
rise

the tendency of the Iatmul to expand by a proliferation of small

offshoots, each resembling the parent body but not connected to it.
In this way Iatmul society survives.
Unfortunately, the deviation-counteracting effect of a self-stabil-
izing circuit that will create a first order change is, like a razor,

double-edged. The naven ceremony confirms the group in its old, set

ways, and thus has long-range implications for the weakening of

the group due to loss of flexibility and risk of error. It is at this point

that the biologists offer a comforting vista with their talk of "variety
pools" and the role of deviance in forcing new solutions. To quote
another systems thinker, Roger Nett:

Since the creative strength of a society must be sought in the capacity of

individuals to evaluate, extend, correct, and ultimately to alter existing
definitions and understandings (a process which is, in effect, deviation) the
problem of ordering a society becomes one of utilizing the vital element
—deviation—in social-organizational context. 12

In the next chapter we will explore further what writers with a

systems orientation have to say about the forces that promote diff-
erentness and those that promote sameness in living systems in
general, not just human systems. The emphasis will be on what the
sociologist Magoroh Maruyama calls the "Second Cybernetics."
Since most theorists of family relationships have focused on the
"First Cybernetics" —deviation-counteracting processes and nega-
tive feedback chains —Maruyama suggests that more attention be
given to this "Second Cybernetics," which he sees as an essential
aspect of change in all living forms.

49
Chapter 3

The Second
Cybernetics

Morphostasis and Morphogenesis

Magoroh Maruyama believes that the survival of any living sys-

tem — that —
any self-maintaining entity depends on two impor-
is,

tant processes. One is "morphostasis," which means that the system

must maintain constancy in the face of environmental vagaries. It
does this through the error-activated process known as negative
feedback. The other process is "morphogenesis," which means that
at times a system must change its basic structure. This process
involves positive feedback or sequences that work to amplify devia-
tion, as in the case of a successful mutation that allows a species to
adapt to changed environmental conditions. 1
The phenomenon of positive feedback has usually been looked
atfrom the point of view of its destructive effects on a given system.
Norbert Wiener discusses it in terms of mechanisms like control
feedback elements in antiaircraft guns, observing that if the feed-
back element pushed beyond some optimum point, it
is will begin
to overcorrect, making wider and wider arcs until the oscillation

SO
THE SECOND CYBERNETICS
2
causes the machinery to break down. Garrett Hardin, a biologist,
analyzes the same process as it applies to social systems. 3 Describing
a number of homeostatic models, both man-made and natural, he
doubts that a true homeostatic system can ever operate freely in
human affairs, because of the tendency for vested interests to build
up. Social power, he observes, is inherently a process of positive
feedback. And herein lies danger. All systems, according to Hardin,
possess a "homeostatic plateau" —limits within which the system is

self-correcting —but beyond the homeostatic plateau, at either ex-

treme, lie and destruction.

positive feedback
Both Hardin and Wiener have the bias of many thinkers who are
grounded in communications theory: They equate any movement
toward randomness or chaos with something undesirable.
Maruyama, along with systems theorists like Walter Buckley and
Albert Speer, believes such movement can have positive value. In
addition to examples of destructive positive feedback cycles, he
cites others that increase the survival potential of a given system.
Thus he gives a cybernetic framing to an evolutionist theory of
deviation that embraces change of any sort.
Another way to look at these two processes is in terms of what
Buckley, following Ashby, calls "variety" and "constraint." Con-
synonymous with pattern, structure, regularity. It goes
straint is
away from a random state, toward what the systems theorist Erwin
Schroedinger calls "negentropy." 4 No living system could survive
without pattern or structure. On the other hand, too much struc-
ture, too much "negentropy," will kill it. must
This is why there
always be, as Buckley explains it, "some sources of mechanism for
variety, to act as a potential pool of adaptive variability to meet the

problem of mapping new or more detailed variety and constraints

in a changeable environment." 5
One good example of too much negentropy is described in a
recent paper by family researcher David Reiss on family "para-
digms" (blueprints, in effect, for facing new situations or making
sense out of old ones). 6 Reiss speaks of one family that recently
migrated from Europe to the United States and encapsulated them-
selves in an apartment encrusted with objects from the past, com-
plaining about the difficulties of life in a dirty, crowded city. Be-
cause of their reliance on a strategy that did not carry within it the

5*
 Foundations of Family Therapy
capacity to shift to meet new circumstances, they were unable to
findnew avenues for enjoying themselves. A contrasting family,
which saw new and difficult events as an opportunity rather than
a threat, solved the dilemma posed by traveling in a foreign land by
one daughter, who knew the language, be with the
insisting that
family whenever transactions with the natives were required.
As described two types of feedback that which wel-
so far, the —
—
comes and that which inhibits change would seem to have oppo-
site functions. Negative feedback is conservative and promotes the

status quo; positive is radical and promotes newness. But this is far
from the whole story. Buckley, in talking about the ''vicious circle
or spiral or escalation," says that "it is not at all certain whether the
resultant will maintain, change or destroy the given system or its

particular structures." 7 One can think of examples: the growth of

monopolies might lead to such total inequity that social revolution
would result, or it could inspire a movement toward antitrust legis-
lation. The death of a religious or political heretic might reinforce
the system he repudiated, or his martyrdom might lead to a revision

of the entire social order. The death or suicide of a family member

might close off possibilities for change in the family, or it might
release unexpected potential for growth.
Maruyama paints an even more complicated picture. He points
out that it is possible to have both positive and negative mutual-
causal loops counterbalancing one another in any given situation
(by "loops" he means a series of mutually caused events in which
the influence of any element comes back to itself through other
elements). As an example he diagram showing forces
offers a vector
and counterforces impinging on the growth of a city. Factors such
as number of people, migration level, modernization, sanitation
facilities, amount of garbage per area, bacteria per area, and number

of diseases form a number of interrelated positive and negative

loops that increase or decrease the population.
Maruyama does not, alas, make any suggestions about how to
predict outcomes from this interplay of loops. He leaves us merely
with the statement that "an understanding of a society or an orga-
nism cannot be attained without studying both types of loops and
the relationships between them." 8

52
THE SECOND CYBERNETICS

Timing and Stages

But there is another way to look at feedback loops. What is

important is not only the relative strength of these loops and the
way they are combined but timing. In assessing a self-correcting
system, a particularly important factor is the balance or unbalance
of the system at any given time. This is, of course, what Jackson and
other family therapists realized when they tried unsuccessfully to
introduce changes into families that were not in crisis and why at

times they would deliberately try to make the family "system"

exceed its runaway.
limits or create a
Family therapist Salvador Minuchin, for instance, describes the
value of inducing a crisis in the case of a family with an asthmatic
He decided to ask the father, who was usually careful to
daughter.
bow down to the expectations of the mother, to come home unex-
pectedly late one night. The father chose a Friday night before a
weekend trip for his experiment, and the mother, a mild-mannered
woman, went after him with a pair of scissors. After this, the focus
of treatment moved and to the other brothers and
to the parents
sisters, and the asthmatic daughter began to improve. 9

Another aspect of feedback processes which is related to timing

is that they often occur in alternations or stages. Bateson calls atten-
tion to "inverse progressive changes," as when an increase in mutual
between a couple reaches some built-in limit and a shift
hostility
toward mutual affection (also limited) takes place. 10 Such an oscilla-
tion usually implies an overall stability.
Maruyama describes a different sort of situation, in which a
deviation-amplifying process can, over a period of time, change into
a deviation-counteracting one. Here there is a drift toward increas-
ing differentiation which at some point somewhat adventi-
loses its

tious nature and stabilizes. One example (my own) is the way many
couples nowadays start living together under the comfortable im-
pression that they can always leave. Sooner or later they find that
time and habit have placed them in as binding a relationship as any
marriage.
Naturally, the process can go the other way, with a formerly

S3
 Foundations of Family Therapy
stable system moving into a period of disequilibrium. Deviation-
amplifying chains characteristic of this sequence seem to divide
along the lines of Maruyama's and Hardin's different ideas about
the nature of positive feedback. There is a gradual process by which
a variation takes hold,and the runaway that develops when a sys-
tem's error-activating mechanisms break down. Of course the two
types of positive feedback may not be related —for instance, a drift
toward deviation may occur all by itself and not in connection with
—
any systemic entity but they may also be stages of a larger process.
An example is the behavior of animal populations that live in an
environment where there an unlimited food supply and few com-
is

peting species to act as a natural check to increase. These popula-

overexpand in the manner of the first type
tions periodically start to
of positive feedback. At a certain point they will suddenly adopt a
—
self-destroying behavior like the lemmings famous march to the
7

sea —as if a limit in the "plateau" regulating their numbers had been
reached and triggered a runaway (literally) to destroy the excess.
But here a difficult question presents itself. Can this sequence
really be said to go in the direction of a deviation-amplifying result?
From the point of view of the lemming population as a whole, if not
from that of the subgroup that gets destroyed, the entire series of
events has operated to reinstate the status quo. But it is also possible
that a deviation-amplifying process can bring about a leap toward
a new and more complex state. At this point we must bring in the
concept of levels.

Levels

So far, we have been looking at the effect of feedback loops on

one particular system. What we need to consider next is that feed-
back processes in living systems must always be viewed in terms of
several levels of systems at once. The fact that there is a hierarchy of living
systems is not a new discovery, although it is not always related to
cybernetic theory. To prevent confusion, let us make it clear that we
are not talking about levels in a descriptive or epistemological sense

54
THE SECOND CYBERNETICS
(as in Levels of Logical Type) but in a structural sense, as in layers
of living systems. The social scientist Herbert Simon has offered the
thought that the complexity of natural phenomena could be better
understood if we realized that we are always dealing with "layers"
or "nests of Chinese blocks," in sequences of increasing inclusion,
such as: individuals, primary groups, organizations, social systems;
or, in biology: gene, cell, organ, organism. Any activity in one of
these layers will obviously be operating simultaneously in at least

one other. 11
Bateson, always aware of levels, makes a similar point when he
notes that the study of interaction always involves at least two
pieces of information, "a statement about participating entities and
a statement about that larger entity which is brought into being by
the fact of interaction." 12
He then adds that an important source of
destructive interaction can be a discrepancy between the goals of
two systems on different levels:

For example, a self-maximizing tendency may lead to the destruction of

some larger system which was instrumental and necessary to the existence
of the self. In special cases, the self-destruction of the smaller entity is
13
instrumental to the survival of the larger system.

The struggle of social theorists to discriminate between an open

aim and an unintended consequence in human events, embodied in
terms like overt/covert and manifest/latent, may in many instances
have to do with this same fact: that any action taking place in a
social field will touch at least two contiguous systems. The sociolo-
gist Robert Merton comes close to suggesting this in an essay on
manifest and latent function. Listing Durkheim as one of many
thinkers who have, without realizing it, used a concept of latent
function, he observes that "Emile Durkheim's similar analysis of the
social functions of punishment is also focused on its latent functions
(consequence for the community) rather than confined to manifest
functions (consequence for the criminal)." 14
Examples of behaviors that affect more than one system at a time
are constantly cited by family researchers; in fact it was the surmise
that the individual sufferer from mental illness could not be under-
stood without looking at the consequences of his distress for the
family group that started family therapy on its way. Following this

55
 Foundations of Family Therapy
line of thought, Haley observed the double consequence of symp-
toms of all kinds: the effect on the individual, which was to make
him less responsible and more helpless; and the effect on his family
relationships, which was to give him a lever for enormous control. 15
The Bateson group used the terms "overt" and "covert," to distin-
guish between openly acknowledged behaviors and those whose
consequences were involuntary or denied.
Holding fast to this concept of structural levels, we can now see
that any feedback may have deviation-amplifying and deviation-
counteracting effects at the same time, depending on which system
one is looking at. Tragic drama is suggestive in regard to this point.
What the Greeks called hubris, translated as overweening pride and
linked to the tragic hero's downfall, closely resembles our old friend
the positive-feedback chain of social power. Once set in motion,
this chain is deviation-amplifyingfrom the point of view of the hero,
whose group is increased to the point that
deviance in relation to his
he is eventually cast out, brought low, or otherwise destroyed. It is
deviation-counteracting from the point of view of his society, in that
out of the ashes of the hero's downfall supposedly rises a new social
peace.An alternative explanation might be that the society uses the
aftermath of the debacle to recalibrate the setting for its own equi-
librium. Thus a tragedy may essentially describe a "morphogenetic"
change (change in the homeostatic setting) rather than a "morpho-
static" change (change governed by the homeostatic setting).
Whichever way we define what is going on, it is clear that with-
out some such multilevel view, we will not begin to understand it.

Such ideas allow us to think differently about the process that links
the deviant and his group. The social typecasting of deviants thus
emerges as an area where two streams of thought, one from general
systems theory and one from sociology, begin to intersect.

The Typecasting of Deviants

Most studies of deviance read like contributions to the sociology

of occupations, except that "occupation" is extended to cover delin-

56
THE SECOND CYBERNETICS
quency, mental illness, and the like. However, a respectable few are
concerned with the circular causal processes that increase the differ-
ence between a person and his group so that he is perceived unfa-
vorably.
Leslie T. Wilkins, in "A Behavioral Theory of Drug Taking/'
explicitly applies Maruyama's mutual causal concepts to the type-
casting process. 16
He explains how an addict "outlaw group" is

created through the mutually reinforcing effect of social definition

on self-image, and how the further step of placing addicts together
in rehabilitation or detention centers amplifies their difference from
the community (and likeness to each other) even more. Using this
framework, Wilkins makes an intelligent criticism of systems of
control that are directed toward changing the individual deviant
rather than the process which creates him.
Family theorists have also developed a literature on deviance
focused on the symptomatic member of a family but they often —
use the concept of scapegoating rather than typecasting. This word-
ing brings up an interesting problem. It is easily noticed that scape-
goating ismutual causal process, despite the implication that the
a
scapegoat is the victim and everyone else is taking advantage of
him. Nevertheless, the word is weighted in favor of the underdog
and is hard to use objectively. Perhaps this is why most research on
deviance does not employ it.
Nevertheless, Ezra Vogel and Norman Bell's "The Emotionally
Disturbed Child as a Family Scapegoat" is an excellent contribution
17
to this literature on family deviants. According to the authors,
emotionally disturbed children are invariably involved in tensions
between their parents. The by projecting their conflicts
parents,
onto the child, maintain a reasonably harmonious relationship, but
the cost to the child may be great. A major contribution of the article
is a description of the way the child is selected and then inducted
into his role. Some chance characteristic of the child — need not
it

even be an obnoxious trait so long as it differentiates him —

will be
singled out and then developed, increasing the contrast between
himself and everyone else in the family. The authors do not see this
as a mutual causal process in Maruyama's sense, but it fits the
definition very well.
Most writing about typecasting and scapegoating has one of two

57
 Foundations of Family Therapy
emphases. The first is on behavior as it functions on different levels
of systems. Researchers studying the family as a system and most
of the contributors to the sociology of deviance write from this
standpoint. The other emphasis, based on traditional, individual-
oriented ideas, gravitates to concepts like "projections," "expecta-
tions" or "roles." Into this category falls much clinical writing on
such behavior as delinquency and mental illness.

From the point of view of family therapy, the individual-oriented

approach badly misrepresents the subject. For instance, to speak of
the "role of the scapegoat"is to present the deviant as a persona with
fixed characteristics rather than a person involved in a process.
"Scapegoating" technically applies to only one stage of a shifting
scenario —the stage where the person is metaphorically cast out of
the village. After all, the term originates from an ancient Hebrew

ritual in which a goat was turned loose in the desert after the sins
of the people had been symbolically laid on its head. The deviant
can begin like a hero and go out like a villain, or vice versa. There
is a positive-negative continuum on which he can be rated depend-

ing on which stage of the deviation process we are looking at, which
sequence the process follows, and the degree to which the social
system is stressed.
At the same time, the character of the deviant may vary in an-
other direction, depending on the way his particular group does its

typecasting. Which symptoms crop up in members of a group is

itself a kind of typecasting. Thus the deviant may appear in many

guises: the mascot, the clown, the sad sack, the erratic genius, the
black sheep, the wise guy, the saint, the idiot, the fool, the impostor,
the malingerer, the boaster, the villain, and so on. Literature and
folklore abound with such figures.
It will thus make any study of deviance whether
a difference in
one chooses by phase or
to depict one's subject in cross-section,
type, or to trace it in terms of a shifting, longitudinally viewed
career. Cne of the problems with early family research was the
attempt to see a typology of families according to symptom: the
"schizophrenic" family, the "delinquent" one. More recently it has
been realized that what is needed are longitudinal studies that show
the varying nature of the family deviant over time, or the use made

58
THE SECOND CYBERNETICS
of different family members in changing succession, as need arises

and persons offer themselves.

The Meaning of Deviance for Social Systems

One view of the meaning of deviance is that it promotes cohesion.

Most writerson the sociology of deviance agree with Emile Durk-
heim that the deviant's main function for the group is to promote
solidarity and highlight rules and norms. A good summary of this
position can be found in R. A. Dentler and Kai T. Erikson's 'The
Functions of Deviance in Groups." 18 Arlene Daniels, in an essay on
scapegoating in a sensitivity training group, makes the further point
that this morale-building function seems to come to the fore in
groups in which expressions of anxiety and hostility are encour-
aged. 19
Another view emphasizes the danger to society. In the same arti-
cle Daniels remarks that the process of scapegoating may only serve

to keep an unworkable system functioning long after it should have

collapsed. Many an outmoded group or sect owes its longevity to
this fact.
A third view of deviance emphasizes its mediation function in
situations where people are in conflict. Family theorists have ob-
served that the presence of a deviant may be vital to the manage-
ment of conflict. Researchers studying or working with families of
schizophrenics were amazed at the way the patient's symptoms
would flare up when attention was focused on some crucial dis-
agreement, particularly one between the parents. Jackson felt that
this type of activity served a diversionary purpose and often termed
20
it a "rescue operation." Family therapy studies, as typified by the
Bell and Vogel article, almost uniformly emphasize the way in
which the parents of an emotionally disturbed child, who are often
in serious, if unadmitted, conflict, can unite around their common
involvement with the child. Thus the hostility between them is
submerged and a surface harmony prevails.

59
 Foundations of Family Therapy
Observations like this led to the belief that whatever else "schizo-
phrenia" might be, it was always associated with a potential split

in the family. Haley speculatively redefines schizophrenia as a

"conflict of groups" and suggests that schizophrenia is a name for
behavior that results from mediating in many warring family trian-
gles. 21 In families where profound differences exist between the

—
parents which often means between whole kin groups a need for —
family unity leads to rewards for ambiguous communication that
keeps the peace. Certain persons will be singled out to handle this
task. Such persons will not make sense to outsiders, and may even
be thought mentally ill, but within the family context this ability
not to make sense is encouraged, presumably because it helps the
family to stay together.

Irresistible Runs

In talking about homeostatic mechanisms, writers like Hardin

assume that if a homeostatic plateau is exceeded, a disastrous devia-
tion-amplifying process will set in and destroy the system. Yet in
families that come into therapy one finds constant small "runs"
positive feedback chains that look as if they might turn into runa-
ways but never do.
These runs may be escalating arguments like those Jackson saw
in his work with what he called (following Bateson) "symmetrical"
couples. 22 Or they can take place in a marriage where husband and
wife have defined their roles in a complementary manner, with one
seeming to be "strong" and the other "weak." Such a pair, as Jack-
son says, "can be viewed as a mutual causative system, whose
complementary communication reinforces the cycles of interaction
between them." 23
One can also see spirals of hostility between parents and a child.
William Taylor cites such an interchange from Haley and Hoffman's
an example of "recurrent states" in
Techniques of Family Therapy as
24
family interaction. Runs like these are nearly irresistible, as any-
one who has ever been caught in one or watched one knows, and

60
THE SECOND CYBERNETICS
they repeat and repeat like a broken record, never reaching any
conclusion.
What are these forms and why do they occur? The answer may
well be that they are a response to a system that is constantly
threatening to exceed a variety of homeostatic limits. Why else

would there be so many positive feedback chains that abort? It is

even possible that the many redundancies of communication noted
by researchers in families with disturbed members, particularly in
the disturbed child/parents triangle, are all feedback chains of this
type.
something in family sys-
This explanation implies that there is

tems resembling homeostatic plateaus. Although no family re-

searcher has, to my knowledge, used the idea of such a plateau in
interpreting family behavior, the sociologist Robert Bales has used
it in speaking of small-group behavior. Although he finds that a
successful meeting need not necessarily have fewer negative than
positive reactions by group members, he remarks that there seems
to be a kind of optimum balance on which success depends. A
departure to either side will bring trouble. Bales finds this most true
when there is an excessive rate of disagreement: "Apparently, when
ill feeling rises above some critical point, a 'chain reaction' or 'vi-

cious circle' tends to set in. Logic and practical demands of the task
25
cease to be governing factors." In fact, says Bales, when they reach
this point groups can hardly accomplish anything.
In the same way, there seems to be a certain range within which
family functioning is maintained. In families in which a member is

deemed "abnormal," this range may be an extremely narrow one.

Dr. Richard Fisch, of the Mental Research Institute in Palo Alto, has
used the term "the ten-foot pole" to describe the narrow range of
closeness and distance that seems to limit the relationships of some
of the couples who come to him for therapy.
up an example of the ten-foot pole in action. Assuming
Let us set
that the relationship between the parents is one of the most impor-
tant variables in the family, the "setting" for closeness and distance
between them would be very important. Suppose a husband is
always exceeding the setting for distance from his wife because of
a prior setting for closeness with one of his parents. One might then
expect to find in the family a recurrent mutual causal sequence of

61
 Foundations of Family Therapy
some sort. This sequence may take the form of the bickering that
Jackson noted, when each spouse, feeling the "victim" of the other,
provokes the very hostilities that justify his or her own. Or it may
take the form of an increasing withdrawal, with distant behavior
provoking even more distant behavior in reciprocal fashion. Before
such a run can become a runaway, with potential for productive
change but also risking destruction of the system, a child or other
family member will often move in to block the escalation —Jack-
son's "rescue operation" —and divert the hostilities and concern of
the parents to himself. Thus he substitutes a safe deviation-ampli-
fying process for one that threatens the family at its core.
One can think of still another sequence. If two parents who
engage in a withdrawal contest go too long without contact, a child
and one parent may get into a run of mutual hostilities which is
blocked when the other parent moves in to defend the child. This
action then serves to reestablish a connection between the parents,
albeit an unfriendly one. One can imagine runs countering runs in
this way in a kind of periodic seesaw, and observation of families
confirms that at times this is exactly what happens.

'
'
'Pathological ' Balance

The questions can still be asked: What keeps patterns like these
so firmly installed? Why don't they break down? Perhaps the an-
swer can be found in our concept of levels: The imbalance in the

nuclear family is serving to correct an imbalance in the larger kin

system, and is embedded in the equilibrium-maintaining mech-
anisms not only of that system but of its subordinate parts. These
may be other nuclear families, dyads within families, single persons,
or body parts belonging to those persons — and just as one cannot
tamper with any one element in an ecosystem without affecting the
whole, so one cannot change much in a family or a member of a
family without affecting a larger field. This includes other social
systems impinging on the family as well. Although we cannot say
that such a "field" has a homeostasis as Jackson thought the family

62
THE SECOND CYBERNETICS
had a homeostasis, the combined effect of many systems leaning on
or pulling against one another may add up to the kind of stability
for which the ecosystem is such a good analogy.
To make this idea a bit more concrete, let us look at the interplay
of feedback influences, both those amplifying and counteracting
deviation, in the hypothetical case of a child whose irrational be-

havior seems to heal a parental split. First of all, the same process
that amplifies the deviance of the child is deviation-counteracting
in regard to the marital dyad. Moving to another level of system,
one could say that the very consequence of having a "sick" member
may be deviation-amplifying for the nuclear family if it interferes
with important family functions. For instance, the family line may
die out with that generation if the sick member is an only child who
becomes unable to have a family of his own.
However, if one shifts to the extended kin group, a deviation-
counteracting effect may reappear. The parents' inability to form a
strong tie may be due to the fact that one or both of them are still

being used to mediate relationships in their own families of origin.

If so, they are stabilizing relationships in these other groups. Given
this prior arrangement, it is possible that all the others have to
follow. Whatever the case, any intervention that tries
it is clear that
to reverse the deviation-amplifying sequences here, without figur-
ing out how to deal with the deviation-counteracting ones, will fare
badly.
To sum up, we can say that a family which is off-balance in regard
to its own system because it is maintaining the balance of other
systems will perpetually be exposed to the destructive effects of
positive feedback chains. In an effort to compensate, some of these
will be used to counter others. Any form that prevents a disastrous
split in the family, as when a child's behavior serves as a diversion,
can be thought of as a corrective circuit for an escalating feedback
chain. Alternative mechanisms might be for one of the spouses to
develop a symptom and heal the split by becoming more dependent
on his partner, or for the family to achieve a united front by scape-
goating some outsider, or by mourning a departed member. Some
families use an uneasy mix of all these strategies.
All families become periodically unbalanced; they have to, as
power relationships between the generations shift. And all families

63
 Foundations of Family Therapy
experience the stresses that produce vicious circles in interpersonal
relationships. What is different in families with members in the
deepest trouble is the way these vicious circles continually repeat,
without ever forcing the family to change in a morphogenetic direc-
tion, because a symptomatic problem or a pattern represented by a
problem person is there to prevent such a change.
And that brings us to a further question: If family pathology can
be so stable, what can finally cause it —as sometimes happens—to
break down? In searching out the answer, we arrive at the place
—
where the sociology of deviance with its process of typecasting
—
and family theory with its process of scapegoating begin to sug- —
gest common conclusions.

When the Corrective Circuit Fails

The reasons why the circuits that keep family pathology in place
sometimes have never been expressed satisfactorily. Many fam-
fail

ily theorists take the view that there are really two parts to achiev-

ing true mental patienthood: a long period of training in the "right"

family setting, learning to mediate in many triangles, and the proper
accreditation by an authorized psychiatric source. The two types of
deviation-amplifying process alluded to before come to mind: (1) a
condition of deviance gradually arrived at, and (2) a crisis that
ushers in a runaway. Of course what constitutes a crisis in a particu-
lar family can only be answered in the specific instance. But one can
surmise that any sudden shift in the arrangement of checks and
balances in the kinship group and its subsystems will cause an upset
that the family may be unable to handle. A very obvious threat of
this sort would be the departure of a key person who seems to be
helping to stabilize the family. Children who are performing in this
way often erupt with disturbances when they reach adolescence,
and the next system up — the community— is often brought in to
restore equilibrium.
Is this a well-deserved end, then, to the pathological balance? Not
always. The officials whom society empowers to act in these situa-

64
THE SECOND CYBERNETICS
tions often authorize the family to go on using the person who is

the key to their stability as before. But there is a difference. Previ-

ously, in our hypothetical family, as the distance between the par-
ents would periodically widen, the child would produce the behav-
ior that caused it to close up again. But —
an important point he —
was not yet your true scapegoat; he was not hated, feared, extruded.
It is society that comes in and turns the "fault" that intermittently

opens up between the family and the child into a permanent chasm.
The scapegoating process —redefined here as a way of relocating the
split in the family — is only made easier by hospitalization or institu-
tionalization. The family is free to continue to have a symptomatic
person without having to deal with his or her inconvenient protests.
Thus does society move in to take a deviation-amplifying role,
replacing the family's relatively benign ceremonies with its own
"degradation ceremonies," as sociologist Erving Goffman calls

them. The person so honored is now stigmatized and placed outside

the pale. But what promotes deviance on the level of one system can
inhibit on another. Society is the beneficiary of the deviation-
it

counteracting effectproduced by a deviant on his group, which is to

reaffirm its solidarity, its belief in itself and in the righteousness of
its ways. But herein lie the very seeds of its destruction (back to a

deviation-amplifying aspect again!), not in terms of the present struc-

ture of the group but of its future ability to adapt and change. One
could say that every single-deviant arrangement is one more coffin
nail for the group.
Thus, this apparent conclusion only raises a new set of questions.
Do these mechanisms for maintaining equilibrium also have the
potential for disrupting it, and in such a case, how can one predict
whether the result will be destruction of the social group or a leap
to a new form? Are there laws governing these forceful cycles? Are
they all the same, or do they differ? We are in much the same state
as when the existence of electricity was discovered, but not until the
principles governing this source ofpower were understood could it
be harnessed and put to use.
In the next chapter we will move back to the mundane question:
How can these principles be applied? We will be back on the clinical
plane again, tracing the findings of researchers who have tried to
describe a typology or continuum on which to range the families

65
 Foundations of Family Therapy
they were working with. Early attempts at typology placed families
in categories defined by symptoms: the "schizophrenic" family, the
"alcoholic" family, the "multiproblem" family. As Reiss points out,
this ties the type of family to the type of problem one individual
in the family is said to possess, and comes perilously close to tradi-
tional psychiatric classifications. 26 Researchers took an important
step when they tried to place families in terms of structures or
sequences rather than particular disorders. Jackson's interactional
typology for couples, and Minuchin's structural typology for dis-
turbed families, though bipolar, nevertheless begin to move our
thinking away from a individual symptom-related orientation and
toward a family-wide view.

66
Chapter 4

Typologies of
Family Structure

Symptom Typologies

Out of the fascination with schizophrenic communication and

how might be fostered by family communication, a new branch
it

of family research arose. This approach focused on microstudy of

interactions, verbal or nonverbal, inan attempt to link communica-
tion style with the dominant type of symptom found in a family:
in other words, to suggest a typology of families by symptom.
Early work in this direction came about in a rather circuitous way.
We have already seen the uses of accident in the Bateson group's
research. Now serendipity enters again. At Yale in the early 1950s,
Theodore Lidz and his coworkers were attempting to map out the
interior workings of the family of the schizophrenic. Since they had
a psychoanalytic orientation, it did not occur to them to see the
family as a whole. Their original plan was to obtain Rorschach
protocols from each family member and to construct a portrait of
the family from a composite of these materials. The team did once
try to see a family with the patient, but this proved unworkable, and
the experiment was not, at that time, repeated. Lidz subsequently
became discouraged about the value of his Rorschach protocols and
the project was temporarily discontinued.

67
 Foundations of Family Therapy
It was during this period of discouragement that Wynne met Lidz
and learned about the research with the Rorschach protocols and
the disastrous family interview. Wynne had also been using an
indirect method of family research, one even more inferential and
roundabout. To study the family of the schizophrenic, he had de-
vised the scheme of having a therapist interview each family mem-
ber and, by analyzing the interlocking and reciprocal nature of the
transferences, try to figure out what the family must be like. This,
like Lidz's protocols, had proved unsatisfactory.
Wynne asked Lidz to send him a number of the Rorschachs,
specifying that they should be from parents of young adult schizo-
phrenics and "normals." A gifted coworker, Margaret Singer,
analyzed these transcripts blind and was able to discriminate accu-
ratelybetween parents with disturbed offspring and those without.
She then performed the even more unusual feat of taking protocols
of disturbed adolescents with a variety of labels (Autistic/
Schizophrenic, Neurotic/Withdrawn, and Delinquent/Acting Out)
and matching these with the correct sets of parents. 1
This early success in predicting symptomatic members from close
analysis of family communication was never sufficiently replicated
by other researchers to qualify as more than a brilliant, suggestive
beginning. But at the time its impact on other researchers was great.
The group in Palo Alto was also trying to isolate communicational
variables associated with different kinds of symptoms. Studies like

those of Wynne and Singer supported the idea that a typology of

on symptoms was more than just a dream, and that
families based
the Rosetta Stone of family communication was about to be
deciphered.
So it was that at the Mental Research Institute in the early 1960s
a fascinating guessing game was played. A researcher would play
back a piece of taped family conversation to a group of colleagues;
for instance, a fragment of a structured interview between a mother
and father. The assembled company would then try to guess
whether these parents might have a schizophrenic offspring, an
—
underachieves a delinquent and they were often correct. If the
conversation was between parents with a young child, the group
might speculate whether the youngster might be expected to have
a psychotic break at, possibly, age twenty. Don Jackson and his

68
TYPOLOGIES OF FAMILY STRUCTURE
fellow researchers were trying to perfect a predictive as well as
postdictive methodology for analyzing communicational character-
istics of families with symptoms. One of Jackson's major interests

just before his death in 1968 was a study of families that produced
members with ulcerative colitis. Once identified, he believed, pat-
terns associated with specific symptoms could be "read" from
samples of family interaction.

Interaction Typologies

Even though Jackson's "matching" experiments were not always

successful, the notion would not die that different kinds of symp-
toms might be linked with different kinds of families. However, it
began to be clear that it was not possible to be too specific about
the symptom. The Palo Alto researchers began to play with a differ-
ent kind of typology, one based on attributes or sequences. Jackson
had already suggested a classification of relationships that crossed
degree of marital satisfaction (or misery) with longevity: Stable-
Satisfactory; Unstable-Satisfactory; Unstable-Unsatisfactory;and
Stable-Unsatisfactory. This typology was dyadic and did not have
a true family-wide application, but it served as a grid for a typology
of couples. 2
Jackson used another couple typology based on Bateson's catego-
ries for schismogenic relationships. In Mirages of Marriage Jackson
defined three basic "modes of interactions": symmetrical, comple-
mentary, and a balanced mix of the two that he called reciprocal.
Each had own potential for pathology, but Jackson felt that the
its

reciprocal mode was preferable, since it allowed for more flexibility

and fitted the basically egalitarian ideals of American society. 3
In Pragmatics of Human and his coauthors
Communication Jackson
recognized that symmetrical and complementary couple sequences
may be found in normal, healthy interaction but may also become
rigid and produce distress. 4 Symmetrical escalations, if taken to an
extreme, amount to constant rejections of each self by the other.
These escalations, when pathological, stop only when one or both

69
 Foundations of Family Therapy
partners become physically or emotionally exhausted, and then
often only long enough to allow the partners to recover their breath.
The authors analyzed the stereotyped arguments of George and
Martha in Edward Albee's Who s Afraid of Virginia Woolf? as a particu-
larly stunning example of this pattern.
The other type, the rigidly complementary sequence, is most
dramatically shown by the sadomasochistic couple, although it is

common in many kinds of marital misery. This form is felt to be

more pathogenic than the other because of the constant discomh-
rmation of the two respective selves. Each partner must fit his own
definition of self into one that complements the other. This arrange-
ment works if one partner is sick or temporarily dependent on the
other, but when it is rigidly invoked it does not allow for growth
and change.
One of the most impressive illustrations of the use of microanal-
ysis to isolate basic schismogenic patterns and to relate these pat-
terns to symptoms is shown in a series of structured interviews
asking, "How did you two meet and marry?" which were given to
parents in clinical and nonclinical groups at the Mental Research
Institute. Pragmatics of Human Communication uses three excerpts from
these interviews to illustrate a couple that interacts in a rigidly
symmetrical fashion, one that interacts in a rigidly complementary
fashion, and one that does not fall into either category but responds
flexibly. Naive people are likely to choose the third excerpt as the
most "pathological" because it appears to show more expression of
conflict, popularly supposed to indicate trouble in marriage. The

opposite is really the case, for the third couple is getting along very
well. The "symmetrical" couple came into therapy because of their
constant fighting, which was having a bad effect on their children,
and because of sex difficulties. The "complementary" couple, al-
though not in a clinical sample, was found to be emotionally distant,
and the wife was extremely depressed. 5
These studies try to decode interaction patterns in wider social
fields. Unfortunately, they do not show how these sequences finish

or top out, rather than leading to a runaway. Watzlawick in Pragmat-

ics of Human Communication does bring up Richardson's fatigue expla-

nation, and also refers to Bateson's idea that one kind of sequence
acts to cancel the other, but these explanations do not go far

7o
TYPOLOGIES OF FAMILY STRUCTURE
enough. 6 In real life, rigid escalations of either type are usually
associated with third-party interferences that block them off, as

well as with "cutoff " behaviors (walking out in a huff, going to get
a drink, and so forth) and "trigger signs" for starting up again. In
none of these writings are the exchanges conceived as circularities;
the snake does not have its tail in its mouth; and so the descriptions
do not seem complete.

Structural Typologies

By the late 1950s the groundwork had been laid for the next step:

a sorting out of organizational features of families that produced

problematic members. Early family research had been with families
of psychotics, but the families that began to be studied in the 1960s,
poor and disadvantaged families, produced problem people who did
not always have so much "what is real" as with "what
trouble with
is right" according to the mores of the larger society.
They also seemed to be organized somewhat differently. The
route of inquiry was now reversed. Instead of being sidetracked by
the idea that there might be an "alcoholic family" or a "schizophre-
nic family" —
that families might differ according to symptom the —
architecture of the family was looked at and categories of organiza-
tion were established. Oddly enough, only then did it begin to seem
that specific types of family structure might indeed be connected
with certain categories of problems.
The first attempt at an organizational typology came from a re-
search project designed by Salvador Minuchin, Braulio Montalvo,
and others to study the families of delinquent boys, which was
reported on in Families of the Slums. 7 These boys' families seemed to
fall into two categories. One was characterized as the "enmeshed"

family, the other the "disengaged" family. Since both typeswere

found among poor families, it seemed that disadvantaged families
could not simply be lumped together under the rubric "culture of
poverty," nor did the label "delinquent" always point to the same
kind of family organization. Families of the Slums attacked the myth

71
 Foundations of Family Therapy
that poverty was synonymous with disorganization, and
necessarily
upheld the principle of equifinality: that the same outcome did not
necessarily mean the same origins, at least in reference to the con-
text of symptomatic behavior.
In the disengaged family, there seemed to be a relative absence
of strong connections,and relationship ties between family mem-
bers were weak The enmeshed family, by contrast,
or nonexistent.
resembled an error-activated system with high resonance between
the parts. The disengaged style gave the researchers the impres-
sion of

an atomistic field; family members have long moments in which they move
as in isolated orbits, unrelated to each other. They act as parts of a system
so loosely interlocked that it challenges the clinician's notion that a change

in one part of a system will be followed by compensatory changes in other

parts. 8

The enmeshed family was characterized by a "tight interlocking"

of members. "Their quality of connectedness is such that at-
its

tempts on the part of one member to change elicit fast complemen-

tary resistance on the part of others." 9
Minuchin's definition of "enmeshed" has a quality reminiscent of
Haley's "First Law" of human relationships, discussed in Chapter
2. Haley saw the homeostatic monitoring of behaviors as a normal

state of affairs in any group with a past and a future, but found it
present to an exaggerated degree in families that fostered psychopa-
thology. It is possible to say that there is such a thing as too much
homeostasis, and the term "enmeshed" seems to embody that idea.
One of the indicators of this state of affairs, according to Minu-
chin, is a lack of differentiation between individual family mem-
bers. As noted in Chapter 1, this idea is similar to Bowen's "undif-
ferentiated family," but "enmeshed" has the metaphoric edge over
"undifferentiated" in suggesting a too-tight connection between
parts rather than a gluey blob. A useful analogy for this hard-to-
describe phenomenon is the situation of the boys in the Never-
Never Land in Peter Pan who all had to turn over in bed at the same
,

time whenever one turned over.

Another effect of enmeshment, which Minuchin emphasizes in
Families and Family Therapy, is to weaken the boundaries that allow

72
TYPOLOGIES OF FAMILY STRUCTURE
family subsystems to work. 10 In brief, the boundary between the
nuclear family and families of origin is not well maintained; the
boundary separating the parents from their children is frequently
invaded in improper ways; and the roles of spouse and parent are
never clearly differentiated, so that neither the spouse subsystem
nor the parent subsystem can operate with ease. Finally, the chil-
dren are not differentiated on the basis of age or maturation level,
so that the sibling subsystem cannot contribute properly to the
socialization process.
In amore microscopic view of the interactions that characterize
an enmeshed family, Minuchin comes up with observations that
support much of what other family therapists and researchers have
described. For one thing, he confirms the striking inability of dyadic
transactions to persist in many disturbed one family that
families. In
included a child with a psychosomatic disorder, he noted that the
family continually deflected conflict through the use of third par-
ties. Whenever two persons disagreed and attempted to work out

some problem, a third would intervene. In this family:

Dyadic transactions rarely occur; interaction is either triadic or group. It is

characterized by a rigid sequence, which promotes a sense of vagueness
and confusion in all family members. For example, if a parent criticizes one
of the children, the other parent or a sibling will join in to protect the child,
and then another family member will join the critic or the criticized. The
original issue becomes diffused, only to emerge again later in a similar
sequence and to be similarly unresolved. 11

One could say that an inability of coalitions to remain stable

would indicate a special kind of structure —one whose systems and
subsystems are so entangled that, whenever there is a conflict of
rules or interests, there will be a tug of war over which set of
alignments will predominate. Since the outer limits of homeostatic
plateaus of various subsystems will continually be overpassed,
there will be many countering movements, or schismogenic reac-
tions of the sort Bateson described in Naven. As a hypothetical
instance, a wife has been talking for an hour on the phone with her
sister. Her husband, who hates his in-laws, gets into a fight with the

older son. The mother intervenes to try to protect the boy, but the
father smacks him and the boy goes to his room and slams the door.

73
 Foundations of Family Therapy
The fight moves with the husband hurling insults at
to the couple,
the wife's family and the wife screaming back. At this point a
younger boy, who is asthma-prone, starts to have an attack. The
wife goes to minister to the boy, who sometimes has to be taken to
the emergency room. This time the attack passes. After the child is

put to bed the father apologizes to the mother, who refuses to speak
to him. However, she phones her sister in the daytime for the next
few days.
We note that there are six escalating sequences here, each one
breaking up the one before it. They alternate between a presumably
symmetrical one (husband/wife) and various complementary ones
(parents/children). The kin group, the marital pair, the parenting
subsystem, and the sibling rescue squad are all participating units.
This event, or one like it, takes place at least once a week.

The Too Richly Cross-Joined System

Ashby, too, focuses on the processes of equilibration in complex

systems. In Design for a Brain, he writes:

Clearly, at any state of the whole, if a single part is not at equilibrium (even
though the remainder are) this part will change, will provide new condi-
tions for other parts, will thus start them moving again, and will thus
prevent that state from being one of equilibrium for the whole. As equilib-
rium of the whole requires that all parts be in equilibrium, we can say
(metaphorically) that every part has a power of veto over the states of equilib-
rium of the whole. 12

7
Thus Bateson's phrase, "the infinite dance of shifting coalitions/
seems applicable to any system composed of many parts and sub-
parts, all linked together in an interdependent way. Systems of this
type find it difficult to negotiate changes in their environment. But
how then to account for a universe full of complex entities whose
adaptiveness has been demonstrated by the fact that they have
survived? Ashby addresses himself to this puzzle and suggests that
there is an optimum degree to which parts of a system should be

74
TYPOLOGIES OF FAMILY STRUCTURE
interdependent if the system is to adapt well, or at all. His thoughts
on this are beautifully relevant to Minuchin's concept of the "en-
meshed" family with its extraordinarily high resonance among
parts.
In Design for a Brain Ashby investigates the possibility of using a
cybernetic model for adaptive systems, in particular a system like
the brain. The book is also a brilliant treatise on homeostatic princi-
ples and an application of these principles to the mechanics of
natural selection and adaptive processes in general.
From the beginning Ashby insists on seeing organism and envi-
ronment as a unified field, two interacting aspects of a whole. He
bids us consider that the dividing line between organism and envi-
ronment is essentially a mental artifact. Using the analogy of a
mechanic with an artificial arm trying to repair an engine, he asks
whether the arm is to be thought of as part of the organism that is
struggling with the machine or part of the machinery with which
the man is struggling.
To describe the totality Ashby chooses a term —the "state-deter-
mined system" —that includes organism, environment, and the in-
variant line of behavior that results when the interplay between the
two is and variables are constant. According to Ashby, the
fixed
state-determined system has no randomness in it. If it is at a certain
state and in certain conditions, it will do a certain thing, and it will
do this thing whenever the state and conditions recur.
Ashby also discusses the attributes of a stable system one that —
persists —
and describes the familiar mechanism of negative feed-
back that, by correcting deviations, keeps the essential variables of
the system within critical limits. Expanding this topic, he speculates
on the attributes of what he calls an "ultrastable" system one that —
not only remains steady in the face of small, continuous disturb-
ances but can reorganize itself to adjust to a large change in the outer

context. We
have already given some consideration to these ideas,
as they impacted upon Bateson, in Chapter 2.
Ashby surmises that a system with the capacity to handle both
kinds of change would have superior adaptive powers. He further
notes that the mechanism that governs such a shift in values would
probably take a step-function form, meaning that it would respond
to a drastic change in the value of some variable that would occur

75
 Foundations of Family Therapy
whenever an element reached a critical state. This second order type
of feedback, which he calls a ''step-mechanism/' would differ from
the first order feedback shown by the merely stable system in that
it would operate only occasionally, when essential variables of the

system overpassed their limits. At this point it would trigger off a

search for a new value that would once more return the system to
a stable state.
To test this idea Ashby constructed a simple, four-unit cybernetic
machine, the "Homeostat," which demonstrated the presence of
both levels of feedback. It not only corrected for minor disturbances
but was able to arrive at a new solution after a gross displacement
of one of its parameters. But, the Homeostat possessed a major
defect: was too richly
it The dilemma of the Homeo-
cross-joined.
stat brings to mind the lines written by the seventeenth-century
poet Francis Thompson in musing upon the interconnectedness of
all things in the universe: 'Thou cans't not pluck a flower without
7
troubling of a star/ One consequence, in regard to the Homeostat,
is ruefully commented upon by Ashby. He points out that a homeo-
static arrangement like the brain, with millions instead of only four
units, would take till infinity to reach a solution on a trial-and-error
basis.
But the worst problem with the Homeostat was that the whole
could not come to equilibrium until all its subparts did. Partial
successes thus could not be retained. Ashby offers the following
analogy:

The example par excellence occurs when the burglar, homeostatically trying
to earn his daily bread by environment
stealing, faces that particular
known as the combination lock. This environment has, of course, been
selected to be as difficult for him as possible; and its peculiar difficulty lies
precisely in the fact that partial successes —getting, say, six letters right out
of seven —count no progression toward the
for nothing. Thus, there can be
solution. Thus, confronted with an environment that does not permit use
13
to be made of partial adaptation, human and Homeostat fail alike.

What, then, is to be made of the fact that the world is full of

organisms that are homeostat-like and yet have shown the power
to adapt? Ashby suggests that this is possible in cases where the
system is not fully joined, or not at all times, and he goes on to state:

76
TYPOLOGIES OF FAMILY STRUCTURE
'The idea so often implicit in physiological writings, that all will be
well if only sufficient cross-connexions are available, is, in this
context, quite wrong." 14
^ It seems to me what the concept of enmeshment
that this is also
is about. The cross-connections between the parts and subparts of

a family system can be so rich that efforts of any element of that

system to find a new solution or to introduce the kind of random
searching that is the essence of trial-and-error adaptation cannot-
succeed.
Consider in this light the predicament of a therapist trying to
change a family that is enmeshed. He or she is in the position of the
experimenter in the early days of the cathode-ray oscilloscope, a
situation Ashby cites in another brilliant illustration of his thesis
that too much joining is not good:

Adjusting the first experimental models was a matter of considerable com-

plexity. An attempt to improve the brightness of the spot might make the
spot also move off the screen. The attempt to bring it back might alter its
rate of sweep and start it oscillating vertically. An attempt to correct this
might make its sweep leave the horizontal; and so on. This system's
line of
variables (brightness of spot, rate of sweep, etc.) were dynamically linked
in a rich and complex manner. Attempts to control it through the available
parameters were difficult precisely because the variables were richly
joined. 15

What Ashby calls a "too richly cross-joined system" 16 can be

treated or changed, he says, when enough communication is

blocked so that the parts become temporarily independent of each

other. The enough if
parts don't have to be actually severed. It is

one of the linking elements stands still and shows no change. The
parts linked by this unchanging element are then separated by a
"wall of constancies." Constancies, Ashby remarks, can "cut a sys-
tem to pieces." 17 In this way he comes to the position that there are
times when an increase in communication would be positively
harmful.
This does not mean communication is a good
that the absence of
thing. Rather, there is a happy medium.
there were not enough If

communication between parts of an organism, or between it and the

environment, it could not survive. For instance, the tongue has to

77
 Foundations of Family Therapy
be told to get out of the way when the teeth bite on a piece of meat.
The fewer the joins between the reacting parts, the fewer the possi-
ble modes of action, but a good level of communication means that
a large repertory of behaviors will be available. These benefits,
among others, have be weighed against the disadvantage that the
to
richer the joins, the more difficult it will be to make new adapta-
tions.
For family therapists, this last point is especially interesting.
Whole schools of family therapy have been founded on the idea
that the more the communication, the better off the family. Actu-
ally, much of family therapy consists of efforts to block communi-
cation, even when the rationale of "more communication" is osten-
sibly being followed. Ashby's statement that a system too much
connected with itself will not be able to change easily parallels the
growing belief of a number of therapists that many families im-
prove merely because the therapist has blocked the usual pathways
and forced them to search for new ones. But these ideas are also
important for the close backing they give to Minuchin's observa-
tions about enmeshed family systems.
His observations that a family is most efficient when its
not at its

parts and subparts are too closely interlocked, and that a functional
family is one in which there are status schisms between generations,
age differentiations between children, demarcations between syb-
systems enacted by the same persons, and a clear boundary around
the nuclear family, seem to fall in with Ashby's general conclusion
that a system needs to guard against too much cross-joining or else
must be able to create a temporary independence of parts. Most
family therapists agree that when a family displays too tightly
connected coalition patterns, this is a recipe for symptoms and
distress.
Bowen, too, describes a therapeutic gambit that strongly resem-
bles Ashby's recipe for creating "constancies" or "cutting a system
to pieces." At times Bowen one family member and
singles out
insists that he or she take a position on some family issue and
maintain that position in spite of opposition by the rest of the
family —which usually appears. This is a tactic designed to achieve
what Bowen calls "differentiation of self," but he also describes it

in a triadic context, saying that if one person in a three-person

78
TYPOLOGIES OF FAMILY STRUCTURE
emotional system can remain detached, tension within that triangle
will resolve itself. One assumes that whether this third person is the
therapist or a family member, Bowen is still talking about what he
,,
calls "detriangling the triangle, forcing one of the elements in a
chain of linked triads to remain still and not react. 18

The Too Poorly Cross-Joined Family

This category, like the particles that physicists believe exist be-
cause theoretically they should, may be a figment of the mind. A
too tightly connected family system argues for the existence of its

opposite, a too loosely connected one. However, an argument can

be made that if such a category just as far from represent-
exists, it is

ing random behavior as that found in the so-called enmeshed fam-

ily. The phrases used to describe such families suggest some kind

of formlessness, some kind of chaos. More unfortunately, the word

"disengaged," though well meant, suggests a quality of not caring.
Minuchin's example of the difference between an enmeshed and a
disengaged family is that in the former the parents get upset if the
child skips dessert, and in the latter they may not notice if he does
not eat all few
day. Yet in families with routines or ceremonies,
individuals can still show toward one another.
great caring
A further confusion develops because one cannot classify a fam-
ily on MinuchhYs scale without reference to the larger context. A

family as loosely structured as the "disengaged" family is said to be

will quickly become locked in with social institutions that form a
more enclosing kind of enmeshed structure, with agency personnel
acting as surrogate parents. The danger of cross-generation joinings,
linked by researchers like Haley to pathology in the family, is as
applicable to this new format as it would be to a family group.
Thus one can argue that there is no such thing as a totally disen-
gaged family. Usually one finds that families whose members do not
become drawn away into jails and hospitals, or get official watchers
assigned to them, have a great deal of structure, though not always
in a conventional form. It is to the credit of Minuchin and his

79
 Foundations of Family Therapy
researchers that they identified the alternate family forms of the
very poor: families with a grandmother and mother acting as par-
ents for children of bothwomen; and similar arrangements in which
a mother is aided by a parental child. These families run into the
same troubles as any other type. That is, symptomatic members
appear in families in which a mother and grandmother are subvert-
ing each other just as often as in families where mother and father
are subverting each other.
Perhaps it would be more useful to think of a continuum of
possibilities rather than categories attwo ends of a pole. One could
never find a family at the far end of such a continuum because it
could not survive under either set of conditions.
In addition, terms like "centripetal" and "centrifugal," first used
by Helm Stierlin to describe organizing principles of families with
adolescents, may be more useful than the terms "enmeshed" and
19
"disengaged." Concrete analogies might also help: bumper cars at
a fair versus a roller coaster — each man colliding with others in the
space of the rink, as opposed to people stuck tightly together by the
force of the ride. I also sometimes think of "rubber band" or "sticky
glue" families as opposed to "colliding molecules" families
"fused" families versus "fragmented" families — always with the
proviso that these types do not present an either/or, but rather
suggest extreme edges against which families might be measured.
In particular, I am partial to the idea that continual, random-
seeming connectedness, whether through noise, touching, alterca-
tions, or overlapping conversations, can be a way of being organized.
Lewis Thomas suggests this in a passage on termite behavior in
Lives of a Cell:

Grouped termites keep touching each other incessantly with their anten-
nae, and this appears to be the central governing mechanism. It is the being
touched that counts, rather than the act of touching. Deprived of antennae,
any termite can become a group termite if touched frequently enough by
others. 20

By contrast, if termites are isolated from each other, this activity

stops and they become quiet and standoffish. If they are in a pair,

they will actually abstain from touching each other, as if this poor
shadow of the real thing were too painful a reminder. This is not

So
TYPOLOGIES OF FAMILY STRUCTURE
to suggest that human beings are like termites, only to show that
in the animal kingdom, patterns of organization may take a number
of equally valid forms.
I would like to take styles
Following this same line of thinking,
of family interaction that seem fragmented and disjointed and show
that they are not random but rich in configurations.

Redundancies in the "Disengaged" Family

Assuming for the moment continuum from

that there really is a
too tightly connected to too loosely connected on which families
can be placed, let me take the position that what may look like
loosely connected families are not without rigidities of their own.
Research on families with schizophrenics has produced evidence to
challenge the idea that these families are without structure. One
could similarly question whether the so-called disorganized family
is without structure. It may have elements as restricted and stereo-
typed as those of the enmeshed family, and for different reasons be
just as resistant to change.
To start with, the model for the "disengaged" family was the
multiproblem poor family that came to the attention of social agen-
cies and clinics because it had, supposedly, "broken down." What
are the features of this family as described in Families of the Slums, still

one of the only books to take a systems view of poor families?

One finding was that most multiproblem families are headed by
isolated mothers without resources. When there is a kin group that
helps out, or a caring boy friend, or a strong grandmother, or an
appropriately used parental child, the family may at times work
badly, but it works, because the mother is not totally alone. It is

when these supporting people are not available that problems get
out of hand. As the authors state in Families of the Slums:

A prominent feature of disengaged families is the isolation of the mother,

who seems unable to contact the external world and draw on extrafamilial
sources of support. In the most extreme forms of this profile, one must look

81
 Foundations of Family Therapy
beyond the chronicity of incompetence in mothering to a family history
usually lacking in anchorage points such as stable work patterns and stable
relationships to a male, friends or other social groups. The mother is
alone. 21

Another feature seems to be a hiatus or gulf between the adult

world and the sibling world. The parents and other grownups seem
to dissociate themselves from responsibility for the children's be-
havior, unless it is personally upsetting to them or brings on unwel-
come community attention. The siblings develop into a collusive
subgroup, sometimes protective and supportive of each other,
sometimes not, but functioning as a slightly wild pack or tribe, good
at outwitting adults' attempts to control them.
When headed by an isolated, overwhelmed mother,
the family is

these attempts seem to be of a global, all-or-nothing kind. A mother

may seem sunk into apathy, ignoring the most incredible amount
of noise and confusion on the part of the children. But at some
moment she will tolerate the situation no longer and will hit out
right and left, often ignoring the major offender, who may be at the
other side of the room, to strike a nonoffender who is near at hand.
Minuchin and Montalvo also comment on the skillful confusion
tactics a group of children in such a family can use to provoke and
annihilate adult authority. Not only parents but therapists are hard
put to match wits with these children. There is a revealing example
in an interview in Families of the Slums. Somebody had taken money
from mother. One child accused another child, who after much
prodding said that he got the money from a friend; more confusing
information was fed in by other children; then the original accuser
said that the accused thought the money was his bus fare, and added
that he always thinks he can take money because the father used
to hide money under the rug, but he is never scolded if he takes it
even though the other children are. By this time the mother and
therapist are lost. The "game" seems much like the proverbial chil-
dren's clapping game: "Who took the cookie from the cookie jar?"
and seems to end either in disruption and no closure, or else, after
much confusion, a particular child, usually the family scapegoat,
will be established as the guilty party.
What principle could unite these different observations, leading
to a hypothesis that would explain the childrens' collusive mischief?

82
TYPOLOGIES OF FAMILY STRUCTURE
Minuchin and Montalvo give us a hint in commenting on the anxi-
ety displayed by a group of children when the mother left the room,
and their consequent noisy, disruptive behavior with much talk on
scary, violent themes. The authors feel that this response represents
the uncertainty caused by the disappearance of the one figure that
acts asan organizing principle for these children. They describe the
interaction in sequence form: The mother leaves, the children be-
come anxious and act up, she then must come back in to reestablish
control, they calm down again. A circular causality is operating.
This presentation is linear, since it describes only the way the
cycle works A truly systemic view would point out
for the children.
how all the pieces and move together. An illuminating instance
fit

comes from a family interview by Harry Aponte. This interview,

recorded on videotape, contains clues that suggest how the redun-
dancies of multiproblem families like this one might be contributing
to an ecological balance of sorts within the family.
The interview is with a poor black family that fully answers to
the description Everybody the mother, six
"multiproblem." —
grown or nearly grown children, and two grandchildren is at risk, —
from breakdown, illness, nerves, violence, accident, or a combina-
tion of all these factors. In addition, the family members are noisy,
disruptive, and hard to control.
At a certain point Aponte asks the mother, "How do you handle
all this?" The mother, who has been apathetic and seemingly un-
concerned as the therapist tries to talk with the children, says, "I put
on my gorilla suit." The children laugh as they describe just how
terrible their mother is when she puts on her gorilla suit.

An incident occurs shortly after this conversation which suggests

that a circular causal sequence is at work, one of those redundancies
that may have todo with family balance. Mother is still apathetic
and looks tired, and the therapist begins to ask about her nerves. At
first the children are somewhat quiet, listening. As she begins to
admit that she has bad nerves and that she is taking pills, they begin
One boy pokes the baby; another boy tries
to act up. to restrain the
baby from kicking back; the baby starts to yell. The therapist asks
the twenty-year-old daughter (the baby's mother) if she can control
him; she says no. At this point the mother gets up and smacks her
daughter's baby with a rolled-up newspaper, rising out of her leth-

83
 Foundations of Family Therapy
argy like some sleeping giant bothered by a gnat. She sits the baby
down with bump, and he makes no further trouble. During this
a
sequence the rest of the children jump and shriek with joy, causing
their mother to reprimand them, after which they calm down and
the mother sits back, more watchful now and definitely in control.
Looked at from one point of view, this is an example of disorgan-
ized behavior. The mother is abdicating her responsibilities; the
children are badly behaved; and when the mother does try to con-
trol them, she overreacts and is only partly effective. But from
another point of view, we may be seeing an ameliorative sequence.
Let us assume, as in this case, that one is dealing with a family with
an isolated, overwhelmed mother who gets into periodic depres-
sions as a form of recuperation from stress. Let us assume also that
the children become correspondingly anxious. A powerful antidote
to both the depression and the anxiety would be disruptive behav-
iors by the children. The cycle goes thus: The mother's depression
carried past a certain point triggers off collusive mischief in the
children; once this mischief rises to a certain level, it will impel the
mother to reassert her power; seeing the mother taking an active
position relieves the children, and as she takes over they quiet
down. Peace will prevail until the mother once more sinks too far
into depression or acts apathetic, and the whole sequence will start
again. 22
This is a very global, primitive cycle, and it appears to be common
in such families. It is also fairly limited in its usefulness, as all family
members pay a price for learning the behaviors it requires. For one
thing, these behaviors automatically get them into trouble in the
outer world. Outside the home —in the classroom, for instance
one frequently finds the following scenario: Teacher writes words
on the board for the children to copy down, then sits quietly at her
desk marking papers. For a child who comes from a home where
quietness by the adult in charge is a danger signal, this creates
anxiety. The child may start making faces at another child or throw
something or whisper an insult. There are children who become
experts at instigating fellow children at such a moment, and other
children who get caught and wind up being sent to the principal.
This is only one possible way to interpret the disorderly behavior
of children in multiproblem families. It represents a sequence that

84
TYPOLOGIES OF FAMILY STRUCTURE
contributes to the stability of the home, even though it is at the

expense of the individuals who get caught in this machinery. We

shall see in later chapters other circular causal cycles, different but
work collaboratively
equally difficult to see, which seem to to keep
the balance (or are the balance) in "enmeshed" families.
The bipolar model we have been discussing has some serious
drawbacks, however, in that it implies a too-simple dichotomy be-
tween random and rigid. Aponte's use of "underorganized" as a
substitute for "disengaged" is an attempt to deal with this problem.
However, underorganized still suggests an either/or continuum, as
if we were simply talking about families having too much or too

little structure or organization.

I am also dissatisfied with the continuum concept itself. It does
not cover enough variables, nor is it rich or interesting enough. In
the next chapter we will consider different, more complex models,
both grid and process formats. In particular, we will address the
growing tendency to think of family worlds on the total system
level, calling upon a new and important concept: the "family
paradigm."

85
Chapter 5

The Concept of
Family Paradigms

The Metaphysics of Pattern

Using this apt phrase, psychologist Paul Dell has recently called
attention to a profound difference between the transactional model of
research on families of schizophrenics and the etiological one. Stud-
ies that try to establish an etiology different from the traditional
individually oriented one have assumed a causal link between family
structure and psychotic behavior. But to say that a particular condition
is contingent on a given type of family organization, or to think that
there is a direct connection, is to make the mistake of linear thinking.
The transactional view, according to Dell, is not concerned with
etiology but amounts to a redefinition of what schizophrenia is.

Transactional researchers take the position that behaviors labeled

schizophrenic are part and parcel of the pattern of family relation-
ships in which the behaviors are embedded, and that neither family
nor afflicted individual can be singled out as the "location" of the
disorder. As Dell says,

The behaviors of family members which together constitute the various

aspects of the pattern are not linearly causal of one another but are co-

86
THE CONCEPT OF FAMILY PARADIGMS
evolutionary. Bateson (i960) and Wynne and Singer (1965a) speak not of
causation but of how the family as a whole fits together. Thus, etiological
constructions of the family theory of schizophrenia (e.g., Fromm-Reich-
mann, Lidz) as embodied in most research to date, neither grasp nor ade-
quately test the transactional position. 1

Dell places David Reiss in the transactional group. In his work

Reiss is about the need to see a family-wide or systemic pat-
clear
7
tern, a "family paradigm/ as an emergent property of family expe-
rience. It is not reducible to the perceptions or reactions of any
individual member of the families. According to Reiss, studies that
link family structure too directly to disorders of individuals, exem-
plified in such terms as "schizophrenic families" and "multiproblem
families," violate that concept.
Building on the experiments of Wynne and Singer, among other
influences, Reiss has tried to research a "metaphysics of pattern" of
his own. In an earlier phase of his work he did indeed attempt to
correlate family interaction styles with disturbances that might be
expected to emerge in individuals in such families. In later thinking
he seems to be broadening his concern with pattern to include a
much richer application ofkey variables that account not just for
families with symptomatic members but for the whole rich variety
of families. In addition, he is beginning to suggest that such disturb-
ances may have to do with the inability of the family paradigm, for
whatever reasons, to process information when the outer (or inner)
field changes too radically. At this point, the paradigm may break

down, with the consequent possibility of extinction of the family

or, in a more positive direction, paradigmatic change, which gives

the family a new lease on life.

Let us start with Reiss's more clinically oriented work. In a re-
search paper linking family interaction styles with individual think-
ing processes, he describes a problem-solving experiment that drew
on three populations: eight families with offspring diagnosed
"schizophrenic"; eight families whose members had "character dis-
orders" (defined by Reiss more concretely as "severe solitary delin-
quencies"); and eight with offspring who had no known disorder. 2
The parents, the symptomatic child, and one sibling were included.
The problem design required a subject to sort a group of fifteen
cards, each containing a patterned sequence of letters or nonsense

87
 Foundations of Family Therapy
syllables (pvk, pmvk, pmsmsvk). The experiment itself consisted of
three parts: an individual task in which each person, sitting in
first,

a separate booth, was asked to sort the group of cards by himself;

then a family task, with each family member getting the same series
of cards to put in order; and a final individual card-sorting task.
During the family task each member would start with two cards and
when these were sorted would press a "finish" button and get
another card, until all fifteen were sorted. Family members could
communicate with each other by earphone, and were encouraged to
share ideas and information. Whether to trade ideas or go it alone,
to adopt a uniform strategy or to use different ones, to wait for
enough cards to suggest an overall pattern or to settle for a sequence
choice based on the first few, were all options a family or its individ-
ual members could choose, depending on family communication
rules. The experiment was set up so that it was easier to arrive at

a correct hypothesis for sorting the cards after they were all, or
mostly all, in, but a rule for ordering them could be guessed at, and

trial-and-error attempts were possible because later cards could

verify that a given sorting arrangement was in fact correct.
This experiment suggested the existence of family mini-universes
which were very different from each other, and which correlated
closely to the clinical groupings. Reiss called these categories "con-
sensus-sensitive," "interpersonal distance-sensitive," and "envi-
ronment-sensitive," terms corresponding to the "schizophrenic"
families, the "delinquent" families, and the "normal" families. The
variables that differentiated these families were selected out by the
test and were congruent with earlier family research findings, espe-
cially those of Wynne. Thus the experiment is not merely interest-
ing in itself but represents a theory-building effort that goes back
several decades.
The three major characteristics that seemed to mark off the
groups of families were: first, family members' ability to use cues
or suggestions from each other; second, their ability to use cues from
the laboratory; and third, their ability to reach closure. The puzzle
was constructed it could be solved most effectively when
so that
sufficient information was obtained and when this information was
shared. It was important for all family members to coordinate their
separate efforts to build useful hypotheses for sorting the cards, and

88
 THE CONCEPT OF FAMILY PARADIGMS
to offer new suggestions or corrections. It was also necessary to pay
attention to information from the environment, in the form of the
cards that were continually being fed in. Clearly the experiment
tested two sets of connections: connections between family mem-
bers and connections between family members and the outer world.
The three types of families responded very differently to the tests,
and the families with clinical problems did most poorly. Reiss found
that families with schizophrenic members were very clued in to
each other but very walled off from everything else. Their internal
sensitivity did not help them with the test but hindered them. In
Reiss's words, "In this kind of family, there is a joint perception that
the analysis and solution of the problem are simply a means to
maintain a close and uninterrupted agreement at all times." 3
/T\ Reiss's explanation for the "consensus-sensitive" family's need
always to be close and in agreement was that they experienced the en-
vironment as threatening and dangerous. The testing situation was
a threat to be warded off; thus people in these families would reach
a solution very quickly, before much information had become avail-
able, and hang on to this solution even when later facts or a better
solution arrived at by another family member contradicted it. Conse-
quently they would most often fail to find an adequate solution to the
problem, not because of insufficient family communication but be-
cause of premature closure. They would rather be wrong than fight.

This syndrome picks out the family characteristics that Wynne

called "pseudomutuality" and the "rubber fence." It also confirms
the quality of "enmeshment" Minuchin described and the
that
tendency toward emotional fusion and undifferentiation that
Bowen found. These researchers, being clinicians, tended to limit
their descriptive models to families with problems. Reiss goes in a
much-needed normalizing direction by extending his categories to
include families that do not necessarily present problematic mem-
bers. As a normative cultural variant of the consensus-sensitive
family, for instance, he cites studies of southern Italian clans, with
theircode of family solidarity and their perception of the outside
world as menacing and unpredictable.
Diametrically opposed to the families with schizophrenic mem-
bers were the families with members who had "severe solitary

Q> delinquencies." These families attended very well to cues from the

89
 Foundations of Family Therapy
external environment but did not attend to cues from each other.
During the tests family members behaved as though it would be a
mistake to accept opinions or hypotheses from their
relatives. There
seemed be an overpowering need to show that one could be
to
independent, could master the environment on one's own. Taking a
more general view, Reiss says that these individuals seemed to share
"a perception that the environment was many pieces as
split into as

there were family members; each member had access to his own piece
and therefore attended to environmental cues from his piece only." 4
These families, the "interpersonal distance-sensitive" families,
seemed to experience the laboratory and the externally given prob-
lem as a way to demonstrate individual mastery. Since the boundary
between the world and members of the interpersonal distance-
sensitive family is not governed by the principle of the protective
stockade, their problem-solving skills tend to be better than those of
the consensus-sensitive families. They have a better appreciation of
"out-there reality," as opposed to the distortions called for by the
family whose main idea is to remain close. But their stance as loners
cripples them They do not take well to the
in another direction.
collective sharing of hypotheses, and may do poorly because an
individual will persist in trying to solve the problem on his own and
so prolong the test time, as well as being limited to his own resources.
Normative examples of the interpersonal distance-sensitive fam-
ily are not well reported on. Reiss cites Minuchin's "disengaged"
family as an example from a population with delinquent children;
however, he singles out characteristics that are not necessarily path-
ogenic, notably the sense of isolation and distance people sometimes
seem to experience in these families. He also comes up with an
example of a middle-class interpersonal distance-sensitive family
(the Littletons), described in sociologists Robert Hess and Gerald
Handel's Family Worlds, who had a disconnected, individualistic style
5
of interaction and tolerated a high degree of conflict and dissonance.
The third category, the "environment-sensitive" family, repre-
sents the relatively problem-free group. People in these families
hold both the values that Reiss emphasized in an optimal balance,
indicating an ability to use cues from other family members and at

the same time to accept and incorporate cues from the environment.
Members of these families can make hypotheses individually or

90
THE CONCEPT OF FAMILY PARADIGMS
jointly, can delay closure until enough alternatives are explored to
validate the conclusion finally arrived at, and can process and share
new information. Unlike the consensus-sensitive family, they are
not constrained by a need for cohesion at all costs, nor are they
hampered by a philosophy of go-it-alone. This group has a sur-
rounding membrane which is tight enough to provide support but
not so impermeable as to block out fresh data. Flexibility about both
degree of internal connectedness and connection with the outside
world seems to be the distinguishing mark of the family that does
best, at least on the problem-solving tasks in Reiss's study.
seems odd to look for cultural models for families in this cate-
It

gory, which is already normal by definition of not having come to

community attention, but Reiss cites studies of eastern European
shtetl culture as affording a world view that embraces a strong sense
of family ties but allows for exploration and mastery of the environ-
ment as well.
Reiss has truly attempted a "metaphysics of pattern" in this de-
scription of family systems, even though this summary of his typol-
ogy and has emphasized only three dimensions: internal
is static

connectedness, external connectedness, and closure. We shall see in

a discussion of his later thinking on family systems that he breaks
out of this model to one which includes variables that govern the
ways in which a family changes paradigms. In the meantime let us
look at the work of another researcher, Eleanor Wertheim, who has
added the change variable in her own way.

A Grid Model for Family Topology

Wertheim's variables have to do with a process taxonomy the —

way a system changes or remains the same over a period of time.
In her article "Family Unit Therapy and the Science and Typology
of Family Systems," Wertheim argues that one must take into ac-
count the change mechanisms of a family as well as its structural
elements. 6
If a family is a rule-bound, change-resistant entity, two

dimensions are important to check. One is the family's morpho-

static (steady-state or homeostatic) tendencies, the other its mor-

91
 Foundations of Family Therapy
phogenetic (rule-changing) capacities. In the dimension Wer- first

theim distinguishes between C onsensual Morphostasis (Mc) repre- ,

se nting a balance between individual goals and family goals, and

Fo rced Morphostasis (Mf), found in a family in which individuals
are bound by ri^id but covert rules that operate in the interest of
t he family . We might think of participatory democracy versus a
more totalitarian form of rule, although the totalitarian society is
different in that the rules are anything but covert.
The second dimension, the rules for changing the rules, incorpo-
rates what we have spoken of as second order change. If a family
is high in Consensual Morphostasis, Wertheim thinks this automat-

ically means that its rules can be changed in a flexible manner to

meet changed circumstances, that morphogenic changes are built

Table 5.1
Classification of Family Systems and Their Predicted Response to Family Therapy (F. T.J

Consen- Forced
Induced sual Mor-
Morpho- Mor- pho- Type Integration Predicted Response to FT.
genesis phosta- stasis
(I.M.) sis (Mc) (Mf) of System Accessibility Duration Outcome

High High Low Open 1 — — —

Integrated
High 2 Acces- Short- Favor-
Fairly Integrated sible Term able

Low Low Partly 3 Acces- Long- Vari-

Open Unintegrated sible Term able
High (Extra- 4 Acces- Short/ Favor-
system- Pseudo-Integrated sible Long- able
ically) Term

Low High Low Partly 5

— — —
Open Integrated
High (Intra- 6 Resist- Short/ Favor-
system- Fairly Integrated ant Long- able
ically) Term

Low Low Closed 7 Unmo- Failure Unfa-

Disintegrated tivated vor-
able
High 8 Resist- Long- Vari-
Pseudointegrated ant Term able

Wertheim, E., "Family Unit Therapy and the Science and Typology of Family Systems," Family
Source: Process

12 (1973), 343-376.

92
THE CONCEPT OF FAMILY PARADIGMS
into its regulatory capacities. Again, as in a democracy, there is

machinery for a new vote to be taken and a new party with a

potentially better program to be elected. One can assume that a
family has this capacity if it appears to be flexible or makes morpho-
genic changes easily, like a horse that takes naturally to jumps.
If a familyjcanno t fin d the means to alter its rules when chang e
is hen we would look for a capacity to accept chang e
called for, t

Fro m the oufside usually from agents of the system the next leve l
,

up, meaning the wider community. Wertheim terms this property

mdu ced~Morphogenesis (IM). Famili es that can use community re-

sources, friends, ministers, doctors, therapists, or elders from their

own kin groups to help them negotiate a morphogenic change are
in this category.
Wertheim adds to her change variables a group of structural
variables that have todo with boundaries, both internal and external.
Thus she includes systems that are both internally and externally
open; systems that are both internally and externally closed; partly
closed systems that are open only internally; and partly open systems
that are open only externally. Mixing the change dimensions with
these structural dimensions, she develops eight family types from
which she then derives eight clinical profiles, as in Table 5.1.
The resulting grid consists of two structures that operate with
optimal flexibility, produce few problems, and are called "inte-
grated"; two that are associated with mild neurotic problems and are
called "fairly integrated"; two that are called "pseudointegrated,"
following Wynne's concept of pseudomutuality, and are associated
with either acute or chronic psychotic disorders; and two that are, res-
pectively, "unintegrated" (families producing social disorders) and
"disintegrated" (families that can hardly be said to be families at all).

Wertheim compares her grid with Reiss's scheme and finds them
compatible, with modifications. The environment-sensitive family
she identifies with her "open-integrated" family; the consensus-
sensitive family with her "closed pseudointegrated" type; the
interpersonal distance-sensitive family with her "externally open-
unintegrated" type.
A way to synthesize these categories is represented in Table 5.2,
which uses Reiss's three categories for family structure but organ-
izes them on the basis of degree and type of connectedness. The

93
 Foundations of Family Therapy
environment-sensitive families are both externally and internally
well connected. The interpersonal distance-sensitive families are
well connected externally but poorly connected internally. The con-
sensus-sensitive families are well connected internally but poorly
connected externally.
Within each type, then, there would be two versions, which
would be differentiated according to Wertheim's mix of cybernetic
elements: homeostatic (morphostatic) versus rule-change (morpho-
genetic) variables.The consensus-sensitive families would be high
on Forced Homeostasis, being rigidly rule-bound, and, of course,
low on Consensual Homeostasis. There would, however, be a diff-
erence between the consensus-sensitive family which was high on
induced rule-change and the "rubber fence" family which rejected
outside attempts at change. The latter would be the most resistant
family, most likely to produce psychotic manifestations and least in
touch with "reality."
For instance, families from religious sects that emphasize values
or beliefs not shared with the surrounding society sometimes also

Table 5.2
Typology of Family Structure

Induced Consensual Forced

Morphogenesis Homeostasis Homeostasis

Environment-Sensitive Family High High Low

A
(externally and internally

c/a
well connected) Low High Low
B

Interpersonal Distance- High Low Low

Sensitive Family A
(externally well connected;
internally poorly connected) Low Low Low
B

Consensus-Sensitive Family High Low High

A
(externally poorly connected;
internally well connected) Low Low High
B

Process Variables

94

T>S-
THE CONCEPT OF FAMILY PARADIGMS
fall into this category. Groups can hold belief systems which, if held
by an individual, might proclaim him psychotic: the ending of the
world on a given date, or communication with beings from other
planets. People in such groups are not necessarily thought of as
crazy since they share their beliefs with others. But one can see that
consensus about basic beliefs would be central to a group's ability
to hold onto members and continue to exist. One could even say
that the more unprovable or implausible a belief system was, com-
pared to the beliefs of the outer society, the more effective it would
be in protecting and walling off the group. There is a usefulness, a
validity, to this kind of shared illusion that may make it easier to
understand the reasons behind consensus-sensitive behavior in
families with psychotic members.
The interpersonal distance-sensitive family would be low on ei-
ther type of homeostasis, since it is poorly connected and does not
possess a high degree of cohesiveness. But the family in this category
with a capacity for induced rule-change would, by definition, be
more amenable to help from the outside than one without this
capacity. As Wertheim points out, the "disintegrated" family type
may be a nonviable form and may not be salvageable. It is hard to
come up with an interpersonal distance family in a pure form;
perhaps none exists, but only strange amalgams. One such amalgam,
a family with many "colliding molecules" characteristics and yet
with some "sticky glue" as well, is the poor black family described
earlier, in Chapter 4. In the videotapes of this family there is one

sequence that seems to illustrate the interpersonal distance-sensitive

principle of "every man's an island" (to misquote John Donne).
The three daughters aged twenty, nineteen, and seventeen, were
describing their feelings of fear and insecurity at night after their
mother had gone out to work. Instead of taking turns watching the
apartment, so that the rest could sleep while one stayed up, all three
stayed up together. was as if this were not a shared situation but
It

three personal and private situations that had nothing to do with

each other. The mother, described by the children as not afraid of
anything, was felt to have all the power, and when she was gone
the daughters felt extremely vulnerable. The older boys displayed
a different version of this unconnected go-it-alone stance. When
the therapist asked, "Can't the boys take care of you?" the girls

95
 Foundations of Family Therapy
complained that the boys (including a twenty-four-year-old who
was part in, part out of the house) didn't seem to worry or care. If
the girls tried to wake them up at night, they would refuse to take
the girls' fears and go back to sleep.
seriously
The very permeable boundary around this family was graphically il-
lustrated by the fact that people could easily get in and threaten their
safety. They lived in a neighborhood of houses poorly locked, streets
badly policed, and territories constantly threatened, but the situation
had a correlate in the fragility of the family boundary and the structur-
al deficits that kept it fragile. The children had not learned how to

form protective groups, plan strategies, organize hierarchically.

To move on to the last category, the environment-sensitive fam-
ily would be high on consensual Homeostasis and low on Forced

Homeostasis, and, like those in the other categories, would differ

only in the degree to which outside agents would be accepted as
reinforcers for internal rule-change. Wertheim suggests that the
family that accepts outside agents is more functional than the other
one in this category, implying a value system in favor of therapy.
It is equally possible that the version of the environment-sensitive
family that rejects induced rule-change is better off than the other
type. Spontaneous shifts may be preferable to induced shifts, as

spontaneous labor, when possible, is preferable to induced labor.

Wertheim uses her scheme to predict types of problems likely to
be found in each of her categories, and she has psychiatric profiles
for all eight. The reduced set of categories represented by Table 5.2
merely distinguishes between the three possible structures de-
scribed by Reiss, divided into two subversions of each, using Wer-
theim's change variables. It is probably too early to claim clinical

profiles for these groupings; it is enough merely to have a grid from

which to start researching such profiles in more detail.

Process Models for Family Organization

The problem with grids is that they are a self-contained and

essentially static model for how families work. There is never any

96
THE CONCEPT OF FAMILY PARADIGMS
hint of how from one box of the grid to another, or
a family gets
even whether this isIf it is, then is there an order to the
possible.
way families change? Are there levels of organization through
which families, if they do change, must pass through, or can they
move anywhere on the board?
These questions are at least addressed by Beavers's "cross-sec-
tional process model/' 7 Beavers's model has three levels of organi-
zation, from families at the bottom of the scale, extremely chaotic
and confused, with poor boundaries and no hierarchy, to a second
level, which is extremely authoritarian, to a top level that is flexible

and adaptive, neither too loose nor too strict.

Beavers draws his model in the form of a sidewise A, with the end
of each bottom leg representing, respectively, "centripetal" families
on the one hand, and "centrifugal" on the other (see Figure 5.1). The
centrifugal group is said to produce sociopathic behaviors while the
centripetal group produces psychotic disturbances. Confused com-
munication, unclear boundaries, and avoidance of power issues cha-
racterizes both these poorly functioning groups. In the mid-section
of the A, where the two styles begin to meet, there are behavior
disorders on the centrifugal side and neurotic behaviors on the other.
However, the model allows for a "continuum of competence," and
the midrange also begins toshow evidence of hierarchy, even though
the structure is dictatorial. At the top of the A, the structure presum-
ably allows for even better functioning by family members.
This model implies a passage from less to more workable struc-
tures. Beavers argues that it is only a step from chaotic to authoritar-
ian forms, and that a flip from one state to another is common and
may also be therapeutically useful. Haley is close to Beavers here
when he advocates in his recent book, Leaving Home, a therapeutic
approach based on getting the parents of "crazy young people" to
behave like virtual tyrants. 8
Another process model, one slanted more toward normal families,
is Kantor and Lehr's typology of open, closed, and random struc-
tures. 9 The model is different from Beavers's because it does not
conceive of dysfunctional families as belonging to different groups
or levels but as flawed variations of
normal types. Kantor and Lehr
assume that families can be categorized according to a choice of
different homeostatic ideals, or ways to approach equilibrium and

97
 Foundations of Family Therapy
Severely
Dysfunctional Midronge Healthy

Figure 5.1
Beavers's Cross-Sectional Process Model
source: R. Beavers, Psychotherapy and Growth. New York: Brunner/Mazel, 1977, p. 96. (This figure has been
modified by Beavers; it will appear in other, as yet unpublished, articles.)

change. The family structure derives from the type of homeostatic

arrangement adopted and is not an invariant piece of architecture.
The flawed forms also derive from that ideal, and differ accordingly.

Kantor and Lehr's three family types resemble the familiar politi-
cal categories of authoritarian, anarchistic, and democratic regimes.
The closed family is highly structured, hierarchical, and rule-gov-
erned; the individual is subordinate to the group. In its flawed
version it becomes a rigid, hollow shell, and if a runaway develops,

98
THE CONCEPT OF FAMILY PARADIGMS
this shell may crack as individuals become rebellious or violent,
sometimes toward others, sometimes turning feelings inward. The
anarchistic or random family sets a high value on personal in-
dividuation. "Do your own thing" is the motto, and there are few
rulesand little attention to boundaries. In the flawed version this
family becomes totally chaotic; turbulence, caprice, and contradic-
tion take over. Still, struggles of individual members to reestablish

some kind of control may end in a shift to an authoritarian, closed

system, or alternatively, fragmentation and dispersal will occur, or
outside authorities will move The democratic or
in to take over.
open system, which seems a golden mean between the two other
styles, balances order with flexibility and the rights of the individ-
ual with those of the group. In its flawed version this family type
tends toward schism and divorce. Its most characteristic stress
comes from the bind that results from taking features from both
closed and random systems which, if not compatible, can lead to
strain and impasse. Kantor and Lehr do not claim that these types
exist in a pure form, but they do believe that families tend to cluster
around these three different categories.
What is developing more and more is a process typology that,
because it emphasizes movement and change, does not depend on
fixed categories and does not assign negative characteristics to fami-
lies as functional/dysfunctional typologies and typologies attached

to symptoms of individual family members do.

Of particular interest in this respect is Reiss's later thinking about
family paradigms, which is reflected in "The Working Family: A
10
Researcher's View of Health in the Household." First, Reiss is

continuing to examine family worlds in a truly systemic sense. In

his research he is continuing to use such variables as the shared
ability of family members to perceive complex configurations when
faced with strange data; the ability to coordinate responses when
trying to make sense of the material presented; and the ability to
delay closure until enough information has been received and
passed around to guarantee the best possible response. Of course his
experiments test family reactions to an unfamiliar environment, so
it could be argued that this does not meet the family paradigm on
its own home ground, when it might look very different. But Reiss
obviously assumes that the strength of such a paradigm is precisely

99
 Foundations of Family Therapy
to help family members cope with the different and the strange.
Second, he is moving away from categorizing families on a scale
from functional to dysfunctional, preferring to see dysfunction in
relation to each family's idiosyncratic
paradigm rather than to judge
it in the light of preconceived ideas of health and illness.

Third, and perhaps most important, is Reiss's scrutiny of what

happens when a paradigm breaks down. He does not believe that
this is ipso facto a bad thing, although he is alert to destructive
consequences. A central issue for him is how family disorder and
breakdown may create the opportunity for its own self-healing
potential. As he says,

More speculatively propose (recognizing both theology and Pollyanna in

I

this proposal) that family crisisfills a positive function in the life of every

family. Though filled with risk, it ultimately opens the family to new
experience, altering their sense of themselves and the outside world and
thereby transforming a paradigm which may have guided them for years
or even generations. 11

This amounts to taking an evolutionary position on paradigm

change.
My own evolutionary position suggests a somewhat fantastic
model that I call Spiral Platters of Family Organization. I owe the
idea for this to a discussion with Paul Dell in Atlanta in 1978. It was
he who first explained to me the thinking leading to an evolutionary
rather than a homeostatic model, thereby seriously influencing the
evolution of my own thinking.

Spiral Platters of Family Organization

For a long time the idea of a typology of families ranging from

too richly joined to too poorly joined, in Ashby's sense, has been
an intriguing one. We may speculate that a family could not exist
at either extreme of such a range. were too tightly interlocked,
If it

it would not allow were too fragmented, it would

for change; if it

be in danger of dispersal. Most families are found toward the middle

of the range, and few are pure examples of either category. All

IOO
THE CONCEPT OF FAMILY PARADIGMS
families must have some structure, no matter how primitive, and all
families must be able to experiment with change.
The best measure of a family that seems to be working well,
whatever its basic category, is whether it can go toward either pole,
depending on what is useful. The "enmeshed" structure that is
mandatory during Thanksgiving dinner may be totally inappropri-
ate the day after, when the teenage son asks if he can eat supper at
a friend's house. Life stages are important considerations, as are
stages of the career of an individual. Family therapist Carl Whitaker
has spoken of the "rotating scapegoat" in describing how each child in
a family, on reaching adolescence, will take turns being the family
problem. He suggests that as long as people take turns and one person
doesn't get stuck in this role, it is not necessarily a bad thing but mere-
ly indicates that the only way this family can let go of its children is

for the children tobecome temporarily impossible. An example relat-

ed to individual careers would be for a wife whose husband is taking
his law exams or medical internship to cope with her loneliness and
depression by having migraines, rather than openly express her frus-
trations in a manner that could endanger the young marriage.
But the main problem with continuums or grids is that they give
no clue to why a family or person undergoes a sudden change. It
often happens that a family or group at the very fragmented end of
our supposed continuum produces individuals who will suddenly
snap into an extremely "enmeshed" format, characterized by shared
fantasies, paranoid reactions, and delusions of grandeur. One has
only to think of the strange leap taken by the German nation (or
a subgroup within it) after the economic and social collapse of the
Weimar Republic into the world's largest "folie a societe," the Third
Reich. One is reminded, too, of the present-day children from ap-
parently permissive homes who become reborn as "children of the
cults." Another example, perhaps somewhat different, is the tend-
ency of boys from extremely fragmented poor families to form
highly authoritarian and structured street gangs.
Cases such as these show clearly that what most typologies
lacked was any portrayal of movement. Beavers's process model at
least implied a passage toward more (or less) evolved structures, and
made a place for the possibility of change. Dell had been arguing
for some time for an evolutionary model in describing family

lOl
 Foundations of Family Therapy
systems, and Reiss's recent paper forcefully backed this concept.
These different ideas suggested a series of platters in a spiraling
cascade. Families could be seen in terms of groups with contrasting
or mixed characteristics, each group clustering at a different level of
being "evolved" (see Figure 5.2). It could be argued that the enme-
shed/disengaged (or centripetal/centrifugal) group would be at the
low end ofa set of levels ranged according to adaptiveness. Families
representing an extreme of either category would fall off and be-
come extinct, or the individuals would tend to fail or die. Alterna-
tively, as we have noted, a person from a centrifugal family might
flip suddenly to the authoritarian side of the next level up. Going
in the opposite direction, a family on the bohemian side might
produce individuals whose families showed up in the level just
below, as fused or enmeshed. In addition, on each level, there is a
possibility of moving from one style to another; some families oscil-
late between centripetal and centrifugal according to circumstance
and need, and some portray a mix, with one grouping in the family
very rigid, the other very loose. The variations are endless.
The reader will see that an intriguing concept is embedded in this
model, the idea of discontinuous change. Development can, of
course, consist of a move from one side to another on the same level.

But to go from one platter to another involves a reorganization so

total as to represent a discontinuity.
Another important concept is implied by the spiral. Movement is
never really circular, since even the most static cycles or dances in
families never come back to line one. The word "spiral" suggests move-
ment that is open-ended. Even if a family is caught in a very narrow
range on the centripetal edge, there may be either movement toward a
mixed state on the same platter, or a leap to a radically new and differ-
ent type of organization on another platter, in a sort of double helix.
Unfortunately, this figure has an intrinsic drawback. It suggests
a set of stairs to Heaven, paralleling the many other attempts to
show that nature follows an ineluctable path toward ever-higher
forms. In psychotherapy, one finds the same idea in theories that
emphasize journeys of the soul toward final goals such as autonomy
or self-actualization. It would be too bad to create a typology of
families based on this same model of infinite progress.
Rather, I would like to see my platters linked together in a cosmic

102
 THE CONCEPT OF FAMILY PARADIGMS

Autocratic Democratic

Anarchistic Authoritarian

Centripetal

Figure 5.2
Spiral Platters of Family Organization

ring, with families or persons located on the "highest" platters able

to skip to positions on the "lowest" platters —even temporarily—on
the supposition that too much perfection stultifies and that one
must sometimes go back to the original chaos that the Greeks be-
lieved existed before the world began. There something hopeful
is

about the pessimistic ideas of the Swiss physicist Roland Fivaz, who
agrees with recent theories suggesting that living systems at the —
price of using up energy and producing wastes —tend to reorganize
at evermore sophisticated levels of complexity, toward more and
more negentropy, in fact. 12 But he offers the happy thought that the

103
 Foundations of Family Therapy
higher the level of complexity achieved, the larger the amount of
entropy (noise, disorder, pollution) it will give off. Past a certain
point, the evolution of a living system (or eco-system) may be
limited, because it will be swallowed up in its own garbage. Then
the whole process will have a chance to start afresh.
A final proviso is that when making any sort of typology we
violate the richness of combinations offered in nature. These designs
are artifacts like Plato's Ideal Forms, which are more useful in imagi-
nation than in reality. It is impossible to prove that a family falls into

this or that category, even if we agree that these categories exist, and
the chances are that any given family will be fluctuating from one
category to another, or have found own unique combination.
its

This disclaimer offered, we will lean the other way and say that,
at the very least, these designs build on and integrate a queer lurch-
ing research, conducted mostly by nonscientists during the past
quarter of a century. One sighs wistfully at the crudeness of the
constructions, hoping that the hypotheses they are based on will
one day be subjected to more rigorous tests.
In particular we must guard against making categories to describe
optimal organization. The "environment-sensitive" family is a con-
cept which, since it assumes normalcy, implies that once a family
is so designated it "possesses" a quality called normality. This is a
grave error, and we have Bateson to thank for continually remind-
ing us to beware of such mistakes. The attributes of a family will
differ according to which point in time (on the family life cycle, for
instance) or space (the particular arena or situation) one views it.

Whether there are visible problems or no visible problems in a

family may merely reflect where the family currently is on the com-
binatory axis of our grid or in the leaping spirals of our dance.
Having satisfied the need for some scheme of placing families, we
can now begin to examine the ideas of researchers interested in one
type of family: the family which we have called "consensus-sensi-
tive," "centripetal," or "enmeshed." The great amount of research
on this category reflects not just interest but availability of such
families for study, since they are a clinically captive population. The
next chapters will emphasize close analysis of the structures and
sequences in these families, and will try to link formal aspects of
family organization to symptoms and distress.

104
Chapter 6

The
Pathological Triad

From Communication to Structure

In i960 John Weakland, one of the original Bateson group re-

searchers, proposed a formal version of the double bind hypothesis
involving three-party interaction. This was and her-
a major shift
alded a growing interest in family structure rather than communica-
tion as a matrix for symptomatology. In his article 'The Double
Bind Hypothesis of Schizophrenia and Three-Party Interaction/'
Weakland observed that according to the original double bind the-
ory, the recipient of the double bind was said to be getting obscurely
conflicting messages on communication on which
different levels of
he was not allowed to comment, nor could he leave the field. 1
Weakland then pointed out that this person might be getting ob-
scurely conflicting messages from at least two people, on which he
was not allowed to comment, nor could he leave the field. This
would be the case with a young schizophrenic and his parents.
Weakland says:

105
 Foundations of Family Therapy
Clearly, parents can, on a given matter, give conflicting messages to a
child. Clearly, it is important for the child, who is in an overall or collec-
tive sense more dependent on both parents than on one, to deal with the
by dealing with the inconsist-
conflicting behavioral influences resulting
ency of these messages. But, equally clearly, one or both parents may
also be giving messages that conceal, deny, or inhibit exploration of the
inconsistency. }
. .

This track allowed Weakland to link the double bind theory with
other branches of research on schizophrenic communication, mainly
that of Wynne and Singer, which focused on the denial, ambiguity,
disqualification, and inconsistency of messages given out by all
members of a family with a schizophrenic. 3 He also cites Lidz's
description of "marital skew/' in which two parents put out the
myth of total harmony and agreement, in spite of masked indica-
tions to the contrary. 4 Weakland also advanced the thesis that the
three-party version of the double bind can be generalized to social
groups other than the family. As evidence, Weakland mentions an
important study, to be examined in detail later: The Mental Hospital,
published in 1954 by the psychiatrist Alfred Stanton and the soci-
ologist Morris Schwartz. 5 In this book, the authors found that agita-
tion in a patient on a ward was associated with conflicting attitudes
and directives about his or her behaviorfrom two authorities who
at thesame time denied their disagreement and defined their overall
position as one of benevolence. Weakland quotes the authors in
describing the position of the patient: "The two most immediately
important persons in his life were, so to speak, pulling him in
opposite directions." This, of course, could apply equally well to a
child whose parents were in covert disagreement about him.
Haley, too, was pulling out of the intense focus on the dyad
which was such a cornerstone of the double bind hypothesis. He
began to look closely at triads, or, as he began to call them (follow-
ing current research on decision making in small groups), "coali-
tions." Haley observed that in families with a symptomatic mem-
ber, the triad that surfaced most frequently was a coalition between
two people, usually of different generations, at the expense of a
third. Simple alliances not involving a third party were rare or did
not persist. Haley noted that one might find a mother speaking for
a child in a way that discredited the father, or the parents might

106
THE PATHOLOGICAL TRIAD
fightwith each other and then turn on the child and accuse him of
causing their difficulties. Worse yet, if such a coalition were to be
labeled or made overt, it would be denied or disqualified. 6
In a later essay that describes the evolution of researchers' ideas
during the life of Bateson's communication project, Haley states
that in the beginning, schizophrenia was described as a reaction to
a learning context in which levels of communication were habitu-
7
ally confused. The type of message or repetitive interchange that
embodied this confusion was called a double bind. As a result of
exposure to such a context, a person would tend to qualify his
messages with an indication that he was saying or doing something
else, and then add a further qualification which contradicted that
one. Thus, as Haley notes, schizophrenic behavior could be seen as
a disorder of levels of communication.
After Haley began to study coalition behavior in families with
schizophrenics, he changed his position and began to suggest that
problems arose not only from a confusion of levels of communica-
tion, but from a confusion between levels of a relationship system. In
an important 1967 paper, 'Toward a Theory of Pathological Sys-
when warring factions exist in
tems," Haley takes the position that
a family or kin member may find himself in the
network, a family
predicament of being punished for choosing sides (since whomever
he did not side with might exact a penalty) and at the same time
8
being punished for not choosing sides. Thus in such a case it might
be necessary for a person to disqualify all his communications.
This idea, like Weakland's, made the irrational behavior of in-
dividuals the consequence of the social structures they inhabited,
not simply a consequence of confused or contradictory messages.
This shift brought family therapy to a new place. Family theory
could no longer be dismissed as some kind of linguistic metapsy-
chology but could be fairly placed with other contemporary behav-
ioral research. In this and other chapters we will link up some of this
other research as it seems relevant. Here we will look specifically at
research on triangles in social groups: the work of Theodore Caplow
on coalitions, James Davis's "cross-pressure hypothesis," Kenneth
Boulding's "avoidance-avoidance conflict," and the like. In addition
we will take up studies of triads in kinship networks and variants
of conflictual triads in other cultures. But let us begin by examining

107
 Foundations of Family Therapy
Haley's description of pathogenic triangles and the part they play
in maintaining family balance.

The Perverse Triangle

In 'Toward a Theory of Pathological Systems/' Haley comments

on a triadic structure that he believes will always cause distress in
a social system. 9 He calls this the "perverse triangle," or cross-
generation coalition, and observes that seems to coincide with
it

undesirable manifestations such as violence, symptomatic behavior,

or dissolution of the system. The characteristics of this triangle are:
1. Itmust contain two persons from the same level in a status
hierarchy and one person from a different level. In the family, this
would mean two members from the same generation and one from
another generation.
2. It must involve two who are on different levels
a coalition of
against the left-out one. A must be made between an
distinction
alliance, which can be based on common interests and not involve
a third party, and a coalition, in which two people join together
against or to the exclusion of a third.
3. The coalition against the third person must be kept hidden.
That is, the behavior that indicates that such a coalition exists will
be denied at the metacommunicative level. In sum, Haley says that
the perverse triangle is one in which the separation between the
generations is breached in a covert way.
Generalizing from the family to other social systems, Haley ob-
serves that in organizations, too, there is likely to be trouble when
a superior secretly joins with a subordinate against a peer. The
manager of a company may not play favorites with his subordinates
and expect to be an effective manager. Similarly, a parent-child
coalition not only undermines the authority of the other parent but
makes the authority of the favoring parent dependent on support
from the child. The inability of parents of an emotionally disturbed
child to act together to enforce discipline is a reflection of their
inability to maintain the generation line.

108
THE PATHOLOGICAL TRIAD
Theodore Caplow, in his book on coalition research, Two Against
One, discusses a similar type of triad which occurs in hierarchical
10 he
organizations. The "improper coalition/' as calls it, is defined
as any three-person coalition that increases the power of one supe-
rior while undermining the legitimate authority of another. An
example is a triad in which A is as strong as B and C combined, and
B and C are of equal strength. The organization containing such a
triad will find it hard to function because of a built-in instability.
As it is set up, A cannot form a coalition with either B or C that
cannot be undermined by the other, yet B and C together cannot
dominate A.
Caplow describes another organizational triad that is closer to
Haley's perverse triangle. In A is stronger than B, and B is stronger
it

than C, but B and C together are stronger than A. The only way A
could forestall a coalition between B and would outweigh C that
him would be to form a coalition would be
with B. Such a coalition
appropriate and would uphold the status order. But since B could
always threaten to make a coalition with C, and C could always
provide the temptation, A could never be sure of his authority. B
too would be stronger than C within that dyad, but not within the
triad, where he would be as much at the mercy of C as A would be.

Neither A nor B nor C would exercise power with any confidence,

and the triad might find it difficult to function at all.

One could argue that this is not formally the same as Haley's
perverse triangle because Haley is talking about two same-level
superiors and one lower-level subordinate. But we very often find
that two authorities will join in representing the interests of an
organization, even though one of them is higher in status, has the
final say, or culturally carries more weight. Thus A and B in Ca-
plow's second triad might be represented by two parents when one
parent's authority is slightly greater than that of the other; or by a
manager and a foreman in a factory, where the two levels are dis-
tinct but both represent the authority structure.
What is strange is that Caplow does not arrive at the idea of a
perverse triangle in his chapters on family triads. He seems almost
to get at it in his description of the improper coalition, but then, in
considering coalitions in family life, never links this form with
family disturbance or individual distress. He discusses a normative

109
 Foundations of Family Therapy
progression of coalition choices in a typical nuclear family, but does
not seem to realize that cross-generation coalitions can at times
Thus he gives an example of a four-person family
create difficulties.
inwhich the father (A) carries most weight; then comes mother (B),
somewhat less dominant; then son (C), who is the older sibling; and
daughter (D), who is the youngest and weakest of the family. Using
a game-playing analogy, with each person out to "win," he assumes
that the player with the crucial decision to make is mother. If she
joins father, the parental dyad will dominate the two children. This
he calls an example of a conservative coalition pattern, because the
appropriate authority structure is upheld. If she chooses to align
herself with son, creating a "revolutionary" coalition pattern, this
will upset the appropriate authority structure. In either case, it is

mother who has the strongest hand.

This line of reasoning is persuasive but carries some assumptions
that should be examined. The implication that family members, like
the players in the experimental coalition games Caplow discusses
elsewhere, have different "weights" and will make alignments to
improve their positions relative to each other is questionable. The
experimental games are done with strangers, not with persons living
in ongoing groups. Caplow talks as though families act like subjects
in experimental games. Thus he can assume that in the "revolution-
ary" family the father has "no reason to join an AD coalition; it

would not enable him to dominate the BC coalition of mother and

son together, and he can already dominate them separately."
This statement further assumes that what drives each individual
is a desire for power or domination. One could argue that in a family

this may be one element but hardly the entire story. The idea that
the self-interest of the individual might at times be subordinated to
the survival of the group, and that his coalition choices might be
influenced accordingly, is never examined. In fact, a father-daughter
coalition in Caplow's family would be a strong probability —even
though it might not outweigh the mother-son coalition with the —
presumed result of helping to balance the parental axis. The family
with "a boy for you and a girl for me" is not just a song writer's
fancy but is common and, pushed to an extreme, indicates difficul-
ties. Similarly, a son who is drawn into closeness with a neglected
mother may not be acting out of a wish to "win" against father, but

HO
THE PATHOLOGICAL TRIAD
out of a desire (not necessarily conscious) to comfort mother, to
protect father from demands that father may not be able to fulfill,

and to act as a buffer in marital strife. All this may enter into his
behavior quite as much as any presumed personal power he may
derive from his attachment to the authority figure.
Caplow himself seems uncomfortable with his focus on the "need
to win" as a major factor in individual decision making. In his
treatment of the family he often takes a leap to the system-wide
level. One rule of triadic relationships, as he observes early in the
book, is that a person cannot be both a coalition partner and an
opponent same network. When he begins to apply this rule
in the
to family groups, a whole new framework appears. If daughter sides
with mother against father, she will not be likely to side with
grandmother, who is on father's side against mother, because this
would subject her to the pressures of divided loyalties. According
to this line of reasoning, alliances are not necessarily based on
motives originating inside the individual but on what Caplow calls
the "strain toward compatibility" across a social network.
Haley is ahead of Caplow in deliberately trying to formulate a
systemic rather than an individually oriented explanation for coali-
tion behavior. He refuses to accept the traditional decision-making
model with its assumptions about interior motivation. He suggests
instead that a conflictual situation arising within different orders of
coalitions will interfere with the proper functioning of a social sys-
tem and that of persons in it. And in describing this, Haley uses a
quite different metaphor from the game-playing metaphor: the
analogy of the Russellian Logical Paradox, with its apparent conflict
between different levels of meaning. Both Haley and Bateson,
whose ideas were at one point intertwined, use this analogy, but
Haley extends it to explain how membership in coalitions within an
organization can under special circumstances put great stress on
certain individuals.
Haley describes this situation as follows: Normally, peers in an
organization are in coalition with each other, a fact expressed by the
presence of administrative levels or, in a family, generation lines. If

an employer joins with one employee to form a coalition against

another, the latter is faced with conflicting definitions of his posi-
tion. Within the larger framework of the organization, his fellow

111
 Foundations of Family Therapy
employee is aligned with him. But at the same time that employee
is siding with a superior against him. As Haley says, "Being forced
to respond when there is a conflict between these two different
orders of coalitions creates distress/' 11 There is a clear similarity
between this formulation and Caplow's insistence on the impor-
tance of compatibility between coalition partners in a group, but
with the difference that Haley links conflictual coalition member-
ship with individual distress. He is beginning to assign more and
more weight to a conflict between loyalties at different levels of
social organizations as a factor in the etiology and maintenance of
disturbed behaviors.
To restate the problem in other terms: Assume that A offers
behavior which fits the operating needs of System One (which

might be the person or might be a subsystem like the parental dyad),

and that conflicting behavior is simultaneously demanded that fits
the needs of System Two (which might be the extended kin group).
If one system is a subset of the other, and the two arenas are not
clearly differentiated or kept apart, confusion will arise. It is the
same kind of confusion that Bateson saw as the heart of the Russel-
lian paradox: the difficulty of discerning the difference between
class and subclass. There will always be a two-level dilemma, never
clearly spelled out, regarding which system's rules are to be obeyed.
For instance, a child, who at home is not required to behave in
any special way at the table, is required to sit still and display
"manners" during a long Sunday dinner at the home of his mother's
parents. He gets confused signals from the overangry scolding of his
mother (much of whose anger is really at her own parents) and
contradicting suggestions from his father (who hates social eti-
quette) that he be allowed to leave the table. The situation escalates
until the child has such a bad tantrum that the parents decide to
leave early and go home. Here the demands of conflicting behavior
from different people and subgroups in the family are clear, as are
the different levels on which these demands are made (for instance,
the mother tells him to sit still, but by overreacting implies that she
really does not agree with what she is telling him).
Haley's leap from a fascination with communication to a fascina-
tion with structure was a crucial one. It allowed him to begin to
consider the possibility that schizophrenia may be, as he put it, the

112
THE PATHOLOGICAL TRIAD
12
result of a conflict of groups. That is, a person exhibiting schizo-

phrenic behavior may be trying to please conflicting groups in

which he has simultaneous membership. This led Haley to study

the formal characteristics of alliances in the schizophrenic's family.

At the Nexus of Warring Triangles

Counting the number of triangles in an average-size extended

family, where there are two parents and two children and each
parent has two parents, Haley noted that any one person in this
group is involved in twenty-one triangles simultaneously. If every-
body is living together harmoniously, there is no problem. But if a
child is at the nexus of two triangles or groups that are in conflict,
he will be in a difficult position. If his mother and maternal grand-

mother are in conflict with his father and his father's mother, he will
have to behave carefully, because if he pleases one group he will
displease the other. If all twenty-one of the triangles the child
inhabits are in divided states, he will have to exhibit conflictual
behavior in order to survive. Such behavior is often considered
crazy or strange.
The dilemma of a person caught between two positions is echoed
by James Davis's "cross-pressure hypothesis." 13 Davis asks us to
consider a person (P) aligned positively with two others (O l7 O z )
but undecided about a fourth person (X). Suppose X is a candidate
for office who is supported by 1
but not by 2 This puts P in a .

quandary. If P votes for X, he will please O x but displease O z If he .

does not vote for X, the reverse will be true. One can diagram this
situation in triadic form (see Figure 6.1). If one assumes that and1

O z represent political blocs as well as themselves as individuals, one

can say that P is under cross-pressure from simultaneous member-
ship in loyalty groups. Clearly much of Davis's thinking comes out
of structural balance theory, which is very close conceptually to
coalition theory and will be discussed in the next chapter.
w Boulding, in his book Conflict and Defense, makes a similar analysis
of what he calls an "avoidance-avoidance conflict." 14 Using the

113
 Foundations of Family Therapy
analogy of an ass between two skunks, he describes the situation
7
as a "trough equilibrium/ with the ass being pushed toward one of
his negative goals every time he moves away from the other. The
ass is in what Boulding terms a stable psychological conflict, or
7
"quandary/ In Boulding's opinion, if the ass (representing a human
being) cannot jump over one of the skunks or otherwise escape, his
behavior may become disjointed and random, and he risks a ner-
vous breakdown. An alternative route is to retreat into the realm of
daydream, fantasy, art, or, in a more pathological manifestation,
schizophrenia or paranoia.

These contributions from different sources show that a remark-

ably similar group of ideas has been explored by different research-
ers. The core of these ideas is that symptomatic or bizarre behavior

can be associated with a situation in which a person is forced to

choose between two paths, each of which carries a penalty. It must
be said, however, that not all of these writers are speaking from the
same premises. There is a fundamental argument over whether the
conflict is between choices on the same level, usually defined as
"ambivalence," or choices on different levels, and herein lies a dis-
tinction of considerable philosophical import.

1. 0<9

I. P votes for X, triangle P-O^-X is congruent but P-O2-X is not.

2. P votes against X, triangle P-0 2 -X is congruent but P-O r X is not.

Figure 6.1
Davis's "Cross-Pressure Hypothesis"

114
THE PATHOLOGICAL TRIAD

Ambivalence versus a Conflict of Levels

Caplow defines ambivalence as "an emotional strain which arises

from interaction with someone who is both an opponent and a
15
coalition partner/' This is essentially a restatement of Davis's
cross-pressure theory in a three-person rather than a four-person
form. The double bind is also popularly equated with ambivalence,
but both Bateson and Haley declare that the contradictory situation
described by the double bind is something quite different. Haley
says that ambivalence as a psychological term refers to the state of
mind of an individual when faced with choices of equal or near
value about which he has mixed feelings. 16 By contrast, he states,

the double bind has to do with a conflictual context within which

it is difficult to function. Haley, following the ideas of the Bateson
group, gives a paradox as an example: someone who is asked to do
something spontaneously. If he does something, it is not happening
spontaneously, and if it is truly spontaneous, how can he be obey-
ing the request?
Bateson, the originator of these ideas, puts the matter somewhat
He describes two types of internal
differently. contradiction which
are commonly found in human communication. 17 One, defined as
"ambivalence/ will occur when two same-level Gestalten, A and
7
B,
are negatively and positively perceived, respectively. However,
they overlap in such a way may be
that the overlapping section
viewed positively when perceived as part of A, and negatively when
The viewer cannot perceive the same section
perceived as part of B.
positivelyand negatively at the same time, but there is nothing to
prevent him from peacefully experiencing whichever Gestalt he
happens to be settled in at the time.
Bateson's second type of internal contradiction involves two diff-
erent levels, one of which is inclusive of the other. Bateson cites as
an example the paradox of formal logic (for example, the sign on
which is written, 'This statement is untrue") and adds that these
forms are "systems of contradiction in which temporary preference
for one pole promotes preference for the other and vice versa." 18 As
a more concrete analogy, Bateson takes the electric bell buzzer. He

115
 Foundations of Family Therapy
describes the way upon an
the electromagnet in the buzzer acts
armature so as to cause an oscillation between two positions which,
for the purposes of his illustration, he labels "yes" and "no." The
armature goes from side to side because the implications of the
"yes" side send the armature back to "no," and vice versa. But
Bateson emphasizes that the "yes" and "no" belong to two different
levels of abstraction: "Yes" refers to position of the armature, and
"no" refers to direction of change. This oscillating motion may be
equitably endured by a bell buzzer, but Bateson points out that it

is hard on humans who are in an analogous position.

One final point Bateson makes about paradoxes has to do with
time. In thinking about a paradox, a person will at first accept the
"yes," but further pondering will drive him to "no," and so on. Thus
from the psychological point of view, the time aspect is important;
we are not dealing with a problem of static indecision, Bateson says,
but one of alternating durations. This time element will bear further
thinking about; for now, it is sufficient to consider that the length
of time a person is allowed to rest at either end of two incompatible
poles could have something to do with the degree of stress he would
experience. One can imagine a situation where this time period
became shorter and shorter, following a progressively faster se-
quence, until incompatible demands are being made on the individ-
ual almost simultaneously. At this point one might expect some
kind of sudden shift, cutoff, or breakdown.
In a covert cross-generation coalition there may be an effect simi-
lar to the bell buzzer. The implication of a child siding with either
parent may be harmful to the system at a level other than the one
where the person is being asked to take sides. What is missed in the
analogy of the ass between two skunks, or in any other model for
a same-level dilemma, is that in family alignments there is a whole
vast web of systems and subsystems which will often be disar-
ranged, whichever side is taken. It is also possible that it will be
disarranged if no side is taken.
The child in a pathological triad, in accepting a coalition with one
parent, may counteract the too-great power of the other parent, but
in doing so he not only undermines the authority of the governing
dyad but may upset the balance of power between the two extended
kin groups. If, on the other hand, he tries to remain close to each

116
THE PATHOLOGICAL TRIAD
parent, and both are negatively involved with each other, he is
caught in a conflict of loyalties. Add the time factor, which suggests
that there may be a point at which judicious alternate siding may
achieve a fatal momentum, and you have a situation where extreme
reactions might occur.
The question that follows is what sort of behavior can be ex-
pected when a person is in the position described above. The answer
brings us back to the communicational model but within the con-
text of levels of organization.

Schizophrenic Communication as a Way of Not Defining Relationships

Haley describes schizophrenic behavior as an effort to avoid

defining one's relationships with other people. 19 Usually, Haley
says, people qualify the statements they make to one another with
metamessages which indicate the following:
1. I

2. am saying something
3. to you
4. in this situation.
Analyzing a transcribed conversation between two hospitalized
schizophrenics, Haley illustrates the many ways in which two de-
termined people can deny any or all of these types of metamessages.
Each of the comments by the two participants is of the variety: This
is not a hospital, it is I am not a patient, I am a man from
an air base;

outer space. In addition, by non sequitur responses to the other's

comments, the validity of these comments is denied. Thus do these
speakers avoid defining their relationship to each other. They deny
that they are speaking, deny that anything is said, deny that it is
said to the other person, and deny that the interchange took place
in this place at this time.
Haley offers an example of a similar response in a family context,
citing the case of a schizophrenic girl who had just been released
from the hospital. 20 On her return, her parents quarreled, and the
mother took the daughter with her to her own mother's house. The

117
 Foundations of Family Therapy
daughter called her father as soon as they arrived to tell him where
she was. When her mother complained that she was siding with her
father, the daughter explained that she had to call him because
before they left she had given him an "odd look" (one of her
symptoms). The where his daughter and wife
father, finding out
had gone, came and persuaded the mother to go back with him.
Before leaving to go home, the mother asked the daughter to go on
an errand; the girl refused and the grandmother went instead. While
her parents were in the next room discussing her refusal, she began
to scream and was subsequently rehospitalized.
In looking at the larger family picture, Haley notes a number of
cross-generation coalitions. The girl's two grandmothers competed
for her loyalty; the father's mother sided with him against his wife;
and the girl sided with her mother's mother against her own mother.
The parents each sought the girl's favor, and each accused her of
siding with the other. Haley points out that if one takes these
circumstances into account, any response on the girl's part could be
seen to carry a penalty. Allowing for this peculiar context, then, the
girl's communications could be seen as eminently adaptive. As

Haley puts it:

At the nexus of warring family functions, what should be an "appropriate

and normal" response to this situation? It would seem to be one in which
the girl would behave in one way to satisfy one faction and in another way
to satisfy another and then disqualify both ways by indicating that she was
not responsible in any case. Such conflicting communication would be
diagnosed as schizophrenic behavior. 21

Haley is not alone in his observation of the way alignments are

denied in this type of family, not just by the patient but by all
members. Similar examples have been cited in research on "thought
disorders" in schizophrenics and their families, although this term
can be seen as a misnaming. Haley's "The Art of Being Schizophre-
nic" is a tribute to the skill with which schizophrenics caricature the
need for concealment in their families. When the family is busily
22

denying some obvious fact, the schizophrenic member will offer

crazy-sounding statements which teeter perilously close to the
truth. At the same time, if someone comes too close to the truth, the

schizophrenic will say or do something crazy.

118
THE PATHOLOGICAL TRIAD
For example, in the case described by Wynne in Chapter 1 (the

case of the catatonic girl), the daughter parried her father's inappro-
priate advances to her by saying, "Petting between teenagers some-
times gets out of hand." The father replied, in equally evasive
fashion, "A father's love is a father's love, a wife's love is a wife's
23
love, and a daughter's love is a daughter's love."
Wynne also remarks ruefully on the tactics of the family when-
ever a therapistwould try to comment on the splits and alignments
that were being so studiously denied. The daughter would answer
such a comment with something like, "If you try to do right, you
don't have to worry," and then add, "I think someone may have
done something to my body." The father would immediately jump
in to ask whether she meant the boy she met at church the summer
before, and the mother would cut across to correct the name of the
boy mentioned by the father. The exchange between the therapist
and the daughter would be effectively disrupted by this kind of
interaction. The therapist might well have begun to wonder
whether this was a pattern of thought disorder or a system of
sabotage.
Clearly, then, denials, ambiguities, covert messagesmust all be
seen within a context where they have a peculiar logic. Such a
context is exemplified by a family in which no relationship with any
family member may be defined or acknowledged without disrup-
tion, stress, or other penalties. This will happen most often when

the organization of the family cannot allow open alliances across the
generation lines because these alliances would threaten important
same-level relationships like (in our society) the marriage. Assum-
ing that this sufficiently answers the question of why these alliances
—which are actually coalitions against others—go underground, let

us move on to a consideration of covert cross-generation coalitions

in the extended family. This will take us back to our initial emphasis
on family structure.

119
 Foundations of Family Therapy

Regularities in Larger Networks

In discussing covert cross-generation coalitions Haley notes that

they never occur alone but always in 24
pairs. In disturbed families
a coalition of a parent with a child will often be matched by a
coalition of another parent with a grandparent. This could be called
a "countervailing coalition/' Haley even states that a parent-child
coalition occurs so often in conjunction with a parent-grandparent
coalition that this hypothesis could be offered: "a breaching of the
generations with a child will coincide with a breaching at the next
generational level." If this is true, says Haley, we can assume that
there are regularities in family networks, that the patterns in one
part of the kin group are formally the same as those in some other
part.
This formulation carries with it an entirely different assumption
about the "cause" of many behaviors which have in the past been
thought to be motivated by an individual's unhappiness or anger.
Haley observes that to examine the question of cause in terms of
somebody's dissatisfaction is to focus once more on the individual.
If one shifts to the larger context, more circular explanations of

"cause" appear. Wife may join child against husband not only be-
cause she is unhappy with husband but because a good relationship
with her husband would have a disruptive effect on her relationship
with her own parents. In this sense, Haley writes, "cause" is a
statement about regularities in larger networks.
In these paragraphs Haley begins to expand on his idea that stress
derives from a conflict between membership in coalitions in social
groups. Caplow's basic premise in describing coalitions is that a
coalition partner may not be an opponent in the same set of rela-
tionships. The argument follows that relationship sets will exhibit
a tendency toward compatibility. If this is true, one would expect
to find a premium placed on strategies which protect a person
against the penalties of responding to incompatible demands. This
may be why formally instituted separations between levels within
social systems are so widespread. In families, as in business organi-
zations, there are many customs and sanctions which enforce a

120
THE PATHOLOGICAL TRIAD
status difference or a generation line. Caplow cites as examples the
ritualized avoidance behavior between persons in delicate kinship
relationships in various societies, and the exaggerated distance be-
tween officers and enlisted men in the army. In some societies joking
behavior is often prescribed between men and their sisters-in-law,
their grandmothers, and their female cross-cousins. This seems to
be a friendly type of avoidance behavior. According to Caplow,
joking seems to deny a dangerous coalition in circumstances where
one might be expected to develop.
A more serious ritual schism is the almost universal "mother-in-
law avoidance" practiced between a man and his wife's mother.
From a coalition standpoint, Caplow says this operates to prevent
an unworkable triad, that is, a coalition in which one or more
members will be forced to assume incompatible positions. In the
husband/wife/mother-in-law triad, the wife is apt to be in a situa-
tion where she must be close to each party at the expense of her
allegiance to the other. There are other considerations too. Caplow
makes the point that

In addition to the hint of inadmissible sexual rivalries, a coalition between

a man and his mother-in-law becomes unthinkable as soon as we identify
it, as we must in linked triads, as a coalition against the father-in-law, the

wife's brother, the husband's mother, and sundry other persons in the
kinship network. 25

The incest tabu is perhaps the most ubiquitous of all the sanctions
against dangerous coalitions. Both Haley and Caplow describe
Freud as an early explorer of family alignments that violate these
sanctions. Of course Freud formulated the problem in a manner
consonant with the individual orientation toward nervous disorders
of his day. Caplow defends Freud's concept of the oedipal conflict
as an instance of a culturally determined coalition found mainly in
modern Western families. Haley reinterprets it as the most generic
example of a secret coalition across generation lines. He further
states that

It could be argued that this pattern is portrayed symbolically as a reflection

of the incest tabu, but one could also argue that the incest tabu is a product
of the recognition that cross-generation coalitions result in distress for all
participants in the family network. 26

121
 Foundations of Family Therapy
From this standpoint, the incest tabu has more to do with making
family structures workable than it has to do with preventing im-
proper sex.

The Perverse Triangle in Different Cultures

We have examined here the idea that when a person in the parent
generation makes a secret bond with someone in the child genera-
tion against the other parent, there will be trouble. The implication
is that no authority structure may have two adjoining generations
in it. This is patently not true. Many women in female-headed
households in our society have workable relationships with their
own mothers and depend on them for help in bringing up their
children. Or they may enlist the help of an older child without
causing pathology.
A look at other cultures is even more enlightening. Caplow men-
tions the important contribution of the sociologist Francis Hsu in
tracing out a typology for kinship and culture that depends on what
he calls the "emphasized relationship" characteristic of that cul-
ture. 27 This relationship would correspond to what I would call the
governing dyad, even though it may not always be formally thought
of as such. Hsu divides cultures into four types, depending on what
this "dominant axis" might be: In Type A, found in many Asian
societies, the dominant axis is the father-son dyad. In Type B,
characteristic of many Western societies, it is the husband-wife
dyad. Type C emphasizes the mother-son dyad, and is exemplified
by traditional Hindu families. Type D consists of a brother-brother
dyad, which is found in many African societies.
One could assume that whatever governing dyad exists in a given
family, it will be workable so long as it is not suborned by coalition
arrangements between one of its members and other family mem-
bers. The term "cross-generation coalition" may be culture-bound
in the sense that strains may arise from same-generation coalitions
where the dominant family type
in societies is organized differently
from our own.

122
THE PATHOLOGICAL TRIAD
emanating from perverse triangles
In other words, the strain lines
will differ depending on the culture. In a study of Hindu family
relationships, Aileen Ross shows that the strain line in this group
lies on the tie between son's wife and son's mother, a relationship

which is highly charged with potential conflict. In this society (or

in it as it used to be), the wife is in a particularly difficult position.
She leaves her family of origin to live with her husband's family
after marriage, and then is under the absolute power of her mother-
in-law. What justice there is operates over time, since the daughter-
in-law can expect redress of her grievances only by living long
enough to be a despot in turn over her own son's wife.
What is illuminating about this example is that in traditional
Hindu society, husband-wife closeness is discouraged as a possible
source of distress to the family system. Aileen Ross describes the
problems faced by a young wife in her new family:

If she was attractive enough to elicit her husband's support, her position

became even more difficult. If he sided with her, the delicate balance of
family relationships was upset, and tensions created which might react
back on her. Her supervision was not given over to her husband for this
might have developed a warm personal relationship between them, which
again might have caused strain in the joint family system. 28

This is one more piece of evidence that the secret breach of

generation or status lines, analyzed as an intrapsychic artifact by
Freud and redefined behaviorally by Haley, is really a theme con-
sisting of many variations. Western society seeks to weld the mar-
ried pair closely, to avoid the dangerous possibility that a child may
enter a subversive coalition with either parent. Traditionally, Hindu
society has sought to place distance between man and wife to avoid
subversion of the all-important mother-son tie.
For other variants one can turn to African society, which the Hsu
typology tells us is generally based on a brother-brother axis. Rob-
ert LeVine, in a study of typical African family tensions, shows that

in many societies the strain line is apt to be between father and son,
who have a culturally prescribed avoidance pattern. LeVine notes
that intergenerational male homicides are the most common form
of murder in many African groups. 29 Part of the reason may derive
from the custom that forces a firstborn male to wait until his father

123
 Foundations of Family Therapy
is dead before he can share any of his father's accumulated wealth.
As there are not too many other opportunities to acquire posses-
sions, this can become a focus for considerable conflict. Bateson
noted that in the Iatmul, fathers and sons were culturally forbidden
to become intimate, and also guessed that this might come from the
mutually opposed interests of the older and younger men. This
opposition is reinforced by the close ties of sons to their mother's
brother's clans. We
have already investigated the connection be-
tween the threat of cleavage between fraternal clans in Iatmul soci-
ety and the social devices that apparently evolved to prevent this.
But we could go further and observe that the problem of social
fission is one that affects all governing structures, and that forms
countering fission in a given culture may have their own potential
depending on the makeup of the governing structure
for distress,
and what forms are used. Some ways to counter fission will be
—
benign an integrative custom like the naven, for instance. Less
benign ways will entail the involvement of a third party, as when
a child's symptom has the apparent result of keeping a pair of
parents from separating or when one of the spouses develops a
symptom, seemingly forestalling such an eventuality in a different
way. If such counteracting behaviors are insufficient, one possibility
is that the "dominant axis" may split along the line of cleavage. In
our society this often means divorce; in African societies it may
mean the splitting off of a fraternal group.
It is important to keep in mind the cultural variants of the "per-
verse triangle." we do not realize that this triad will differ depend-
If

ing on the way families are organized in a society, we will begin to

think that the family structures associated with symptoms in our
society are the only ones. We must also guard against seeing symp-
toms in a linear causal sense, having the "purpose" of saving a
marriage, for example. All we can say is that they are sensitive
responses which are usually associated with relationship dilemmas
in hierarchical groups, and which join with many other interlocking
factors to favor family balance.

Stepping back now to the behavior of triads in social networks,

we go on to deal with an important mini-industry in the field of

social psychology: the theory of structural balance. Much of Cap-

124
THE PATHOLOGICAL TRIAD
low's thinking in Two Against One is heavily influenced by balance
theory, although he only mentions it in a footnote. Haley, too,
seems to be struggling with the idea that there are laws governing
the compatibility of relationships across social networks; and Davis
is purely in the tradition of balance theory. The effort of the next
chapter will be to select useful pieces of this intricate and fascinat-
ing set of ideas, without getting too seduced by its elegant logic.

125
Chapter 7

The Rules of Congruence

for Triads

Balance Theory and Family Theory

The major difficulty with applying structural balance theory to an

understanding of families is that this theory was originally devised
within a framework of individual psychology. The laws of related-
ness described by balance theory were used to promote a widely
shared premise: that a state of "cognitive dissonance" or perceptual
inconsistency creates discomfort which a person will attempt to
correct. The main emphasis was on the focal person's attitudes,
sentiments, and cognitions. As a result, explanations having to do
— —
with his social context his family or other groups became invisi-
ble. However, the theory contains core ideas that offer clues to some

of the rules governing relationships in families and kin groups, and

is particularly applicable to triadic behavior patterns in families
with psychotic members. The patterns that balance theory predicts
widen the possibilities of clinical intervention.
The tenets of balance theory, derived by Dorwin Cartwright and
Frank Harary from research by Fritz Heider, rest on the premise that
linked sets of relationships abhor interior contradictions. 1 Heider's
interest was in cognitive fields, and he hypothesized that such fields

would tend toward a consistency of attitudes or sentiments. He

126
THE RULES OF CONGRUENCE FOR TRIADS
intended to formulate rules of congruence that might govern the
way an individual perceived persons and other entities, whether
material or abstract. Cartwright and Harary stress the difference
between this emphasis and that of researchers like T. M. Newcomb,
who have extended the theory to explain social as well as cognitive
fields. For instance, if Person P liked Person O and Person X, but

Persons O and X were enemies, there would presumably be pressure

on one of the parties to shift his attitude until a situation where
there were no conflicting loyalties prevailed. But this would not
necessarily mean that Persons O and X would have to become
friends. "Balance" is not synonymous with harmony. The rules for
the interpersonal interpretation of balance theory are often put this
way (see Figure 7.1):

Balanced Triads Unbalanced Triads

A A

A
A

/ \
2. B^ ^C B^ ^C

A
A
/ \
/ \
/ \
/ \
3. B^ ^C

A is me j
B and C are the first and
second parties in the proposition.)

Figure 7.1
Balanced and Unbalanced Triads

127
 Foundations of Family Therapy
The friend of my friend is my friend.
i.

2. The enemy of my friend is my enemy.

3. The friend of my enemy is my enemy.

4. The enemy of my enemy is my friend.

This is essentially a theory of coalitions. Under these rules, triads
are balanced, or, as it might be more descriptive to say, congruent, in
two cases: (1) when all relationships between the three possible
pairs are positive (a word defined by balance theorists in terms of
liking, similarity, affinity); (2) in the classic "two against one" situa-
tion, where two of the parties are friends but have a negative atti-
tude toward the third ("negative" being defined in terms of hostil-
ity, opposition, distance). Triads are unbalanced in only two cases: (1)
if there are two positive relationships and one negative one; (2) if
all three relationships are negative. Some reservations about these
terms "positive" and "negative" will be explored later; for now, the
balance theory definitions indicate adequately we
what is meant. If

apply this formula to the P-O-X triad described above, it is clear

that three routes can be taken to make this triad congruent. O might
become friendly to X, or the relationship between either P and O,
orP and X, could become unfriendly (see Figure 7.2).

Unbalanced — the enemy Balanced — the friend

1.
my 2
of friend cannot be of my friend is my friend.
my friend.

Balanced — the enemy Balanced — the enemy

3. X
of my friend is my enemy. of my enemy is my friend

Figure 7.2
Balanced Solutions to Unstable P-O-X

128
THE RULES OF CONGRUENCE FOR TRIADS
To make these propositions more concrete, if Betty and John and
Bill are all getting along, fine. But if John and Betty have a falling-

out, there will be pressure on Bill to take sides, especially if they are
all membersof an ongoing group. If Bill cannot play mediator and
reestablish harmony, or if Bill finds the stresses of a divided loyalty
too hard to take, he will probably end up on one side or the other.
This is a balanced form, the classic two-against-one. Of course, Bill

may not like his position. If so, it would be good if he could leave
the field and say, "A plague on both your houses."
But not everyone in this situation has these options. If Bill is a
child and Betty and John are his parents and drag him into their
quarrel, all his choices will be dismal. He might side with Betty and
lose John, or vice versa. Or he might seesaw between them, always
careful not to alienate either by seeming too close to the other. I
have called this option (the triangle with two positive sides) the
"inadmissible triangle" because of its potential for stress. Alterna-
tively, Bill could create a closeness between the warring pair if he
could worry them by getting ill or behaving strangely. He might also
unite them if he acted troublesome and bad. But none of these
courses is a very comfortable one, even though, technically, each of
them is "balanced."
Balance theory applies not only to single triangles. Cartwright
and Harary take Heider's formulation and extend it in an important
way. They are trying to describe rules that will cover not only a unit
of three but larger configurations as well. Thus they take Heider's
original statement that a balanced state exists among a set of three
entities if relationships between all three are positive, or if two are
negative and one is and expand it
positive, to include units of n
number of elements.
For our purposes Cartwright and Harary's original theorems and
proofs can be restated very simply in terms of triangles. Let us start
with a triangle that is balanced by having one side positive and two
sides negative, and then decide to add any number of points, linking
them together in a grid and still following the rules for congruence.
The result is that two, and only two, coalitions will appear. Two
clusters of points joined by positive lines will be separated by nega-
tive lines, and this will be true no matter how many points are added
(see Figure 7.3). The reader who doubts this may want to play

129
 Foundations of Family Therapy

1^ 7^ B
X /
1

1^
X I

i. B 2. C iJo

3. D 4. D

No matter how many points are added

only two coalitions result.

Figure 7.3
Two Coalitions

//
\ ^N. ^r 1

\\
1/
F

Figure 7.4
Single-Deviant Arrangement

130
THE RULES OF CONGRUENCE FOR TRIADS
around with triangles, adding new points and labeling arcs between
them "positive" or "negative" according to the rules outlined above.
The result will be two opposing groups no matter how large the
structure becomes. The apparent exception is when one of the two
subsets consists of only one point and all the other points are in the
other subset (see Figure 7.4).
The interest for social psychologists and clinicians is that these
two forms occur naturally in families and other small groups. One
often finds forces polarizing a social field so that everybody is on
one side or another. As an alternative form which seemingly coun-
teracts the first, everybody will join together against one member.
This will solidify previously warring factions into a single bloc.
William Taylor comments on these patterns in an article that relates
balance theory to interaction sequences in families in which one
member is symptomatic. He suggests that balance theory may offer
some key to understanding the process whereby a group makes use
of what he calls the "single deviant arrangement" as a way to
prevent a split or civil war. 2
One proviso that critics of balance theory make is that all the
elements of the structure must be connected, otherwise there will
be no pressure for relationships to be compatible. This point coin-
cides with Caplow's assertion that in any set of triads the prohibi-
tion against the same person being an opponent and a partner at the
same time is not true if the people in question meet in different
systems of action. 3 Balance theory seems mainly to apply to ex-
tremely close-knit relationship systems, not to loose ones, collec-
tions of strangers and the like. In addressing himself to this question
James Davis has come up with a formulation which suggests that
"balanced" forms are special cases under an overall rule for what he
calls "clustering." 4

131
 Foundations of Family Therapy

Clustering Theory

Balance theorists have noted that the tendency of balanced struc-

tures to divide down the middle is one explanation for the peculiar
phenomenon of polarization. Simmel commented on this facet of
social behavior many years ago: "Periods of excitement generally
place the whole of public life under the slogan, 'Who is not with
me is against me/ The consequence of this is a division of elements
into two parties/' 5 But there was no set of laws explaining this
activity until the advent of balance theory.
The trouble with applying these laws arises when one asks how
a society can manage to cohere at all if there is a tendency for
polarization to occur whenever a negative relationship appears in
any part. Coser's work on the importance of multiple group affilia-
•k tions provides a partial answer. 6 Coser's point is that an extremely
closed group risks polarization because it does not allow the exis-
tence of differentiated subgroups. If, he says, one conflict cuts
through a society, this will put into question its basic consensual
agreement, thus endangering its existence. On the other hand, if

attachments to many groups are allowed, thus creating a multitude

of competing loyalties, these will act as a balancing force, prevent-
ing a deep cleavage along one axis. Coser notes that in a closed
society, where all loyalties are focused on one or two ideas, objects,
or persons, there is greater danger of polarization (or, one might add,
the alternative of scapegoating) than in an open society, where
many loyalties are allowed.
Davis extends this line of thinking in his article on clustering
theory when he notes that polarization is only one way that a group
can be divided. It is common to find in a society many cliques or
islands of close-knit persons with mild hostility or distance between
them. Davis modifies the basic propositions of balance theory to
include this possibility. Specifically, he changes the rules for con-
gruence to include the negative triad. He states that the first three
conditions cited by balance theorists still hold: a friend of a friend
willbe a friend; an enemy of a friend will be an enemy; a friend of
an enemy will be an enemy. The fourth proposal, that an enemy of

132
THE RULES OF CONGRUENCE FOR TRIADS
an enemy will be a friend, he excludes. An enemy of an enemy may
be an enemy. This last rule may result in a structure that is "clus-
—
tered" containing many isolated groups rather than "balanced."
7
—
To take an example from a kin network, we might diagram two
types of family structure, one clustered and one balanced (see Figure
7.5).
Balanced Family Clustered Family

A B A B

/ \
/ \
\
\
/ ! \
^ \
/
c/" ^»0 c£ *
/ \
1

I
^D
\ / \
\ /
\ /
\ /
\ /

E F

No negative triads allowed. Three generations are independent of each other,

negative triads are allowed.

Figure 7.5
Graphs for Balanced and Clustered Families

In the balanced group, the "negative" arcs usually occur between

parents and children. Between grandparents and grandchildren,
however, the lines are usually positive, meaning that these are rela-
tionships of particular warmth. This is a common situation for
three-generation families in our culture. Apple has made cross-
cultural studies which indicate that if the parents have primary
responsibility for disciplining the children, there will be an attitude
of respectful distance between these generations, while the grand-
parent-child relationship will be one of friendly equality. 8 If, as in
a few cases, the grandparents wield authority, the "negative" status
between them and the children while the child-parent
line will exist
relationships will be more companionable. Kinship terminology
often enshrines these differences in terms like "cool" and "warm."
As in balance theory, relationships across such a network will tend
to be compatible: the warm relatives of my warm relatives will be
warm to me, etc.

On the other hand, a "clustered" kin group of three generations

would be depicted by a structure in which same-generation persons

133
 Foundations of Family Therapy
were positively aligned with each other and there would be "nega-
tive" arcs between cross-generation persons. The difference would
be that the arcs skipping a generation would also be "negative."
Thus you would have a set of three generations, none of which is
closely tied to any These are
other. both normal ways for a family
to be organized —that is, one would not expect pathological behav-
iors to arise in them. But the latter form is not likely to occur in
family groups, given the tendency of intimate relationships to de-
velop mild compatibility. In less tight-knit, hierarchical groups like
bureaucracies, the rules for clustering would be more likely to
apply, albeit with strong sanctions against too much peer closeness
and an emphasis on vertical loyalty to superiors. Davis also men-
tions Elizabeth Bott's study of English working-class families in a
London neighborhood (where every two families known by a third
family knew each other) as a different example of clustering. 9 The
resulting pattern of connectedness exactly satisfied Davis's theory
in that there were many little social islands with no special connec-
tions to each other.
The concept of clustering gives some backing for the suspicion
that balance theory, which predicts that all relationships in a given
set must be compatible, applies only in some cases. When we see
families in which the rules for balance theory strictly apply, this is
a signal to look more closely, for it is exactly in families with severe
pathology that these rules seem to be most strongly enforced.

Balance Theory and Family Pathology

Caplow and Haley agree that there is a tendency for relationships

to be compatible across any family network. If so, what is the
difference between a relationship structure that induces distress and
one that does not? One
could explain this by saying, as Haley does,
that in the distressed family all or most relationships seem to turn
into coalition relationships. Mother-child are close at father's ex-
pense. Parents can get together only in opposition to a child. The
nuclear family can achieve solidarity only by scapegoating a grand-

134
THE RULES OF CONGRUENCE FOR TRIADS
parent. Conversely, a conflict between two parties will be interfered
with by a third party who will side with one or the other of the
combatants or will deflect attention to himself or herself. Closeness
and apartness are both important aspects of family relationships,
but it seems that these processes in a distressed family are always
triadic, because no closeness, or any apartness either, is ever com-
fortable between two people.
Another factor is intensity. In families that produce symptomatic
members, there will be some triads in which the relationships are
experienced as if the family were a highly charged electromagnetic
field —
and the terms "positive" and "negative" are metaphorically
apt in this connection. Relationship values may shift abruptly, but,

whichever value appears, it will have the character of an invisible

force. By contrast, in a family that functions normally the align-
ments are much milder and far less binding. As a result, people seem
free to make alliances appropriate to necessary operations of the
moment, or stages in time. This is perhaps why clinicians so often
use the term "flexible" to identify a well-functioning family and
"rigid" to identify a dysfunctional one.
It should be pointed out, however, that inflexibility may not
always be attached to particular alignments but rather to the order
in which they appear. There may be a dominant coalition pattern
for one or more family triads, but these triads would take different
forms so that they would look different at different stages of the
total sequence. Because of this shifting quality, a family exhibiting
extreme pathology may seem chaotic and confused, and the rigidity
will be noticeable only in the redundancy of sequences across time.
A fourth factor is degree of connectedness of the family relation-
ship Sometimes a pressure for compatibility will spread to all
field.

corners of the network and not be limited to just one or two adjacent
triads. When Bowen noted this phenomenon he found it happening
most often in times of crisis, with more and more persons brought
into the triangulation process by Kalman Flo-
a sort of contagion.
menhaft and David Kaplan, describing families in which a member
has been brought to a psychiatric facility for hospitalization, com-
ment on the tendency for relatives to appear from nowhere and to
involve themselves in ways highly influential to the outcome of the
case. 10

135
 Foundations of Family Therapy
The pressure toward compatibility cannot be understood without
reference to the importance of hierarchy or levels in family struc-
ture. If there is a tendency toward rigid cross-level joinings in a
family, this will mean that there will be a pressure for "negative"
same-level ties. But this will create a conflict between family sub-
systems. If mother is ignored by father and turns to son for close-
ness, this may be useful for her, but it will create an improper
coalition which will weaken the executive dyad. This coalition may
not be useful for the child either, especially when he begins to move
away from home. Families have a special mandate to keep genera-
tion lines distinct, because each new generation has the task of
eventually disengaging itself from its predecessor.

However, intense cross-generation alignments are not the only

source of difficulty. Many kinds of family trouble arise from the
reverse situation, where there is extreme peer closeness and a huge
generation gap, one might almost say gulf. This can occur in family
structures where parents are extremely disconnected from their
children and the siblings form an overdose subgroup or gang, as in
Minuchin's "disengaged" families.
The point isno type of structure is itself bad or good. What
that
is good is a reasonable degree of independence (in Ashby's sense)

between parts. A family that works well seems to have built-in

separating devices that interfere with a tendency for relationships
to follow rigid coalition rules. The generation line, like other status
schisms, puts a distance between parts that might become too close.
So does the demarcation around a subgroup such as the spouses.
When parts can temporarily split off, so that mother can be close
to baby at feeding time without making father jealous, or husband
can visit his mother without wife feeling threatened, this means
that a certain separation of relationship areas is occurring. We then
have Caplow's optimal situation, where players can be opponents
or partners without disloyalty because they are meeting in different
activity systems.
To summarize, a family that produces members with psychotic or
psychoneurotic symptoms is also likely to provide us with the fol-
lowing clues: (1) a high degree of family connectedness; (2) covert
coalitions which cross generation lines; (3) closeness and distance
between family members determined by rules for congruence of

136
THE RULES OF CONGRUENCE FOR TRIADS
coalitions; (4) third parties interfering with or deflecting conflict or
closeness between pairs; and (5) relationships with a high intensity
factor.
It is becoming clear, however, that many of the axioms of balance
theory are better dealt with by coalition theory. This leads to the
questions: why has there not been an explicit joining of the two
areas of research, especially in Caplow's Two Against One. A close
reading indicates that at times Caplow does follow a line of reason-
ing based on balance theory, but, much of the research on coalitions
he describes comes from game theory. Hence there is an implicit war
of metaphors in his book. The "causation" implied by a person's
desire to win a game is not in the same universe as the "causation"
implied by the strain toward compatibility characteristic of "bal-
anced" family networks.
Scholars in the field are still struggling with many other questions
raised by balance theory. Howard Taylor's exhaustive study, Balance
in Small Groups,surveys these questions and attempts some an-
swers. 11
For our discussion we need to keep in mind primarily the
premise that close-knit relationships will tend toward compatibil-
ity. Clinical observation indicates that when this state of affairs is
carried to an extreme, so that no person in a social group can shift
his allegiances without affecting everybody else, the group is in
trouble. In family systems, in particular, there will be trouble if

there are intense, invariant cross-generation involvements.

Trade-offs for Balanced Forms

One observation that seems to be borne out by clinical studies is

that a polarizing process in a group tends to change to a scapegoat-

ing process, with obvious advantages for the larger unit. Family
with notions of the child as scapegoat or family
literature is replete
healer who "rescues" the family from the threat of parental splitting
or worse, but these formulations place the motivation, albeit uncon-
scious, within the individual. A more systemic view would see the
matter in terms of two formal patterns.

137
 Foundations of Family Therapy
Here it would be well to go back and reconsider Bateson's ideas
on the matter. In discussing symmetrical versus complementary
patterns, he suggested that each pattern might operate to check the
exponential tendencies of the other, and even proposed that the two
types of schismogenesis might be psychologically incompatible.
Putting this concept into an evolutionary framework, one could
speculate that those groups who survived were able to substitute
one schismogenic pattern for another, rather than accept the conse-
quences of either schismatic pattern getting out of hand.
The result, which seems sometimes like a sacrifice of the individ-
ual for the group or for some other person, is not always made
explicit. And yet it is this trade-off that family therapists often find
themselves confronting. Clinical examples abound of the process
whereby parental conflictseems to be averted by the symptomatic
behavior of a child, or sometimes a spouse. If the symptom carries
a benevolent label of "sick," this will create a concerned, united
group in place of a conflicted one. If it carries a label of "bad," this
member.
also unites the family, but in opposition to the problematic
There is always an element of "bad" in the "sick" label and vice
versa, but one definition usually prevails as the dominant one. At
the risk of suggesting a functional use for a symptom in a family,
there does seem to be a link between symptoms and family balance.
In some cases there seems to be an oscillation between family
conflict and a symptomatic display. One process takes over when
the other threatens to get out of hand, in a kind of alternating
mechanism that prevents the worse consequences of either state.
Again we see the applicability of Bateson's self-corrective circuits.
Haley's case of the schizophrenic girl, discussed in Chapter 6, is a

vivid example of parental conflict that was resolved when the

daughter returned to the hospital. The family had polarized, only
to become reunited with the once more in the position of odd
girl

man out. Cycles of hospitalization can be seen as an alternation

between threats of splitting to single deviant arrangements that
temporarily heal the split, back to splitting ad infinitum. This oscilla-
tionfrom symmetrical to complementary movements seems to de-
pend on a substratum of fixed coalition sequences. One goal of
therapy would be to attack the rigidity of coalition sequences, not
merely to interfere with specific behaviors like scapegoating. A

i3 8
THE RULES OF CONGRUENCE FOR TRIADS
value of seeing the formal structure that surrounds a symptom is

that this structure has to be seen in a circular rather than a linear

fashion, and thus may be interfered with in more than one way to
produce symptomatic relief.
In any case, it is clear that rigid coalition sequences such as we
have described in this chapter have positive as well as negative
value for the family or other entity; otherwise, with the dangers
they entail, they would not be tolerated. Haley suggests an answer,
saying that in most cases the subordinate party in a cross-generation
coalition seems to be involved in a split or conflict which is dividing
two superior parties. This split is not always between the two supe-
rior members of the triangle but can occur between two larger
groups with which each of these persons is identified. The triadic
mechanisms in mediating conflicts in social fields will be the focus
of the next chapter.

139
Chapter 8

Triads and the

Management of Conflict

The Natural Triad

Up to now we have been looking at triads that relate to distress

in a social system with the implication that these forms are them-
selves the source of some malignant tension. The position taken
here is that they are neither good nor bad. They are natural regula-
—
tory mechanisms which may or may not depending on the point
—
of view exact too high a price. They take a benign shape when a
group is functioning well, and what we have come to think of as a
pathological shape when it is not.
Let us start with the "natural triad" of Morris Freilich, which
Caplow mentions in Two Against One, as a good example of a benign
triangle. Freilich, an anthropologist with an interest in triads, no-
ticed a peculiar three-person arrangement that occurred over and
over again in kinship groups in many countries. What is interesting
about this triangle is that it has many of the same basic characteris-
tics as Haley's "perverse triangle" and Caplow's "improper coali-
tion." There is a close tie between a superior and a subordinate;

140
TRIADS AND THE MANAGEMENT OF CONFLICT
there is a hostility or distance between the other superior and the
subordinate; and there is a clear difference of attitudebetween the
two superiors.
Freilich formally describes the cast of characters of his triangle as
follows:
1. a High Status Authority (HSA)
2. a High Status Friend (HSF)
3. a Low Status Subordinate (LSS) 1
Freilich posits a positive or friendly relationship between HSF
and LSS and a negative or distant one between LSS and HSA.
Regardless of the kinship structure of a society, the person who
wields authority over a child, whether his father, his uncle, or his
grandfather, is the HSA, while another relative without that re-
sponsibility would play the part of the HSF —perhaps mother's
brother or sister, or a grandmother, or some such. Freilich observes
that similar sets of relationships are represented in societies like our
own by triads within occupations or institutions. In a prison it might
be "warden-chaplain-prisoner"; in a hospital, "psychiatrist-social
worker-patient"; in the army, "officer-chaplain-G.I."; in a univer-
sity, "authoritarian professor-friendly professor-student."
Freilich discusses themany-sided uses of this triadic form. For
one thing, he says, it is a kind of buffer, upholding the hierarchical
nature of a society that contains some who lead and some who
follow, while mitigating the strains between the levels. The HSF, he
says, is a power balancer within the group, mediating between the
severity of the demands of the group and the needs of the individ-
ual. In our society, a grandparent, not usually being the disci-
plinarian of the child, can act as the HSF. But if the grandfather, for
cultural or other reasons, is the main authority, the father or mother
can be a chum or pal.
In addition, the HSF is a tension-reducer. If the HSA creates
tension in the group, the HSF ameliorates it. Freilich turns to the
concept of "expressive" and "instrumental" leaders originated by
Talcott Parsons and Robert Bales, who wrote that in every group
there will be one person who is the "task specialist" and one who
is the "social-emotional specialist." Both Caplow and Freilich
equate these roles with two necessary but conflicting polarities.
There is the program of the individual and his own interest, and

141
 Foundations of Family Therapy
there is the program of the organization and its survival. Caplow
argues that, at times, it is healthy for a "revolutionary" coalition of
subordinates to override the sanctioned program of the organiza-
tion. At other times, especially when the group is in danger from
without, the authority structure of the organization takes prece-
dence.
We could say that these polarities represent Ashby's adaptive
mechanisms for the survival of any organism. The administrative
side of the continuum falls into Ashby's category of "constraint,"
the rules and regulations necessary for the maintenance of the sys-
tem. The individual side falls into the category of "variety," the
pool of idiosyncratic elements from which new solutions can be
drawn when the system is facing previously unknown circum-
stances. These polarities are the systole and diastole of the tension
between stability and change.
Parsons, applying his version of these polarities to the American
family, gives the mother the expressive role and the father the
instrumental one. Empirical research has found this not always to
be true, and latterly, of course, changes in parenting styles and sex
roles make the linking of any position with any gender a doubtful
enterprise.
Freilich points out that Bott's studies of family networks may
clarify at least some of these issues. In families embedded in close-
knit kin networks (usually working-class or ethnic groups) there is

a fairly rigid differentiation of labor with instrumental and expres-

sive roles clearly parceled out to father and mother respectively.
Nuclear families with sparse or loose networks have a less rigid
division of labor, and the parents' roles are more interchangeable.
But whatever the arrangement, the possibility of an alternation
between relatively "authoritarian" and "permissive" positions acts
as a system of checks and balances which may be integral to the
survival of any group.
Freilich tries to incorporate the principles of balance theory in
explaining how his "natural triad" works —but this poses some
problems. Heider and other balance theorists are correct, the HSA,
If

HSF, LSS triangle will have built-in difficulties. As soon as we posit

a positive relationship between HSF and LSS, this means that the
relationship between the two authorities will have to be negative,

142
TRIADS AND THE MANAGEMENT OF CONFLICT
or else the triad will have to change, either with all sides becoming
positive or with the HSA becoming friendlier with the LSS, and the
HSF becoming more distant.

To guard against the pressure for change in such directions, Frei-

7
lich suggests that ''formality systems' equivalent to Caplow's status
schisms between levels in a hierarchy, or Haley's generation line,

will act as role protectors. As an example of my own, a student in

trouble with the principal of his school might be befriended by the
guidance counselor. Balance theory predicts that the friendlier the
student and the counselor become, the more likely it is that the
counselor will side with the student and that relations between
counselor and principal will become tense, thus subverting the au-
thority structure of the school. Usually the pressure for solidarity
among school personnel will keep that tendency in check, so that
principal and counselor do not become divided over handling the
case, with possible worsening of the student's problems. This is an
illustration of how a "formality system" or status line operates.
As this example suggests, Freilich's formulation applies mainly to
a normal situation. It is appropriate for the leaders of a social system
to represent two contrasting positions, either of which may be
needed, depending on circumstances. The HSA (the principal in this
example) is important at times when the survival of the group is at

The HSF (the counselor) is important when exceptions must

stake.
be made on behalf of the individual, or when outside forces or
interior stresses point to the need for change. If the "formality
systems" are working right, they will counteract the strain toward
compatibility predicted by balance theory, even when the leaders
disagree.
But the "formality systems" do not always have sufficient weight
when conflicts between leaders or in-groups become too great, or
when there are pressures to form, or to intensify, cross-level or
cross-generation coalitions. Here we will see an economical twist by
which the natural triad, as described by Freilich, becomes the "per-
verse triangle" of Haley, and a third party employed which div-
is

erts the threat of splitting or a war. It is at this point that we must

return to the extraordinary insights of Alfred Stanton and Morris
Schwartz, mentioned in Chapter 6, in their description of what they
called "the special case": a configuration which seemed to act to

H3
 Foundations of Family Therapy
encapsulate a conflict and prevent the social fabric from being torn
apart.

The Problem of the Special Case

In The Mental Hospital Stanton and Schwartz explore the influence

of triadic forms in mitigating conflict. 2 Published in 1954, the book
was intended as a "social structural" study of a large institution. The
authors were, by and large, committed to an organization theory
framework that put the focus on elements such as chains of com-
mand, formal versus informal structures, flow charts for decision
making, lines of communication, arrangements for handling con-
flicts, questions of morale, and all the other problems that interest
the student of administration.
However, for family researchers, the study turned out to be an
unexpected corroboration of some of their hypotheses. Clinicians
working with families had long suspected that a hidden conflict
between parents might have something to do with the symptomatic
behavior of a child. Stanton and Schwartz's findings, which linked
outbreaks of pathological behavior on a hospital ward to unaired
disagreements among members of the staff, seemed to provide a
suggestive confirmation for these hunches.
Oddly enough, few if any other organization theorists picked this
—
idea up which is why Stanton and Schwartz's work is so unusual.
No study in small group or organization theory literature has ex-
plored in such detail the workings of the three-person form that
Haley calls the "perverse triangle," Caplow calls the "improper
coalition," and Stanton and Schwartz call the "problem of the
special case."
What is so special about their "special case"? The phrase itself

seems to lack color. It is almost as if the authors did not know that
what they had stumbled on deserved a more glorious signature.
Their book would have been, in fact, merely another organization
theory study had not they allowed themselves to be sidetracked
when an unidentifiable object swam into view.

144
TRIADS AND THE MANAGEMENT OF CONFLICT
This object appeared during what began as an inquiry into
first

breakdowns in staff morale. It seemed always to be around "the

problem of a special case" that both patient outbursts on the wards
and crises in the administrative life of the hospital clustered. This
around these events and start docu-
led the authors to put a circle
menting their circumstances. They noticed that the problem invari-
ably occurred when a patient was treated in such a way as to be
marked off as a favorite or pet of some authority. It was not the
same thing as "unique treatment." Sometimes a person had an ob-
jective reason for being treated differentially, as when he needed a
special diet for a physical condition. To be a true "special case," the
following features had to be present:
1. "special person" treatment
2. one who administers that treatment
3. one who protests it

4. an audience group whose norms are violated

Often the irritation around special treatment would be expressed
most loudly when an item in short supply was specially given, so
that others got less, or when the regimen ordered involved extra
work for staff. When it was seen as connected with a patient's
demand for particular recognition, or when it involved waiving of
a hospital rule that others had to follow, this would also cause
resentment.
The problem of the special case was also curious in that it seemed
to appear only under certain circumstances. For instance, if matters
on a ward were unusually disorganized or weren't going well, a staff
member might start to favor one patient at such a time. Others
would then begin to criticize the staff member, and feelings against
the patient would run high. A division would make itself felt along
a line of cleavage implicit in most organizations. In this hospital as
in other institutions, persons in authority tended to cluster around
two poles: some represented the official program, with all its rules
and regulations, and others took a more permissive attitude, arguing
that each case must be judged individually. Stanton and Schwartz
noted that at the center of each "special case" disturbance, there
would be two persons on the staff who represented those two poles.
The authors also noted that these persons were unable to deal
directly with each other over the issues that divided them and —
H5
 Foundations of Family Therapy
there was usually one issue which was not directly related
at least
to patient care —
but that they chose to do battle through a third
party, so to speak. Thus the one whose style came closest to the
permissive pole would side with the patient against the unfeeling
bureaucracy and would engineer all kinds of favors for him. The
other, believing in an across-the-board enforcement of rules, would
insist that the patient be treated like everybody else. The "protec-
tive" party might be and the "punitive" party a nurse,
a therapist
since these two opponents often represented the two opposing
groups within the institution. After a while the staff closest to the
main participants would begin to take sides and a polarization
would appear along authoritarian versus permissive lines. The pa-
tient would usually respond by becoming extremely agitated. Al-
though he might be seen as a successful manipulator, playing both
ends to his own advantage, his position was actually not enviable.
He was forced to respond to contradictory definitions of his situa-
tion by two superiors who each had power over his fate. This often
resulted in the patient becoming extremely disruptive and upsetting
his entire ward.
The authors go on to observe that the disruption might not only
spread downward among the patient population, but upward, in-
vading each echelon of the administration. As the controversy be-
tween the two staff members mounted, each of them would talk to
(or, as Bowen would say, "triangle in") more and more personnel.

If the major figures were influential in the hospital structure, the

entire staff might end up divided into two warring camps, with the
patient as the cause celebre. Subsidiary disagreements would cluster
about the main one, differentiated from it primarily by the fact that
their resolution would leave the core situation relatively unchanged.
The resolution of the main problem could happen in several ways.
One of the two combatants could go over the head of his opponent
to protest to a superior, usually by threatening to hand in his resig-
nation. Sometimes one of the two parties would begin to attract
more and more opprobrium to himself until he ended up a "minor-
ity of one." In this case, if his resignation was accepted, it was
tantamount to an extrusion, an example of Taylor's single deviant
arrangement. This would in essence free the patient, who had until
then been the major recipient of stress.

146
TRIADS AND THE MANAGEMENT OF CONFLICT
But more effective way to resolve the crisis, according to the
a
authors, was for the opposing pair to settle their differences directly,
face to face. The striking fact about the "special case" situation was
that the real disagreement, whatever it might be, was something the
principals were not always aware of. The authors began to believe
that this matter of covert disagreement was central to the special
case, if not actually its cause. They go so far as to state that:

All patients who were the center of attention for the ward for several days
or longer during the period of study were the subjects of such a covert
disagreement. The most striking finding was that pathologically excited
patients were quite regularly the subjects of secret, affectively important
staff disagreement; and, equally regularly, their excitement terminated,
usually abruptly, when the staff members were brought to discuss seri-
3
ously their points of disagreement with each other.

The authors present day-by-day account of one such

a detailed,
difficulty between a staff administrator and the patient's therapist,
which documents the above assertion. The state of the patient fluc-
tuated according to the intensity of the unexpressed disagreement
between these two staff members until they finally talked matters
out, whereupon the patient's excitability immediately subsided.
Stanton and Schwartz also discuss a related phenomenon: the
collective disturbance, which in one case involved a change that was
imposed by administration in the name of economy but that was
seen as detrimental to patient welfare by many of the staff. Person-
nel took sides for and against this change, and after it was finally
instituted a crisis in morale occurred, not only among staff but
among It was during this period that the
the patient population.
agitation of one patient ward spread contagiously, until the
on a
whole ward was in a turmoil. A therapist who was an outspoken
critic of administration policy had disputed the treatment of this

patient with a member of the administration. The authors make the

point that in this case the background of the patient's agitation was
not only the conflict between the therapist and the administrator
but the crisis involving a collective polarization of the entire institu-
tion. It did not matter that the disputed policies had been dictated
by sources quite outside the hospital itself.

H7
 Foundations of Family Therapy

The Mirror-Image Disagreement

In Chapter 2 we paid attention to one very important aspect of

social groups: the appearance of self-reinforcing repeating se-
quences. Haley, Caplow, and Freilich, in the works cited, tended to
describe their triangles in static terms, like pieces of architecture or
bits of Euclidian geometry. Yet, on a closer look, these triangles are
not static but embody the type of mutual-causal processes referred
to above. Stanton and Schwartz were the first researchers to link the
rigid triadic forms characteristic of social contexts where symptoms
appear to these peculiar redundancies. They became fascinated with
the dynamics of their ''special case" and invented a phrase to de-
scribe the polarization that invariably took place around it: the
"mirror-image disagreement." If, during the management of a "spe-
cial case," the two authorities took opposing views along authori-
tarian versus permissive lines, a deviation-amplifying process
would ensue that would increase their differences exponentially.
The more the protector protected a patient, the more the punisher
would punish him. But a central rule seemed to be that the parties
must remain polarized. If one party changed position, the other
would, apparently unconsciously, make a simultaneous about-face,
so that the structure of contraries in which the patient was locked
remained the same. By this process, an originally small conflict
between two authorities could become enormous. This could hap-
pen even in benign circumstances, with ludicrous results.
One could say that the patient, caught in such a vicious cycle, is

the unwitting agent who exponentially transforms a difference or

conflictbetween two staff members into a full-blown mirror-image
disagreement. But if one is to remain circular, one should not forget
the part the patient plays in maintaining and intensifying this dis-
agreement. In some strange way the intensity of the conflict gets
disssipated by going through the patient, almost as if he were a
lightning rod, and with the unconscious connivance of the apparent
"victim."
This amelioration of the conflict does not happen without a price,
however. In a brilliant insight, Stanton and Schwartz surmise that

148
TRIADS AND THE MANAGEMENT OF CONFLICT
this type of polarization becomes internalized and surfaces in the
patient in the form of a pathological condition known as "dissocia-

tion." This simply means that a person begins to perceive events

and persons around him in purely black and white terms. One
person will seem all bad, another all good. The patient will see
himself, too, as alternately "bad" or "good."
The authors observe, however, that the patient may not be per-
ceiving the facts as unrealistically as supposed. As they observe, two
important persons in the patient's life are in fact pulling him in
opposite directions. If these two persons are authorities with fateful
power over him, and one of them sees him in need of strict controls
and punishment and the other as deserving of favors and kindness,
a social context emerges in which the patient's "dissociated" fanta-
sies may be justified. This is not only true in regard to the perception
of the favoring person as good and the nonfavoring person as bad,
but may also pertain to conflicting notions about treatment or about
the self. The authors conclude by saying:

If our hypothesis is correct that the patient's dissociation is a reflection of,

and a mode of participation in, a social field which itself is seriously split,
it accounts for the sudden cessation of excitement following any resolution
4
of this split in the social field.

Stanton and Schwartz thus share with Haley the idea that certain
behaviors associated with schizophrenia are an appropriate reaction
to a real split between real others with whom the patient is inti-

mately engaged, if not deeply dependent on. This is very different

from assuming that there is a split in his personality, or a dissocia-
tive process taking place in his brain. In particular, the idea is pre-
sented that the person at the nexus of a pathological triad is some-
how deflecting splits in the wider social field —between kin groups
in the case of the family, or between professional groups in the case
of an institution. It is of interest that the line of fission on which
the split takes place is simply an intensification of the polarities
inherent in the dual executive structure which seems to be most
functional for any group. The dynamic model of Stanton and
Schwartz, which deals with an escalation of intensities along this
line of social cleavage, also explains the severity of the symptom
displayed by the person who is mediating the split, which may

149
 Foundations of Family Therapy
psychically, if not literally, tear him apart. It also offers the thought
that this person may indeed be an unwitting sacrifice, so that he,
rather than the body politic, absorbs the split.

Stanton and Schwartz concerned themselves with pathological

escalations in the context of a hospital ward. The logical next step
isto examine such escalations within a family that has a member
with a psychiatric symptom to see if the family acts in the same way
symptomatic behavior. Of all clinician-researchers,
as a matrix for
Salvador Minuchin has done the most interesting work in this area.

Conflict -Detouring Triads

A recent contribution to the literature connecting triads to the

management of conflict is Minuchin's Psychosomatic Families, a
ground-breaking study of children with psychosomatic disorders:
asthma, diabetes, and anorexia nervosa. 5 Minuchin and his cowork-
ers started from the hypothesis that children could be used to ob-
scure or deflect parental conflict. In analyzing relationship configu-
rations which coincided with symptoms in a child, Minuchin
7
formulated a typology of what he called "rigid triads/
These "rigid triads" are: "Triangulation," "Parent-Child Coali-
tion," "Detouring-Attacking," and "Detouring-Supportive" (see
Figure 8.1). "Triangulation" describes a situation where two par-
ents, in overt or covert conflict, are each attempting to enlist the
child's sympathy or support against the other. This form would
correspond to what I have called the "inadmissible triangle" of
balance theory, the triangle with two positive sides, connoting in-
tense conflict of loyalty. "Parent-Child" is a more open expression
of parental conflict, even though the family may come for treatment
with a child problem. One parent will side with the child against
the other parent, and at times it is difficult to determine whether the
child or the outsider spouse is in more difficulty. The intense close-
ness of the child to the preferred parent can result in symp-
tomatology, however, especially when the natural process of grow-
ing up begins to put stress on the parent-child stasis.

150
TRIADS AND THE MANAGEMENT OF CONFLICT

Triangulation • Parent-Child Coalition

Detouring- Attacking Detouring -Supportive

\ /
\ /
\/
•
C

Figure 8.1
Minuchin: "Rigid" Triads

There are two types of "Detouring" triads. In a "Detouring-

Attacking" triad the parents are most often perceived by the clini-

cian as scapegoating the child. The behavior the child shows is

disruptive or "bad," and the parents band together to control him,

even though one parent is often apt to disagree with the other
parent over how to handle him and both may handle him inconsist-
ently. Most behavior disorders in children fall into this category. In
a "Detouring-Supportive" triad the parents are able to mask their
differences by focusing on a child who is defined as "sick," and for
whom the parents show an enormous, overprotective concern. This
brings them and is a frequent feature of families in
close together
which tension is expressed through psychosomatic disorders. All
these triads, or permutations of them, can be found in families with
psychosomatic children, but they are prevalent in families where
children have other problems as well.

IS*
 Foundations of Family Therapy

The Transfer of Stress

Using this triadic framework, Minuchin's team came up with a

research idea that ties parent-child interactions to symptom produc-
tion in the child. The particular test devised was originally directed
toward a group of diabetic children. It had been noted that the
presence of free fatty acids (FFA) in the blood could be an indicator
of emotional arousal, and a concentration of these substances had
long been linked to the onset of diabetic acidosis. A measurement
of the plasma elevation of this substance might be the clue that
would test out what could lead to the physiological changes known
to be associated with diabetic attacks.
There were forty-five research families in all: ten with children
with intractable asthma; nine with families with superlabile diabe-
tes; and eleven with children with severe anorexia. The control

group consisted of seven families with diabetic children whose ill-

ness was well controlled, and eight families with behavior-problem

diabetics whose condition was not life-threatening. The purpose of
the experiment was to provide evidence for Minuchin's hypothesis
that the symptom of the child is intimately connected with the
presence or absence of stresses between his parents. The larger
research purpose of the study was to prove that a therapy that
concentrated on changing the structure of relationships which con-
strained the child would also alleviate the symptom.
The question that especially interests us here, however, is the
smaller one dramatized by the structured interview. This interview
was designed so that blood samples of parents and symptomatic
child could be taken at regular intervals. After a baseline FFA level

was member, the child was placed be-

established for each family
hind a one-way screen to watch while an interviewer induced an
argument between the parents. After half an hour the child was
asked to come into the room, and parents and child were asked to
work together on a solution of the argument.
The researchers found that the symptomatic children in the ex-
perimental group showed a far higher rise in elevations of plasma

152
TRIADS AND THE MANAGEMENT OF CONFLICT
FFA than children in the control group. In addition, their FFA level
went higher and took longer to return to baseline after the interview
was over than anyone else in the family, while the FFA level of the
"higher parent" began to go down shortly and sharply after the
symptomatic child entered the room. Examination of family in-
teraction data (all sessions were videotaped) confirmed the hypoth-
esis that in some way the parents "passed on" their emotional

arousal to the child, as one would transfer a heavy weight. In the

experimental families almost all conversation involved the child,
while in the control families much more interaction took place
between the parents.
Although there is no known direct link between levels of free
fatty acids in the blood and the symptoms connected with asthma
and anorexia, the children with these disorders behaved much the
same in the interview as the diabetics. For instance, the FFA profiles
of all the anorectic children, averaged out, rose in a high, steep curve
as the parents argued; the level of the "normal" diabetics stayed
below baseline level; while that of the "behavioral" diabetics rose
slightly above it. This experiment, limited though it may be in
population, is the first to my knowledge that has ever established
a direct tie between an interactional sequence involving a sympto-
matic child and chemical changes associated with his illness.

Therapeutic interventions based on Minuchin's hypothesis that

a child'ssymptom can be linked with parental conflict proved
unusually successful. The goal of therapy was a structural one: to
disengage the child from his position between the parents and to
help the parents deal with their problems more directly. Follow-up
studies of the group of fifty families of anorectic children over an
eight-year period showed
86 percent of these children had
that
recovered. In addition, they were doing well on other indices of
normal functioning. Most of the group reached normal weight
within less than a year of treatment, some in the first weeks after
therapy started. To date, there have been no deaths, which contrasts
with the standard fatality rate for anorectics of 12 percent. In addi-
tion, the return to social and personal functioning contrasts with
the average 40-60 percent success rate of individual therapy treat-
ment programs, in which the anorectic may regain normal

153
 Foundations of Family Therapy
weight but continues to show symptoms and to function poorly.
At this point it is useful to reconnect with the idea, suggested by
Stanton and Schwartz, that a child's symptom may be associated
with a mirror-image disagreement between the parents (overt or
covert) which can be enormously stressful if it involves the child.
In the four cases illustrating therapy with anorectics in MinuchhYs
book, this form repeats itself over and over in the beginning inter-
views. It is almost as if there were a hidden program in some famlies,
no matter what social group they come from. Otherwise, how do all
these families, from such diverse backgrounds, come up with such
a similar reaction to threats to family stability? Over and over in the
initial lunch session (a standard feature of MinuchhYs treatment

program for anorectics) there is one authoritative parent who tries

to force the child to eat and another who pulls back, gives way,
attempts to calm the child and soften the other parent. The child is
caught in the classic "ballot box" situation: If he eats, he will be
voting for one parent, and if he does not, he will be voting for the
other. In addition, the action escalates so that he is being pulled
harder and harder in two directions.
Nevertheless, it is important to realize that we are not dealing
here with a simple triangle or cycle, but a complex force field with
surprisingly similar characteristics from family to family. It is al-

most as if there were sets of instructions in families — perhaps in all

social groups —having to do with the ordering of behaviors in the

face of change.
We have seen that the most obvious single characteristic of fami-
lies with "disturbed" members is their apparent lawlessness, most
strikingly conveyed by the lack of boundaries or appropriate status
lines. The family is governed —
if that is the word by a powerful —
politics of secret coalitions across the generations. What is so in-
triguing about the families in which this subterranean structure
prevails is that one also finds processes equivalent to Bateson's

corrective circuits, with symmetrical escalations tending to polarize

the family, blocked by complementary sequences of a counteracting
nature; threats of civil war or reciprocal violence, blocked by symp-
tomatic displays or, alternatively, the emergence of the kind of
solidarity which only a common enemy or an outside catastrophe
can create. Of all these forms, the most common check to violence

154
TRIADS AND THE MANAGEMENT OF CONFLICT
or splittingis, as we have noted, the "single deviant arrangement,"

in which the group attains unity at the expense of a symptomatic

member. At the same time, the family fails to evolve toward an
organization more appropriate to its stage. How to intervene in
these schismogenic operations, and in the fateful spirals that charac-
terize them, will be the material for the next few chapters.

155
Chapter 9

The Simple Bind and

Discontinuous Change

Evolutionary Feedback

At this mid-point in our discussion we shall shift from a struc-

turaltaxonomy to a process taxonomy, in line with Bateson's model
for mapping phenomena at increasing levels of complexity: the
"zigzag ladder of dialectic between form and process/' 1 It has been
useful to note the early impressionistic descriptions of the flow of
interaction in families that produce disturbed persons, and then to
try to codify structures from the information provided by this flow.
In families with psychotic members, in particular, generalizations
about deviant and normative structures are ways of imposing an
external order upon flow. One can snatch out of the booming,
buzzing confusion of family interaction certain clear redundancies
and 'There it is, and there, and there again!"
say,
The emphasis on structure, however, does not reflect the power
of living systems to reorganize in sudden and transcendent ways. A
recent paper by Paul Dell and Harold Goolishian examines the
concept of "evolutionary feedback," a term developed by the physi-

i$6
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
cist Prigogine to describe a "basic, nonequilibrium ordering princi-
ple that governs the forming and unfolding of systems at all lev-
els/' One can turn to Bateson's Mind and Nature and find a similar
2

description in his comparison between epigenesis and evolution:

In contrast with epigenesis and tautology, which constitute the worlds of

replication, there is art, learning and evolu-
the whole realm of creativity,
tion, in which the ongoing processes of change feed on the random. The
essence of epigenesis is predictable repetition; the essence of learning and
evolution is exploration and change. 3

7
Prigogine's concept of "order through fluctuation/ as Dell de-
scribes it, emphasizes not stability and homeostatis but the idea of
discontinuous change:

... at any point in time, the system functions in a particular way with
way of functioning has a
fluctuations around that point. This particular
range of stability within which fluctuations are damped down and the
system remains more or less unchanged. Should a fluctuation become
amplified, however, it may exceed the existing range of stability and lead
the entire system into a new dynamic range of functioning. An autocata-
lytic step or surge into positive feed back is n^r\^r\ fn nhf^in f uch instabil-
ityj

Dell's point is that the cybernetic analogy based on a mechanical

model of closed-system feedback is limited and inaccurate. There is

a different cybernetics of living systems which was incompletely

explained by the negative feedback view. This point is dramatized
by the sudden, step-wise leaps to new integrations characteristic of
such systems, which are not only unpredictable but irreversible.
The conceptual emphasis is not on processes that tend toward equi-
librium, but rather on self-organizing processes that reach toward
new evolutionary stages.
What makes this argument so crucial is that families that come
for treatment with distress in one or more members seem to be
having difficulty —
with evolving they are or seem nonevolved,
stuck in an outmoded stage. Perhaps it is this being stuck that made
the early version of the homeostatic model so convincing to thera-
pists working with troubled families. The emphasis in those fami-
lies is on maintaining equilibrium, too much so. For those families

that become more and more like a homeostatically controlled piece

^57
 Foundations of Family Therapy
of machinery, the task of therapy should be to make available the
power inherent in all living systems to transcend the stuckness and
move to a different stage.
To recast our cybernetic analogy within an evolutionary frame-
work is certainly in itself an evolutionary step forward in family
theory and theory of change. For one thing, it fits the process we
are trying to describe far better than the static model of error-
activated feedback mechanisms does. For another, it affords a far
more satisfying rationale for the success of some of the so-called
paradoxical approaches to therapy that produce rapid shifts in fami-
lies or individuals. These shifts can take place with incredible sud-
denness, and indeed seem to be self-generated. To go further into
this subject, let us turn to the ideas of another physicist who has
written about discontinuous change, John Piatt.

Hierarchical Growth

One property that families share with other complex systems is

that they do not change in a smooth, unbroken line but in discon-

tinuous leaps. Piatt, in an imaginative paper, speaks of a process
physics in which the emphasis is not on static structure but on what
he calls a "flow hierarchy": forms that maintain a steady state while
matter, energy, and information continually flow through them. 5 A
bit of thought will convince the reader that families, too, are like
waterfalls or cascades, with the many-tiered pattern of the genera-
tions persisting as an overall structure, even though individuals pass
through it as they are born, grow old, and die.

Piatt argues that many natural systems are of this type, and that
change in such systems occurs in a startling and sudden way. He
cites falling in love, acts of creation, conversions, evolutionary
leaps, reformations, and revolutions as examples, and says that
when a system is conflicted or dysfunctional this may not necessar-
ily portend disaster but may indicate that pressure toward a new
and more complex integration is mounting.
Piatt makes a useful distinction between three kinds of change,

i58
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
each of which depends on the way the entity in question is orga-
nized. If the entity is externally designed (a watch, for example),
then any change will have to be imposed by an outside agent, the
watchmaker, who may take apart and reassemble the watch. If it is
internally designed (a plant that contains a genetic blueprint), then
only mutations of the gene pattern can produce a change.
A third model for change is found in living systems that, unlike
plants, have the capacity to evolve to new and unpredictable levels
of organization. In such systems, change is not prefixed but takes
the form of a transformation, a sudden appearance of more func-
tionally organized patterns that did not exist before. Piatt calls this
type of change "time emergence/' One might think of a kaleido-
scope, which keeps the same geometric pattern as the tube is turned
until all at once a small particle shifts in response to gravity and the
whole pattern changes to an entirely new one. The most interesting
feature of a kaleidoscope is that one can never return to an earlier
pattern, and this is consonant with the way systems that have what
Ashby calls "bimodal feedback mechanisms" operate. 6 Such sys-
tems will remain stable as long as the environment around them
does not change, or as long as internal elements within do not
change; but if either of these events occurs, the system will break
down or else respond by shifting to a new "setting" that will meet
the demands of the new field. The change in the setting creates a
discontinuity because the range of behaviors, the "grammar" for
allowable activities, has changed. Thus a set of completely different
patterns, options, and possibilities emerges. The new organization
is usually more complex than the previous one, but it too is rule-

governed and will not change again until new pressures from the
field enforce a new leap. Not to sound too purposive, it must be

emphasized that the source of newness most often comes from some
random element. As Bateson says: "The ongoing processes of
change feed on the random." 7
The natural history of a leap or transformation is usually this:
First, the patterns that have kept the system in a steady state relative
to its environment begin to work badly. New conditions arise for
which these patterns were not designed. Ad hoc solutions are tried
and sometimes work, but usually have to be abandoned. Irritation
grows over small but persisting difficulties. The accumulation of

159
 Foundations of Family Therapy
dissonance eventually forces the entire system over an edge, into a
tendency brings on ever-intensify-
state of crisis, as the stabilizing
ing corrective sweeps that get out of control. The end point of what
cybernetic engineers call a runaway is that the system breaks down,
or creates a new way to monitor the same homeostasis, or spontane-
ously leaps to an integration that will deal better with the changed
field.

Families are notable examples of entities that change through

leaps. The individuals making up a family are growing (at least

partly) according to an internal biological design, but the larger

groupings within the family, the subsystems and the generations,
must endure major shifts in relation to each other. The task of the
family is to produce and train new sets of humans to be indepen-

dent, form new families, and repeat the process, as the old set loses
power, declines, and dies. Family life is a multigenerational chang-
ing of the guard. And although this process is at times a smooth one,
like the transitions of political parties in a democracy, it is more
often fraught with danger and disruption. Most families do not leap
to new integrations with ease, and the "transformations" Piatt re-
fers to are by no means self-assured. This brings us to the research
of sociologists and clinicians studying the family life cycle.

Expectable Life Stage Crises

The family was discovered by a circuitous route. Of

life cycle
major importance was the work of Erik Erikson during the 1940s
and 1950s, whose depiction of individual life stages, and of the
interplay between these stages and the shaping processes of social
institutions, challenged the narrow focus of intrapsychic theories of
8
development. After World War II, clinicians studying individuals'
responses to stress began to question the notion that some people
had better coping patterns or better "ego strengths" than others.
One of the first pioneers in this area, Erich Lindemann, noticed that
the difference between a normal and an abnormal grief reaction had
to do with the overall makeup of the family network of the be-

160
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
reaved one, not with his coping mechanisms as shown by previous
attempts to handle stress. In his classic study of survivors and rela-
tives of victims of the Cocoanut Grove fire, Lindemann notes that

Not infrequently the person who died represented a key person in a social
system; his death was followed by disintegration of this social system and
by a profound alteration of the living and social conditions for the be-
reaved. 9

The intensity of a grief reaction did not have to be tied in with a

previous neurotic history, but was linked to the type of loss for the
person involved.
Researchers who came and Lindemann began to see
after Erikson
that in terms of the family life cycle, one man's adolescence might
coincide with his mother's change of life, and possibly his grand-
mother's demise. Rhona Rapaport has singled out as stress events
"the critical transition points in the normal, expectable development of
the family life cycle: getting married, birth of the first child, children
going to school, death of a spouse, or children leaving home." 10
Reuben Hill, at the University of Minnesota, has found differ-
ences in the way families respond to these stages. He speculates
about the factors that predispose a family to treat a normal life stage
11
stress as a crisis. For instance, a child going to kindergarten might
in some families produce a crisis, as the retirement of the head of
the household would in others. A growing number of researchers,
such as Michael Solomon, have extended these observations to
show that psychiatric and medical symptoms tend to cluster mag-
12
netically about those times.
Building on these observations, Thomas Eliot contributes the idea
that a crisis in a family often follows a revision of membership. He
offers two unusual terms: crisis of dismemberment, meaning when
a family loses someone; and crisis of accession, meaning an addition
to the group. 13 Lindemann's work, as the passage just quoted makes
clear, falls into the category of dismemberment, or, to use a less

grisly word, separation. A crisis of accession is exemplified in stud-

14
ies like that of E. E. LeMasters on "Parenthood as Crisis.
Not only a loss, it then appeared, but the acquisition of new
family members could trigger an upset. In 1967, a now classic study
by T. H. Holmes and R. H. Rahe, who compiled a "Social Readjust-

161
 Foundations of Family Therapy
ment Rating Scale," indicated that there was no correlation between
the negative perception of an event and the degree of stress that was
15
attached to it. Out of a list of forty- three life stress events, rated
by 394 subjects in terms of intensity and length of time necessary
to accommodate to them, ten out of the top fourteen involved
gaining or losing a family member. It is fascinating to realize that
events with presumably positive meanings, like "marital reconcilia-
tion,"ranked as more stressful on the scale than some with negative
connotations, like "difficulties with sex." 16
If it is true that crises tend to erupt at times when a family is faced
with normal revision of membership, then it is logical to assume
a
that these crises would be most intense in families that have diffi-
culty reorganizing —letting members go or taking in new ones. It is

only one step more to the surmise of family researchers like Haley
that symptomatic behaviors tend to surface at points in the family
life cycle when the process of disengagement of one generation
from another is prevented or held up. 17 For instance, members of a
family in which a child is one of the possible factors that is mediat-
ing a parental conflict may resist or even block the child's departure.
A symptom seems to be a compromise between staying and leaving;
the child becomes incapacitated to a greater or lesser degree and
never really leaves home, or may leave but find it hard to negotiate
the transition to marriage and fall back, or else a child of the new
marriage may have to serve as mediator in turn. One can often see
the truth of the biblical statement: "The fathers have eaten a sour
grape, and the children's teeth are set on edge." Sometimes one frail,
psychotic child seems to be holding an entire kin network on his
shoulders, like the key person in a family high-wire act, displaying
incredible strength and impeccable sense of balance.
We may now justifiably ask what the arrangement is that some-
how prevents people in a family from making the leap to a new
integration? The answer is suggested by the concept of another kind
of shift, which occurs when an entity is about to exceed its parame-
ters or break. For this we shall have to turn to Ashby and his idea
of step-mechanisms.

162
THE SIMPLE BIND AND DISCONTINUOUS CHANGE

The Concept of Step-Mechanisms

In Design for a Brain, Ashby described four types of movement by

which natural forms or substances pass from one state to another. 18

A "full-function" moves in a progressive fashion without a finite

interval of constancy between states, like a barometer. A "step-

function" has intervals of constancy separated by discontinuous
jumps, like a set of stairs. A "part-function" is like a step-function

except that from one state to another the line is progressive, rather
than instantaneous. A "null-function" indicates simply an absence
of movement or change.
Ashby comments that many step-functions occur in the natural
world. His examples include the tendency of an elastic band to break
when the proportion of pull versus length reaches a certain point, or
blow when the circuit is loaded beyond a certain number
of a fuse to
of amperes. Looking at more complex entities, such as machines,
Ashby notices that some of their variables may exhibit a sudden shift
in character whenever they reach a certain value that he calls a
"critical state." In fact, he says, it is common for systems to show
step-function changes whenever their variables are driven too far
from some usual value. He speculates that it would be useful for a
system to have at least one such element. For instance, in the electri-
cal wiring of a house, if there is no circuit breaker, the whole system
will break down and have to be replaced. But if there is a circuit
breaker, only a fuse will blow, and when that is replaced (assuming
the overload has been corrected), the system will still be functioning.
Ashby calls this type of arrangement a step-mechanism.
One difficulty with Ashby's ideas is that he was not really con-
cerned with living systems at the group level and above but was
trying to devise a cybernetic model that would account for the
evolution and structure of the brain.Thus most of his examples are
drawn from the worlds of biology, chemistry, and physics, and one
has to pull his ideas out of context to make them apply to social
systems. But without some notion similar to the step-mechanism,
the sudden shifts in behavior one often sees in families with symp-
tomatic members could never be explained.

16)
 Foundations of Family Therapy
In the family, one essential variable is the relationship between
members of the executive dyad, who are usually the parents. This
relationship probably has particular arrangements regarding the
management of such dimensions as closeness/distance and balance
of power, which limit the behaviors allowed in this dyad. Let us
hypothesize that one of these sets of limits is constantly being
overpassed. With an even-Stephen or symmetrical couple, a slight
advantage accruing to one person may provoke an escalation that,
if not blocked, could end in violence or divorce. With a one-
up/one-down complementary couple, too much inequality may
or
produce depression in the "low" spouse and concomitant anxiety in
the "high" one. Whatever the nature of the plateau (and it is usually
not a pure example of either of these models), there will be a
"critical state" that represents some value beyond which the couple

as a system may not go and remain intact.

At this point different things can happen. A couple may have
techniques for handling this threat, like a cooling-off period for an
angry symmetrical couple or a "good fight" for a distant comple-
mentary one. Another technique would be for one of the spouses
to develop a severe or chronic symptom, which will also prevent a
split, though at a cost. However, it often happens that a third party,

very likely a child, becomes drawn into the conflict. Once this
happens, the child's discomfort grows while parental tensions
lessen. Some minimal cue indicating parental conflict may incite
anxiety in the child, who reacts with irritating behavior. At this
point one of the parents may start to attack him or her, while the
other moves to the child's defense. Caught in the tightening spiral,
the child may respond with a physical or emotional symptom. This
will cause the parents to stop their covert struggle and unite. A very
real issue joins the couple, since the child's well-being is at stake.
Their getting together, especially if it is accompanied by supportive
behavior, allows the child's anxiety to diminish.
In this example one could say that warning signals are at work
whenever a feedback chain reaches a critical state in a set of rela-
tionships. These signals forestall events that might endanger rela-

tionships important to the group. For instance, the child's symptom

is a warning signal that diverts the parents from having a fight.
But what if the child's discomfort proceeds to a level that is

164
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
unacceptable, and a positive feedback chain develops that cannot be
countered by the usual family responses? Here we move up to the

next where the interface is not between the child and his
level,

parents but between the family and the wider society. Ashby
writes:

A common, though despised, property of every machine is that it may

"break." when a machine "breaks," the representative point
... In general,

has met some critical state, and the corresponding step-function has
changed value. ... As is well known, almost any machine or physical
system will break if its variables are driven far enough from their usual
value. 19

It is what is generally called a nervous breakdown

possible that
is what Ashby is talking about. In a family,
similar in function to
the individual's "breakdown" operates as a step-mechanism signal-
ing the failure of the family's stabilizing arrangements and often
activating interventions from the larger system, the community.
Here is where helpers in various guises come in and an attempt is
made to repair the broken element, the person.
But to go back to the image of the electric circuit, as long as it

continues to be overloaded, it will not do any good to fix or replace

the fuse. Sometimes the problem is temporary; the overload has
been due to a sudden plugging in of an extra appliance (a mother-
in-law visiting, for instance), and once that is taken away, the
system will return to normal. But often the change is permanent.
Somebody has died, or there is an irreversible shift in family cir-
cumstances, or a family member has reached a new maturation
level. Then the family must make a shift in its overall organization
to meet the new demands. Otherwise the person's symptomatic
behavior may continue, or another problematic behavior may re-
place it. In a family with a troubled member, we may be dealing
with a fear on the part of family members that the leap to the next
stage may impair some important family member or subsystem, or
threaten the survival of the family itself.

It follows that symptomatic displays can be thought of negatively

as aborted transformations or positively as negotiations around the
possibility of change. In his Prison Notebooks, the Marxist philosopher
Antonio Gramsci writes: "The crisis consists precisely in the fact

165
 Foundations of Family Therapy
that the old is dying and the new cannot be born; in this interreg-
num a great variety ofmorbid symptoms appears/' 20 A sympto-
matic redundancy is an arrangement that usually springs up to
handle this interregnum between the old and the new. It represents
a compromise between p ressures for and against change. The symp-
to m is only the most visible aspect of a connected flow of behaviors
and acts as a primary irritant that both monitors the options for
c hange, lest too rapid movement imperil someone in the family, and

also keeps the ne cessity for change constantly alive What results .

then is a turmoil of behaviors that spiral rather than cycle around

the possibility of a leap. Sometimes the leap is taken simply because
of some accidental shift brought about by the spiral, which is al-
ways moving forward in time. Even a very narrow, bunched-up
spiral that chronically circles around some central point is still al-
ways shifting and is never without some potential for change. Reiss,
in his paper on family paradigms, describes a nice example of this
change-by-accident. A teenager in a family fell seriously ill while
the family was on a camping trip. The father, who had been seen
as nice but boyishly undependable, took over during the emer-
gency, carving out a new and lastingly authoritative role for him-
21
self.

The next question is how to help the family to make a leap up,

rather than continue in the chronic spiral, and to achieve a transfor-

mation to a new stage that will obviate the presence of symptoms
or distress.

Paradoxical Injunctions and the "Sweat Box"

Piatt, as we saw, was stressing the positive —even extraordinary

—capacity of living systems to achieve transformations that go
beyond what could previously have been predicted or achieved,
thus not only "saving the day" but pointing the way toward a new
one. Ashby was looking at a different kind of shift, perhaps equally
extraordinary: the ability of one element of a system to "break" if

166
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
too much pressure for change has been introduced. In a family or
other group, the shift to a symptomatic configuration saves the day,
but it does not always point the way to a new one. It may be seen
as a nonevolution, or failed leap, since it not only keeps the family
from making a new integration but seems to happen at the expense
of one family member, who has often sentimentally been thought
of as the "scapegoat." On the other hand, it may be valued as
the only persistent pressure toward change that is going on in the
family.
The question for therapy then becomes: How does one disrupt an
arrangement that in some ways promotes family stability (mor-
phostasis) and instead help the family achieve a transformation that
will represent a more complex integration (morphogenesis)? Here a
discussion of what Richard Rabkin calls "saltology" (from the Latin
saltus, "to leap"), and that might more prosaically be called "leap

theory," is in order. Also important in this connection is some

extremely good thinking Rabkin has done in relating transforma-
tions or leaps to the appearance of that communicational oddity the
"paradoxical injunction."
In a paper called "A Critique of the Clinical Use of the Double
Bind," Rabkin has presented a refreshing examination of the origi-
nal double-bind concept. 22 This paper reclassifies most of the exam-
ples used by clinician-researchers to illustrate double binds into
masked hostility, sarcasm, strategic deceit, and ordinary "damned-
if-you-do, damned-if-you-don't" dilemmas.
A case can be made one of these dilemmas,
for equating at least
the paradoxical injunction, with the double bind. A paradoxical
injunction is a statement that intrinsically contradicts itself unless
teased apart into a "report" level and a "how this report ismeant"
level, with the second level inclusive of the first. An example, from
Sluzki et al/s article on transactional disqualification, is the follow-
ing exchange:

son (to mother): You treat me like a child.

mother: But you are my child. 23

The mother's answer on the "report" level is absolutely true, but

in the context of the exchange the mother is doing some devious

167
 Foundations of Family Therapy
reframing: the son wrong; the son is unfairly criticizing her; the
is

son should furthermore accept these distortions of his original mes-

sage because he is her child. This would be a fair example of what the
researchers in Palo Alto would have thought of as a double bind,
whether Sluzki does or not. And the double bind was, as we know,
associated with manifestations of irrational behaviors such as
schizophrenia.
Nevertheless, there is Rabkin points
a terrible simplification here.
out that the paradoxical injunction is a form of communication that
all parents and all children (all superiors and all subordinates, for

that matter) encounter at some time in their lives without going

insane. Of course they may get upset —but ideally they shouldn't,
Rabkin argues, because the paradoxical injunction is the best our
poor language can do to suggest that a systems change is required.
Rabkin takes an example clinicians have used to equate a para-
doxical injunction with a double bind. The parent says to the child,
at a point when the child is about to pass into the gray area of
adolescence: "I insist you go to school because you enjoy the beau-
ties of learning. " (The Bateson group in Palo Alto used a similar
example, a New Yorker cartoon in which an employer is telling an
uncomfortable-looking employee, "But, Jones, I don't want you to
agree withme because I say so, but because you see it my way.")
Rabkin then quotes Arthur Koestler on the process of creation.
Before a creative leap can occur, says Koestler, all previous path-
ways must be blocked. It is only from the accumulated intensity of
the stress that pressure to take the leap will occur. 24
Seen in this light, the paradoxical injunction appears to be the
communicational form most likely to create sufficient pressure for
change. The paradoxical injunction of parent to adolescent child
says, in effect, "I want you to be independent, but I want you to

want that independently of my wanting that." What might be

called, for want of a better term, a "simple bind" is set up. The
receiver is directed to remain simultaneously in a symmetrical and
a complementary relationship with the communicant. This being
impossible, a leap must be taken to what Rabkin calls an "achieve-
ment," his word for Piatt's transformation or new integration.
The impossible situations that the Zen Master sets up for the
student are understandable in this The Master says to the
light.

168
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
student: "Here is a stick. If you say it is real, I will hit you with it.
Ifyou say it is not real, I you with it.
will hit If you say nothing,
I you with it." One response
will hit is for the student simply to take
away the stick. The whole point is for the student to become
"equal" to the Master, but this cannot be done by an order from the
Master, or from within the Master-student relationship at all. The
student must somehow get the idea "on his own" that this is the
course he must take. In line with this thinking, one should reserve
the term "paradoxical injunction" or "simple bind" for the confus-
ing directive that often appears as a harbinger of a leap to a new
stage, and the term "double bind" for communication sequences
that block this leap, or imply unthinkable consequences should it
occur.
The introduction of this concept of the simple bind solves many
issues thathave perplexed researchers and clinicians for years. For
one thing, there is no longer the vexing question: If paradoxical
communication is operating in art, fantasy, play, and most creative
activity, how do we distinguish between forms of paradoxical com-
munication that are associated with schizophrenic communication
and forms that are associated with the achievements of the artist or
the prophet? For another, we have a way to explain the idea of the
therapeutic double bind or counterparadox, which has been likened
to homeopathic medicine: The cure resembles the disease. A thera-
peutic double bind might be rephrased as a reinstatement of the
conditions of a simple bind, although this time within a different
context: the relationship between the therapist and the client or
family. The bind is reimposed, the period of confusion is gone
through, the family or client takes the requisite leap, and the new
integration is then rewarded, rather than invalidated or dismissed,
or is its own reward.
An example of this process is described by Bateson in an essay
on "learning to learn." 25
Bateson had become interested in por-
poises who were trained to show "operant conditioning" to the
public by exhibiting special behaviors, hearing a whistle, and then
receiving a fish. The porpoises possessed a considerable repertoire
of these behaviors. Bateson realized that these animals, since they
did not produce the same behavior every time, must have "learned
to learn" how to produce a piece of conspicuous behavior. He asked

169
 Foundations of Family Therapy
to watch the process by which was taught to do this, and
a porpoise
in fact created an experimental situation in which to conduct his
observations.
was to reward the porpoise for a piece of conspic-
First the trainer

uous behavior. The animal quickly learned that raising her head
would produce a fish, and several repetitions reinforced this impres-
sion. However, the next time the porpoise came in and repeated the
behavior, no fish. The trainer would wait for the animal to produce
a new piece of conspicuous behavior —
perhaps an annoyed tail flap
— and then would reward that. The behavior was reinforced three
times in the session in which it occurred but not in the next. Re-
wards occurred only when the porpoise again produced a piece of
unusual behavior.
This process was evidently so disturbing to both man and beast
that the trainer kept breaking the rules to reinforce the creature at
times that were not appropriate. The porpoise, in turn, began to act
more and more agitated as attempts to gain a previously reinforced
reward would prove futile, exhibiting behaviors that, in a human,
might be called psychotic.
Before the fifteenth session, however, a remarkable event took
place. The porpoise rushed about the tank, appearing intensely
excited. When she came on for her performance, she put on an
which had never been
elaborate display of eight behaviors, three of
noticed in this species before. Bateson makes the point that the
disruption of habitual patterns of stimuli and response can be in-
tensely upsetting if this disruption constantly puts the creature in
the wrong in the context of an important relationship. But he adds
that if the disruption and pain do not cause the animal to break

down, the experience may produce a creative leap, a fact noted also
by Wynne in his essay "On the Anguich and Creative Passions of
Not Escaping the Double Bind." 26
This example reinforces the notion that a prerequisite for crea-
tive leaps in complex systems is a period of confusion accom-
panied by self-contradictory messages, inconsistencies, and, above
all, paradoxical injunctions: I command you to be independent; I

want you to spontaneously love me; I order you to be the domi-

nant one. These messages, with their threatening implications that

170
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
the relationshipbetween the communicants may be endangered if
the change does not take place, can be called the "sweat box/' The
"sweat box," in mild or severe form, often seems to be necessary
before morphogenetic or basic structural changes can take place in
a person, in a family, or in larger systems like tribes or nations.
It is important to note that if and when a move in an appropriate

direction is taken, there must be immediate confirmation and re-

ward. The essence of the double bind is to disconfirm a leap once
taken, to indicate that change is not desired, or to disqualify the
whole event. Thus the double bind could be described as a simple
bind that is continually imposed and then continually lifted; pres-
sure to change followed by injunctions not to change; a yes-do,
no-don't kind of thing that produces the disruption and pain that
Bateson argued were untenable for humans and other creatures.
Rabkin, carrying this idea further, states that a paradoxical injunc-
tion that brings about a systems change followed by a paradoxical
injunction to undo that systems change might well result in intense
disorganization in the recipient of such messages.
Take the example of a mother caught in a struggle with an adoles-
cent son. She wishes him to display more adult ("symmetrical")
behavior. But if she enjoins him to do so, she is defining him as a
child (a "complementary" relationship). There is no way out of this
difficulty, as every exasperated parent and resentful teenager

knows, except through some shift whereby both find that they are
relating more pleasantly and more as peers than as parent and child,
at least in the area the struggle was about. This shift can take place

suddenly, or a long back-and-forth battle may be required. But the

necessary condition is that the shift in the rule governing their
relationship should happen "spontaneously," since for the mother
to enforce it, or for the child to seize it, would merely reaffirm their
previous situation.
If the parent giving the original paradoxical messages responds
positively to an integration of the relationship at a more equal level,
then this is dilemma. There has been
a successful resolution of the
no double bind, or at least no harmful one. But if at the moment
the child and mother do reach that desired state, one of them, or
someone else in the family, signals that this is bad or might be

171
 Foundations of Family Therapy
dangerous, then you have the preconditions for a double bind. And
then you have the appearance of symptoms embedded in cycles in
which the pressure for change builds up, followed by injunctions
against change, in endless sequence like a stuck record: the famous
"game without end."
The way a simple bind might either become resolved or else turn
into a symptom can be illustrated by this hypothetical case. Thir-
teen-year-old Peter begins to sleep late in the morning and be late
for school. His mother becomes tired of pushing him to get up and
finally says, "Why do I always have to kick you out of bed to go
to school? Act like a grownup. You ought to want to go to school
for the sake of your own future. Your father used to get up at six
and run a paper route before he even got to school in zero degree —
weather," and so on.
This is a bind (simple variety), because if Peter "acts like a
grownup" he is demonstrating a symmetrical relationship, but at
the same time, if he does go to school, it is in response to his
mother's demand, and his relationship to her is thereby defined as
complementary. What he does do is become even more reluctant to
go to school. His mother oscillates between washing her hands of
him and going after him, a process that only escalates the tension
between them. The school staff telephones to say that Peter is
beginning to cut whole days, thus putting on even more pressure.
Father, who can usually sleep later than Peter and hates to get up
early, is constantly awakened by the morning fusses. Although he
prefers to stay out of his wife's dealings with his son, he begins to
protest. "Lay off the boy," he says to his wife, "You're only making
things worse." He compares her to his father, who made his own
adolescence miserable by insisting that he get up and take the paper
route. He says that he can sympathize with the boy. This statement
brings out the latent split within most parenting dyads, the split
between a permissive stance and an authoritarian stance. The
mother, intensifying her position, says, "It's about time you stopped
babying the boy." Father says, "It's about time you stopped nagging
him." They end up shouting and get into a state of unresolved anger
with each other. Peter draws the covers up over his head and suc-
ceeds again in not going to school.

172
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
This is the normal type of confusion a family faces when children
become adolescent. It is usually resolved if the parents can over-
come their differences and establish a united front. Perhaps adoles-
cent rebellion not only serves to establish beginning independence
for a child, but offers an issue that the parents, who by a natural
process will one day be child-free again, can use to test out the
nature and strength of the bond between them. It seems not to
matter which way the parents go; the situation is solved if the
parents can say, "It's your ownmess it up and take the conse-
life,

quences/ or "Get to school and no more nonsense." Somehow,

7

from this microtest of whether the parents (or others in the family
who will be affected) will survive their son's eventual departure, he
gets sufficient confirmation to really begin to leave, and the school
issue drops away. The boy may find that an attractive female
schoolmate waits at the same bus stop. Suddenly it is no longer,
"Why don't you get up and go to school?" but "Why aren't you ever
at home any more?"

Here is the alternative scenario that might establish a symptom.

The boy does get up and go to school. He finds the female school-
mate and also regains his interest in studying (an unlikely story, but
this remains a hypothetical case). However, the father begins to feel

more and more depressed. His work is not going well, and his ulcer
begins to act up. It seems that this is the last child at home and the
one the father was especially close to, all the more in that he has
a rather domineering wife and chooses to remain distant from her
rather than fight anything out openly. The father experienced a
small feeling of elation when the boy defied his mother over not
going to school in a way that was never possible for him when he
was growing up. The boy is very important to him. The mother, too,
is strangely caught up in the fight she has with her son. It is as

though he is able to stand up to her in a way that her husband never

can, and although she is angry, she gains a kind of satisfaction from
his assertiveness. With her husband, there is only shadow boxing;
with her son, someone is really there.
At the same time, perhaps both are unconsciously aware that
the boy's growing up means the emergence of many difficult is-
sues between them, and the father's ulcer seems to signal that he

173
 Foundations of Family Therapy
willprobably turn his feelings about these issues inward, rather
than hazard an open conflict with his wife. A sense of ominous
possibilities fills the air. The little at night and com-
father eats
plains about his ulcer. When
he does, the mother seems annoyed
rather than sympathetic and says, "I'm sick of your always going
on about your ulcer and never going to the doctor about it. I al-
ways have to push you to make an appointment. Why can't you
take responsibility for your own problems making the
instead of
whole family miserable?" The father becomes moody and quiet,
and the son feels his own stomach tighten. He says, "I don't want
any more supper/' and starts to leave the table. Mother says,
"You sit there till we're all finished." Father says, "Let him go, for
God's sake, do you have to run everybody's life like you run
mine?" The evening ends with the boy in his room, depressed, the
father watching TV in silence, and the mother furiously washing
the dishes.
The next day the boy complains that he has an attack of nausea
and cannot go to school; in fact, he throws up. The parents fight
about whether or not he should be made to go school. In the end,
he stays home. This is the beginning of a school phobia. Two
months later, having tried everything and on the advice of the
school, the parents start looking for a psychotherapist. What the
psychotherapist decides falls outside the lines of this story, but a
contextual reading of the situation would be to perceive that the
boy's appropriate behavior in going to school was not rewarded.
Instead, intimations of catastrophe (parental discord, father's ill-

ness) erupted. The polarization of views, permissiveness versus pu-

nitive action, increased, with the boy's symptom now at the center,

maintaining these parental behaviors and being maintained by them

in a self-perpetuating loop. The bind quite evidently cannot be
resolved by a creative leap, such as the boy's falling in love (an
involuntary act that could be seen as an appropriate response to a
simple bind: "He" did not decide to go back to school; "falling in
love" is what decided it). Indeed, the hints of catastrophe increase
when he mentions that he has met a wonderful girl. The leap that
should be made is invalidated not by any one villain, but by the
context, which covertly frames his eventual departure as a betrayal,

a harmful thing.

174
THE SIMPLE BIND AND DISCONTINUOUS CHANGE
This, then, is an example of the way a quite ordinary problem of
growing up could become a symptom. Now let us take a closer look
at the larger configurations that seem to accompany most sympto-
matic displays: the patterns of behavior that reinforce a problem
while at the same time attacking it, and that exemplify in systems
terms the double-binding sequences said to "drive men mad/ 7

175
Chapter 10

The Thing
in the Bushes

What-to-Change

Up to now the family therapy movement has done better in the

area of how-to-change-it than of what-to-change. Descriptions of
the creature that family therapists are out to get have been notori-
ously unsatisfactory. Clinicians know that there is something rus-
tlingabout in the bushes, but nobody has done a good job of finding
itand explaining what it is. It has eluded efforts to put it in terms
of communication patterns (the double bind, for instance), as well
as attempts to be more global and tie it to a type of family structure
(Minuchin's "enmeshed" family; Bowen's "undifferentiated family
ego mass"). Qualities or traits indicating a family guaranteed to
produce dysfunction, such as Wynne's "pseudomutuality" and
Bowen's "fusion," are suggestive but poorly attached to any partic-
ular symptomatic configuration.
As the search continued, the triadic concepts of coalition theory
seemed to point to a more useful unit, one larger than the inter-
change but smaller than the family. Constructions like Haley's

176
THE THING IN THE BUSHES
cross-generation coalition, or Minuchin's depiction of "rigid triads/'
seemed to be going in the right direction but were static. On the
other hand, the emphasis of Watzlawick, Weakland, and Fisch in
Palo Alto on tracking the behaviors around the problem, although
process oriented, were not tied sufficiently to the larger context.
The Palo Alto group, in adopting a cybernetic analogy, seemed to
be going in the right direction. A symptom or problem looked "as
if" it were controlling or monitoring behaviors in a face-to-face
relationship, so that they do not pass beyond certain limits. Con-
versely, it looked "as if" the problem were being supported and
controlled by the context in which it appeared. But the context is
in actuality an ecological field made up of more than one level of
system, and the problem acts as a contrary presence, urging the very
changes it apparently prevents, and ambiguous regarding the level
on which change, if it does take place, may occur. Bateson puts it

very well:

''Stability" may be achieved either by rigidity or by continual repetition

of some cycle of smaller changes, which will return to a status quo ante after
every disturbance. Nature avoids (temporarily) what looks like irreversible
change by accepting ephemeral change. "The bamboo bends before the
wind," in Japanese metaphor; and death itself is avoided by a quick change
from individual subject to class. Nature, to personify the system, allows
old man Death (also personified) to have his individual victims while she
substitutes that more abstract entity, the class of taxon, to kill which Death
must work faster than the reproductive systems of the creatures. Finally,
if Death should have his victory over the species, Nature will say, "Just

what I needed for my ecosystem." 1

This argument allows us to get out of a linear trap. At each level

of structure stability and change have different implications; if we
choose to emphasize one implication over another, we are "chop-
ping up the ecology," to use Batesonian terms again. Nevertheless,
to see more clearly, one must at times put a little circle around a
phenomenon, as one would single out an area of the abdomen to
sterilize and prepare before doing surgery, temporarily forgetting

that the organ we are going to tamper with or take out is linked with
a living human being, in a family, in a world. So this chapter will
be an exercise in "chopping up the ecology." We shall be looking
at a problem in a triadic context even though this does injustice to

177
 Foundations of Family Therapy
the richness of concentric rings and levels within which each behav-
ior is embedded.

The Mystery of the Essential Variable

There is no one value or factor in the family that a symptom can

be said to be associated with. The closest concrete proof that there
are any such variables at all has been offered by Minuchin in his
experiment linking stress reduction in parents to a successful at-
tempt to pull the symptomatic child into their argument. One vari-
7
able here would be "conflict between the parents/ and the assump-
tion is that it must be kept from surfacing, for whatever mysterious
reasons. Minuchin's families were a psychosomatic sample, though,
and there is between somatization of stress and conflict
a strong link
avoidance. This variable would apply only to families in which
open conflict is toxic.
There are other families with constant arguments between the
parents or other family members and extreme symptomatology in
the child. In many such cases the arguments occur only around the
symptomatic behavior. Alternatively, incipient battles between the
parents seem to be averted or diverted by the symptomatic behav-
ior. Whether the conflict is open or hidden, the symptom still ap-

pears to be part of a recursive cycle or set of many such cycles,

which seem to hover around the possibility of change.
The strains in the family which seem to trigger a child's symptom
are not always (or only) between a pair of parents. They may in-
volve a mother and a grandmother, or a mother and a parental child,
or a wife and her mother-in-law. The opposing parties may be two
competing clans, as in Romeo and Juliet, or may be outside the family,
as when two therapists disagree about the handling of a case. There
seems to be one common feature in all these cases: The symptom
arises in a more expendable party when the relationship between at
least two other parties —
who often constitute an executive unit or
—
are otherwise extremely important to the group is threatened. The
nature of the threat can be answered if one knows what the possible

178
THE THING IN THE BUSHES
consequences of the disappearance of the symptom might be, but
this is not anything that can be predicted. One must guess. In the
case of the Capulets and Montagues, one could argue that had not
the lovers died, the two clans might have engaged in a destructive
war; instead they peacefully united. If the husband whose ulcer
seems to unite his wifeand mother recovered, the marriage might
be threatened, as the conflict between wife and mother (or, poten-
tially, between any other pair) might surface. It does not matter if

the danger seems real or not.

In the case of a single parent,one finds that abdication or depres-
sion in the parent is one variable that seems to need to be kept
within bounds. Sometimes much of the child rearing is left to a
parental child. In such cases, a "bad" child (one who causes acute
disruptions, creating trouble at school, starting fires, and so on)
probably behaves this way whenever the mother abdicates her re-
sponsibilities or becomes too depressed. The disruptive behavior
not only seems to bring mother back into the picture but unites
mother and parental child against the culprit. The cycle is clear,
since this coalition inspires the "bad" child to make fresh trouble
as soon as the mother once more starts to abdicate, leaving the
parental child once more in a vulnerable position.
The Interfering Grandmother, a common feature of single-parent
families in which the young mother is very dependent on her own
mother, offers a variant. Here one variable is the maintenance of the
relationship between the two women. Often a child who is
"spoiled" by the grandmother becomes the stabilizing factor. His
difficult behavior helps to keep the young mother dependent on her

own mother, while the favoritism shown by the grandmother to-

ward the child forces a wedge between them. Closeness and distance
are monitored by this particular arrangement. If mother acts too
independent, the child defies her, forcing her to rely on grand-
mother to help keep the child in line; this in turn works to make the
mother feel even more like "leaving" and the cycle continues.

179
 Foundations of Family Therapy

The Homeostatic Seesaw

Let us, for the sake of simplicity, confine ourselves to one type
example: the case of the child whose symptoms or problems seem
to be monitoring a mother-father relationship. In a large proportion
of these cases the parents present what looks like a very uneven
marriage. One partner seems to be the "strong one" while the other
ismore dependent. Family researchers like Robert Ravich have ob-
served that the "complementary" or one-up/one-down pair makes
up one of the largest groups in their clinical population. 2 Attributes
of this structure seem to be: (i) an intense clinging; the pair can be
deeply unhappy, but will put up with utmost misery rather than
separate; (2) in many cases an equally intense avoidance of conflict
or behavior that might call the relationship into question; and (3)
an inordinately high proportion of children who are disturbed. The
marriage can seem very happy, with both partners apparently con-
tent, and yet they may have a psychotic child. However, their
apparent contentment sometimes disappears if the child stops being
a problem. In such cases one might almost say that the more severe
the child's problem, the more severe the "trade-off" will be in terms
of somatic or psychological illness in a spouse or other relative, or
the emergence of marital difficulties.
Looking at the supposed happy pair with the disturbed child, we
may well ask, "How can this be? How can there be a conflict that
does not appear and that the participants do not experience?" The
answer is provided by a fascinating study by Cynthia Wild and her
coworkers that focuses on communication disorders in families with
a symptomatic member. 3 In a research project comparing the com-
munication styles of families of hospitalized schizophrenic males
with those of a control group of hospitalized males with character
disorders, the authors found, among other things, that the behavior
of an unusual number of the fathers with schizophrenic sons could
be described as "overcontrolling," while the communications of the
mothers were classified as "amorphous." This combination allowed
one spouse, the seem dominant, but a close reading of
father, to
conversations showed that mothers, by the use of non sequiturs,

180
THE THING IN THE BUSHES
scattered thinking, topic changes, and the like, were able to nullify
any decisions the fathers might try to make. The authors point out
7

that these behaviors have a mutual causal effect: "Mothers vague-

ness increases the likelihood that fathers will take over and control
situations,and fathers' arbitrary and often irrational style of control
77
7
increases mothers vagueness. One can see how these linked
behaviors would minimize any appearance of disagreement be-
tween the parents, while still fostering an intense struggle. Wild
speculates that this arrangement might also account for some of the
confused thinking of their sons.
As we observed earlier, this kind of uneven marital arrangement
seems to produce a disproportionate share of disturbed children. We
might think of a child caught in the situation Wild describes not so
much as a victim of a confused environment but as part of a family
balancing we stick to our cybernetic analogy, and consider the
act. If

relationship between the parents to be governed by set limits, like

a seesaw that can go only so high or so low, we see how a child can
influence the tilt. If he sides with one parent against another, or
makes trouble for one of them with the covert support of the other,
this behavior influences the balance of power between them.
Suppose the mother, in the marital subsystem, takes the one-
down position. Add the child as the secret ally of the one-down
parent, and add also a behavior that seems guaranteed to provoke
the one-up parent. Rage and bluster though he or she will, the
one-up parent cannot do anything with the child, and his authority
is reduced to nothing. The seesaw tilts upward. But by and by it will

get too high, whereupon the authoritative spouse probably will

begin to push the partner down have the child come
again, only to
in before the seesaw gets too low. The cycle is no longer dyadic, as
7
in Wild s study, but involves a triangle. At the same time, we must
guard against the linear assumption that the child's behavior stabil-
izes the marriage. We are dealing with circular chains in which no
one element controls or serves another.
When a couple's struggle is open, the symptomatic display of the
child seems to block the fighting. Such a couple may only start to
bicker and the child will come in with his usual number, an asthma
attack, an obnoxious behavior, or whatever. If the problem is a
somatic one, the parents are more apt to unite in anxious concern.

181
 Foundations of Family Therapy
If it is a disruptive behavior, they may unite to scold the child. But
scratch the surface and the mirror-image disagreement about the
problem, or at least about the management of the problem, will
come out. Beneath the appearance of unity, one parent can be seen
to be less upset or more protective; the other is more upset or more
punitive. However, it is important to note that the parents' fight
does not call their own issues into question, since they are merely
disagreeing over the problem of the child.
This "homeostatic seesaw" can be seen as a mutually sustained
imbalance that holds the parents together. Symmetrical couples
seem to have little trouble fighting —in fact, that is what usually
brings them into treatment, rather than a somatic symptom or prob-
lems with a child —and they also have less trouble deciding to
divorce. Complementary one-up/one-down couples are far more
or
locked together, with the supposedly more powerful partner in
reality just as fragile and as dependent as the other. Seen only in a
linear dimension, the child's behavior keeps this seesaw within safe
confines. If it become too uneven, the one-down parent
were to
might become depressed or develop a symptom. If, on the other
hand, it became too even, the couple would be more symmetrical
and there would be the danger of splitting, or (in the case of abusive
spouses) violence might erupt, which would endanger one partner.
Oddly enough, if the child's symptom disappears, it is the one-up
parent who is at most risk for a symptom, as though it were his job
in the absence of the child to prevent the seesaw from going into
reverse tilt and endangering the other spouse.
A proviso must be made here. Families in which severe blowups
occur periodically between the parents do not necessarily present
"symmetrical" spouse relationships. In particular, the presence of a
severely symptomatic child predicts the opposite. If one looks
closely,one finds that the "pseudohostility" of these families is part
of the sequence that enshrines a symptom. There is probably a
hidden seesaw operating, and the child "knows" the cue that indi-
cates when it is time for him to come in and break up the quarrel
between his parents' or other family members.
Lidz's classic article, "Marital Schism and Marital Skew," de-
scribes the spouse relationships in eight families with hospitalized
schizophrenic children. 4 In some the conflict went underground,

182
THE THING IN THE BUSHES
with one parent deferring to the other; in others the parents fought
openly. But even in the "schism" cases, one spouse was described
as more controlling and the other as more compliant. If our model
is correct, one outcome of the child's symptom was that the realistic
possibility of splitting never arose.
A simplified example of such a cycle might be something like this.
Six-year-old Tommy has tantrums and is difficult to control. Close
inquiry into the context of the tantrums reveals that they tend to
get worse at suppertime. Father, who is an old-fashioned pater-
familias,works long hours and comes home fairly late. Mother, a
nonassertive, domestic person, sets great store by a family dinner,
and so she makes Tommy wait for his meal till the family is ready.
She much care into making a good supper. The preceding
also puts
Friday,we learn, she made a dish which her husband specially likes.
That night, when it is time to eat, she calls the family to table. As
usual, Father is so deep in the newspaper that he has to be called
more than once (we frequent reaction of
learn later that this is a
Father to any request made to him by his wife). Tommy sits down
with Mother, and she begins serving. She is annoyed because Father
has still not come to the table, so she reacts with irritation when
Tommy Tommy," she
whines, "I don't like that stuff." "I'm sorry,
says, "it's all we have for supper." Tommy still refuses to eat. Father
puts the paper down, comes to the table, and says, "Tommy, eat
your food!" Tommy looks at Father and pushes food off the plate.
Father says, "Okay, no dessert."
Now Tommy throws himself on the floor and begins to kick and
scream. Father grabs Tommy's arm and drags him upstairs. He puts
him in his room and slams the door, coming down to take his place
at the table. Instead of serving him, his wife is standing with a
pained look on her face. "What is it, for God's sake?" asks Father.
She responds by asking, "Why are you always so hard on the boy?"
Father, in a rage, leaves the house and spends the evening at the
corner bar. Mother fixes a plate of ice cream and goes upstairs to
Tommy to calm him down. When Father comes home, late at night,
she is in bed asleep, and for two days she is very cool to him. Father
affects not to notice but starts to go out of his way to be nice to both
wife and son, and so calm is restored, until the next tantrum.
If one looks at this episode as a recurring dramatic event, one is

183
 Foundations of Family Therapy
likely to ask, "What is the meaning of Tommy's behavior for differ-
ent people in this family?" Some guesses could be made. The
mother seems to feel that she has very few rights regarding her
husband, and one assumes that she takes a one-down position in the
marriage. However, as tensions build up, one assumes that there
will be a greater tendency for the cues to arise that set off a tantrum.
Mother, in siding with the boy, makes her husband feel not only
ineffective but criticized and excluded. One might say that the
seesaw becomes rebalanced by the addition of the child and his
problem. In the parental subsystem, the mother temporarily "wins."
At the same time Tommy's misbehavior is reinforced, both by
Mother's extra comforting and by a very obvious reduction of ten-
sion all around. If one goes outside of this very artificial depiction
of a parent/child triangle and includes other family members or
important persons such as Father's mother or an older sister, one
gets a far more complicated set of interlocking feedback chains, but
the general principle remains the same. An unusual closeness be-
tween Father and his mother, or the beginning of a struggle between
Mother and the daughter who was always her best helper, may be
aspects of a family dilemma of which the boy's symptom is only the
most visible sign.

Couple Cycles

Since a piece of the total symptomatic sequence is often a regula-

tory arrangement between spouses, some attention should be given
to research in this area. One could, and probably should, make an
argument for not treating couples as a separate universe. It is en-
tirely possible that no purely dyadic cycle exists independent of
third parties. These paragraphs, therefore, are a separate section
only because couples have so often been studied (and worked with
in therapy) as independent entities, rather than as pieces of more
complex balancing acts. And we must remember here that some of
the Bateson group's most interesting studies of complementary and

184
THE THING IN THE BUSHES
symmetrical sequences were based on observations of spouse in-
teractions.
Jackson in particular had a genius for describing couple interac-
tion as it related to workable, unworkable, or difficult marriages.
Since the profile of the couple can determine the design of therapeu-
tic interventions, it is worth expanding here on his couple typology.
In Mirages of Marriage, Jackson arranges couple types from best to
worst. The Stable-Satisfactory marriage is on top, with its two
subgroups: the "Heavenly Twins" and the "Collaborative Geni-
uses." The Stable-Unsatisfactory marriage contains the "Spare-
Time Battlers" and the "Pawnbrokers." The Unstable-Unsatisfac-
tory marriage is characterized by two miserable pairs: the "Weary
Wranglers" and the "Psychosomatic Avoiders." At the bottom of
the heap are two love-matches made in hell, the "Gruesome Two-
some" and the "Paranoid Predators." 5
The interest of this rather dashing set of categories is that they
do not assume a rigid set of attributes, but assign what we have
come to see as a "world view" or "paradigm" to these couples.
Jackson made it clear that no one couple is ever a pure representative
of any of these forms. In fact he had an uncanny ability to spot
positive elements even in unhappy marriages. For instance, he con-
cedes that the "Spare-Time Battlers" usually do not seek profes-
sional help; they may fight, but they get enough out of the family
aspects of the marriage to keep them going, and they are less apt to
have problems with The "Pawnbrokers" are the covert mar-
sex.
riages of convenience which people will put up with as the lesser
evil to remaining alone. The "Weary Wranglers" are far more path-
ogenic, and probably would show up in Lidz's category of "marital
schism" which Lidz associated with psychosis in a child. Jackson
agrees that these couples may have from severe
a child suffering
pathology. The "Psychosomatic Avoiders" underground ver-
are an
sion of the Wranglers; they are unable to express anger openly, and
often one of the spouses or both has a stress-related psychosomatic
complaint. Sexual difficulties and drinking problems are another
way in which these couples express their discontents. The Avoiders
often present themselves in a "sick/well" formation, or take the
familiar one-up/one-down position of apparent victim and appar-

18
 Foundations of Family Therapy
ent victimizer. I would add that these couples may have highly
disturbed children if a spouse does not take on the symptomatic
role.

But the couples who are the most disturbed, in Jackson's eyes, do
not seem disturbed to themselves. The most mystifying category is

the"Gruesome Twosome/' who are perfect turtle doves, have never

had a cross word in twenty years, and only present themselves in
therapy because of a symptomatic child, most often a psychotic one.
The last group, the "Paranoid Predators," maintain their closeness
in opposition to a supposedly hostile world, and again their rela-
tionship, though they perceive it as happy, may be extremely toxic
to their children.
Jackson makes one further important point about these catego-
ries: Couples can move up or down this ladder as they go through
life. The Stable/Unsatisfactory Couple, for instance, may drop into
the category of Unstable/Unsatisfactory if one of the spouses finds
a new partner and realizes that there is more to life than he or she
had assumed. And in therapy, even though some categories seem to
promise less hope for relief than others, there is always the chance
that a couple may be helped to get to a less stressful place on the
ladder of discontent. Of course there is one proviso. The apparently
contented marriages in the last two categories, if they come into
therapy with a disturbed child, may become less contented if ther-
apy works. The trade-off for a nonsymptomatic child may be that
one of the spouses will become emotionally disturbed or medically
ill, or that the couple will break up. But this is still, perhaps, a better
human solution than the previous one.
A more recent attempt to link therapy with a typology of married
couples comes out of Robert Ravich's Interpersonal Behavior
Game-Test, or Train Game. 6 The game is set up so that the two
partners each have toy trains going in opposite directions, with one
section of track passable by only one train at a time. There is an
alternate route that takes longer, but neither person can see the
other's side of the board; and each partner can lower a gate to the
other person's access to the direct route. Collisions occur frequently
as a result. The payoff is based on the amount of time it takes both
partners to get their trains to the end point. Obviously this is a game

186
THE THING IN THE BUSHES
in which cooperation and communication are more essential than
competition.
Ravich isolates three major patterns. The ''Competitive" pattern
resembles a price war, as each partner tries to inflict the most harm
on the other to gain the most for himself. Of course in this case the
couple's joint score will be extremely low. The second pattern is

"Dominant-Submissive," with one spouse habitually letting the

other take the direct route, either taking the alternate one himself
or waiting till the direct route is free. The third pattern is a "Cooper-
ative" one, in which the couple takes turns using the direct route
in a polite manner. Couples who use the two latter patterns often
score highly.
Ravich interviews these couples after they take the test, and has
found that their behavior with the trains reflects the way they
handle issues in other areas of their lives. However, none of these
patterns seems to predict for or against married bliss. The competi-
tive pattern, which seems most destructive, may not be so,
like the
for these partners are at least in contact with each other and there
is a certain balance in their respective strengths. The uneven pair

may do better in arriving at decisions, but the submissive spouse

often pays dearly in the form of depression or other symptoms. The
evenly cooperating couple should present the ideal state, but even
there a hitch may arise. Ravich has found that such couples are
often using this pattern for mutual avoidance, and that one of the
spouses is apt to be having an affair on the side. In a study of a small
group of couples that took turns by using the direct route-alternate
route pattern, insuring minimum contact, he found that every one
ended in divorce. Ravich's therapeutic stance is that whatever pat-
tern a couple is using, the therapist should help them to add others
to their repertoire, variety and flexibility being the goal rather than
any particular form of decision making.
Ravich's research did not lead him to a model built on cyclical
lines, even though he often noted sequence regularities. For in-

stance, there was one version of the Dominant-Submissive pattern

which he called the Flipflop Phenomenon. One spouse would habit-
ually take the direct route with the other dutifully following after,
but after a certain number of trials the roles would be exchanged,

187
 Foundations of Family Therapy
with the submissive partner taking the direct route. This flipflop
would usually occur in a cyclical pattern, as predictable, Ravich
comments, as the variation of the tides. He feels that this is an
antistress mechanism that mitigates the uneven balance of these
one-up/one-down marriages and may account for the longevity of
these unions even though they constitute the most clinically un-
happy group. Interestingly, he finds that couples with problems
related to drug or alcohol abuse usually fall into the Dominant-
Submissive category, and they also, one assumes, use this flipflop
pattern for relief.

This brings us to another observation. Most work on couple and

marital problems has stressed the problems of one or both of the
pair,and not the sequence of behaviors in which these problems are
embedded. It is perhaps because this sequence is such an obvious
one with alcoholic couples that the literature on this problem has
been the first, outside of the family field, to focus on the interaction
context. There has been a notable shift in ideas on how to treat
alcoholism from working with the individual, to working with the
dry spouse, to working with the whole context of maintenance
people, including the would-be rescuers who only spur the alco-
holic on to more heroic boozing.
One recent study of alcoholic couples by Steinglass et al. indicates
that their cycle has a wet period and a dry period, as a tropical
climate has wet and dry seasons. 7 Both are essential to the ecology
of the couple. Certain behaviors can take place only during the
"wet" season (like sex or fighting) and are prohibited during the
"dry" season. One can also notice that the shift includes a monitor-
ing of relative power positions, much like the child-parent seesaw
described previously. The apparently one-up position of the re-
sponsible dry spouse is effectively challenged during the "wet"
season, even though it is reinstated with a vengeance the morning
after. And like any other symptom, the drinking glues both spouses
together, since the drinker is automatically defined as weak and
needing care.
There growing evidence that spouse abuse is also a cyclic phe-
is

nomenon. Berman, Pittman, and Ratliffe suggest that here again is

an "overadequate" spouse (the abused one) and an "underade-
quate" one (the abuser). 8 The abuser is often a man who feels or is

188
THE THING IN THE BUSHES
socially, culturally, and financially inferior to his wife. But the wife
also feels very insecure, and apparently needs a man who is ex-
tremely dependent on her. Triggering behavior seems to occur when
the one-down spouse low (when the wife has gotten a
feels too

raise, has been out seeing friends too much, or simply starts acting

too independent). An episode of physical abuse follows, which

reequilibrates the relationship and is in some way accepted by the
abused one. The aftermath of the beating can be remorse, forgive-
ness, renewed tenderness, at best; at worst, the "superior" spouse
will have a sense of intimidation and powerlessness which may
nevertheless act as a security bond for the relationship. Evidence for
this curious premise is the extraordinary attachment these partners
have for each other. If the wife leaves, the husband will scour the
world for her, and she, even when safely ensconced in a shelter, will
often manage to leave some clue that allows him to find her. If the
woman does escape, there is evidence that such a man will simply
find another woman to fit the vacant niche.
What we may be seeing skewed relationship in extre-
there is a
mis; a complementary schismogenesis which is always tilting two
ways: either toward annihilation of the abused by the abuser or
abandonment of the abuser by the abused, in an endless oscillation.
Unlike the alcoholic couple cycle, this one has not yet been "discov-
7
ered/ and so the usual treatment plan calls for an assumption that
the woman is a victim and will cooperate in efforts to leave this
unsatisfactory relationship. This assumption ignores the profound
significance the arrangement may have for both partners and the
potentially devastating consequences of change. A natural history
of this cycle, which will lead to more intelligent strategies to break
it, has yet to be done.
Another fascinating class of couple cycles falls in the purview of
doctors who treat somatic problems: ulcers, headaches, arthritis,
cardiac problems, and a host of other far-from-minor ills. The num-
ber of marriages held together by a somatic illness is legion. Stranger
yet are cases where both spouses and are
are in a suffering contest
competing over who is the sicker one. Here the competition is to be
theone-down partner, yet the competition itself is symmetrical.
Some grim pairs will literally compete with each other to the death,
if need be.

189
 Foundations of Family Therapy
Last, not least, are the couples whose bond seems to rest upon
psychiatric symptoms — depressions, periodic psychotic episodes,
obsessions, phobias, anxiety attacks — the price many people will
pay for the assurance that they will never be alone. Insufficient
attention has been given to the cyclical nature of these problems in
the life of a couple, and to the interlocking behaviors of the presum-
ably "well" spouse; what are his or her benefits from the arrange-
ment, and what is the price? One of the few serious studies of
couples in which the wife was periodically hospitalized for
psychotic episodes is that of H. Sampson, S. L. Messinger, and R.

D. Towne. 9 Their documentation of the cyclical nature of these

episodes and the part they played in the family relationship picture
deserves attention. Of most interest is their classic account of a
group of wives whose periodic departure into the hospital allowed
the entry of the husband's mother into the home. The husband-
mother tie seemed to be monitored by the hospitalizations and
recoveries of the wife, just as the husband-wife tie was monitored
by the availability of the older woman, example of a
in a fine
two-generation adult relationship cycle. Couple problems, in any
case, often involve other family members or third parties in roles
that are crucial in maintaining the problem, and this is true to an
extent which has not been sufficiently realized.
Some other books and articles on couple therapy that should not
be missed include Carlos Sluzki's highly original piece "Couples
Therapy: Prescription for a Systems Experience," which expresses
a strategic conceptualframework through the use of clinical vig-
nettes; Peggy Papp's inventive "The Use of Fantasy in a Couples
Group"; Norman and Betty Paul's A Marital Puzzle, which focuses
upon unresolved mourning; and Clifford Sager's Marriage Contracts
10
and Couple Therapy, a highly regarded text in the field.

Power as a Family Issue

A discussion of what brings about distress in a family system

leads directly to a critique of the basic assumptions some researchers

190
THE THING IN THE BUSHES
have made about family interaction. Much of the work we have
discussed in this book seems to posit power issues as the basis for
—
family difficulties whether expressed as open discord or hidden by
disqualifications. But power is never an absolute item; it always has
to be "power for what?" In an absolute monarchy the answer is
simple: "Power to make my subjects do as I say." In a war between
countries "power to subdue,
it is if not crush, my adversary." In a
game it is "power to win."
But in a family, even an authoritarian patriarchy of the old school,
the issues are not so simple, because the objectives of a family, even
allowing for cultural difference, are not like those of parties who
have no stake in each other's well-being. One
assume also has to
that the family does have a particular "good" which everybody
wants a part of, or a "product" which no other institution can
replace. But what are these goods or products? What can a family
do that the state or some other group cannot do? A family can exist
without being an economic unit; a family can exist without being
a child-rearing unit; these and many other functions can be taken
over by other parties or organizations.
There is only one invisible but important task which few other
institutions can perform as well. This has to do with an orderly
access to intimacy. It may also be related to an invisible systole and
diastole of connecting and withdrawing shared by all the social
animals. This unconscious but orderly arrangement can be a func-
tion of the nuclear family, but it can also extend to the borders of
the face-to-face community in which the family lives, or include
connections with extended kin.
It is possible that this "social envelope," to quote Kai T. Erikson,
is as important for individual survival as amniotic fluid is to an
11
unborn child. and poignant study of the after-
In a masterful
effects of a communal disaster, the Buffalo Creek flood that deva-
stated an entire mining community in West Virginia, Erikson makes
the point that the survivors were unable to get over the impact of
this event not only because houses and people were destroyed, but
because the fabric of the community, rooted in history, proximity,
and time, was ruined beyond repair. This is the first time, to my
knowledge, that a social scientist has made such a strong statement,
backed by such ineluctable evidence, that individuals need a net-

191
 Foundations of Family Therapy
work of people, obligations, customs, houses, furnishings, objects of
reference, space-works, time-works, all of which go to make up
each person's "social envelope. " Without the surroundings to which
he is accustomed, the individual may go on living but his will to live
may not. To a man and woman the survivors of this flood continued
without meaning and without hope
to experience life as disjointed,
long after the event. They seemed to have a mass neurosis.
But Erikson points out that this looks like a neurosis
if one only
examines each individual complaint. Taken as an aggregate, the
findings point toward a reality, not a neurosis. The people who
survived the flood were relocated, but in flimsy trailer camps with-
out reference to old neighbors or old neighborhoods. Those linkages
that might have remained to build on were ignored. And Erikson
doubts that enough linkages did remain to have allowed this partic-
ular group of people, which had been grounded in a special commu-
nal arrangement, to survive in any functional way at all.
Moving from the particular to the general, it seems probable that
a "social envelope," like the family, is distinguished from all other
kinds of social organizations in at least one respect: the regularity
of the rhythm that pulls individuals together and pushes them
apart. The person who has written most cogently on the makeup of
this interactional ebb and flow is Eliot D. Chappie. In a book called
Culture and Biological Man, Chappie describes interaction rhythms in
ongoing relationships as analogous to and connected with circadian
rhythms in physiological processes. 12 If the two are, in fact, con-
nected, an upset in biological rhythms may coincide with an upset
in social rhythms and vice Both types of rhythms require a
versa.
high degree of internal calibration, and there are optimum frequen-
cies and intensities involved.
Chappie has hypothesized that in social interaction there may
even be a daily "interaction quota" for each individual. If a person
does not get some as yet undefined degree of this requirement, it

may bring about untold deleterious effects on the physiological

—
rhythms of his body affecting blood sugar, sleep patterns, and the
like, which may in turn get out of kilter with each other.

At any rate, it is reasonable to hypothesize that any kind of

satisfactory interaction between people in a family would involve

192
THE THING IN THE BUSHES
a balance of giving and getting, and of being touched and being left

alone. Chappie goes so far as to say:

Each individual needs to interact for so much time, with so many people,
as well as to experience intervals when he is by himself and not interacting.
. Even if each person gets the quota of interaction which his daily rhythm
. .

requires, he also is seeking interaction with his complements. Any old

interaction will not do; he needs to utilize his endogenous rhythms of
action and inaction, at a tempo within the natural limits of his repertoire,
and thus experience a maximum degree of synchronization by the other
13
person.

It would be hard to prove that there is a quota of interaction on

an individual basis, but it is easy to see that there is a highly
stereotyped sequence and frequency of contact for each dyad or
cluster in a family. Once the clockwork of interaction frequencies
within a family has evolved, there would necessarily be a premium
on keeping that particular regularity going. The need for people to
get together on necessary tasks and to accomplish important daily
business itself creates a need for scheduled contact. Simply because
''primitive" peoples do not use clocks and watches to calibrate their
touchings does not mean that they do not use other mechanisms.
Rather, these mechanisms operate below the level of consciousness,
and are built into the periodic routines of everyday life.
And even as physiological processes seem to build their rhythms
on geophysical cues like diurnal and lunar cycles, or the turn of the
seasons, so social interactions also seem to follow periodicities ex-
ternal to themselves. Intimacy between couples often follows peaks
and lows according to the woman's menstrual cycle; women in the
same family tend to synchronize their periods; and the seeming
superstition of workers on mental hospital wards that patients will
become most agitated at the full moon may have a similar explana-
tion in an ebb and flow of social tension following a lunar monthly
clock. The adaptive usefulness of some type of calibration makes it
unnecessary to posit a "need" for rhythm in a person or an individ-
ual "quota" for interaction.
Thus the ability to control access routes to other people —to both
get enough interaction and block it off when necessary —may be of
193
 Foundations of Family Therapy
far greater importance than has been realized. If there were no
question about the equitable distribution of supplies, then the sup-
ply routes would not be in question. But if the supplies are in
question, anxiety about the means for getting them may become
intense.
Consider the following scenario. A mother who is overinvolved
with her baby son, perhaps because his father pays her very little
attention, might show behaviors like picking him up when he wants
to sleep, talking to him when he wants to be alone, waving things
at him, poking him, dressing and undressing him unnecessarily, and
so forth. Depending on the type of temperament this child was born
with, he might develop behaviors for blocking her out: avoiding eye
contact, stiffening on being picked up, resisting being dressed. Later
on, he might refine these behaviors into ways of walling himself off
(and walling off others), become a stubborn, inward kind of person,
perhaps a "workaholic" or an "absent-minded professor," or enter
a career in which contact with people is slight. He might, however,
look like a "strong, silent type" to some unwary female who had,
contrastingly, been ignored by the adults in her early life, and who
had learned to cling to every little shred of attachment as if it were
a prize. How well would complement each other at first
this pair
and how inevitably the struggle for "access control" would over-
shadow their later life. It seems to me that most of the so-called
power struggles that researchers have noticed in disturbed families
are of this type and over this kind of issue. And it also seems clear
that because of the interdependent nature of the "goods" competed
for, there is no way to win unilaterally.

An experiment with newlyweds devised by Harold Rausch and

a group of colleagues supports this thesis. 14 The research was de-
signed to study the behaviors of husbands and wives as they coped
with interpersonal conflict. To this end, some quasinaturalistic
scenes were set up which the couples in the experiment were asked
to role-play. Two scenes dramatized a conflict of plans or interests
involving very specific issues (whether to go out to dinner or to eat
at home, for instance). But the third and fourth scenes involved
exactly this area of access routes we have been talking about. Scene
Three was called Husband Distant. The instructor would tell the

194
THE THING IN THE BUSHES
husband to imagine that he wanted to be alone that evening and
would do anything to accomplish that goal, while the wife was told
that she was to try to engage her husband in any possible way. In
Scene Four, Wife Distant, the reverse was to take place. In addition
to the outcome of these scenes, the tactics spouses used during them
were recorded: avoidance, coercion, reconciliation, backing down,
and so on.
One small group of these newly married couples were called
"discordant," because they reported a high degree of conflict and
marital dissatisfaction. In the issue-oriented scenes, the wives were
quite domineering and used coercive tactics to a much greater de-
gree than wives in the other groups. The husbands, by contrast,
behaved very meekly and readily gave in to the wives. But during
the distance-maintaining scenes, the husbands turned from lambs
to tigers. They not only clung to their instructions to remain distant
at all costs, using coercion and other aggressive tactics which they

had not previously displayed, but continued to use these tactics

when it was the wives' turn to be distant. The wives, in contrast,
behaved far less coercively in response to their newly assertive
husbands than their female counterparts in the other groups. And
when it was their turn to maintain distance, the strength of their
husbands' attacks on their fortresses led to escalating battles and
mutual recriminations.
This little piece of the experiment illustrates the difference be-
tween getting one's own way (power in decision making) and being
in control of access routes (power in seeking or blocking intimacy).
It also shows that the one who looks like the dominant one can keep
his or her own fortress intact but may have a struggle to conquer
the other person's. Ravich's Train Game shows this feature clearly.
Even though a spouse might let the other partner go first on the
direct route, he or she could always lower a gate so that the train
could not get to its destination. In Rausch's experiment one could
see that the wives in the "discordant" group were only apparently
dominant. In the matter of the crucial issue, access to intimacy, they
were powerless except for the option of playing tit-for-tat.
This is only one small example of how struggles around issues of
intimacy could affect a young family right from the start. It is

W
 Foundations of Family Therapy
possible that all major struggles in families with psychiatric disturb-
ances have to do with this issue, which lies at the heart of the
ecology of family life.

Having reached this point, we may consider the present chapter

a kind of watershed or Continental Divide. It is time to call a halt

to this abstract exploration of issues of change and targets of

change, and move to consideration of models of intervention. From
here on, we will be concentrating on issues
7
of therapy —in fact, how
to catch 'The Thing in the Bushes/

196
Chapter 11

Breaking the
Symptomatic Cycle

First and Second Order Change

In describing symptomatic sequences we are dealing with ar-

rangements that influence the schismogenic tendencies of relation-
ship processes, as Bateson first surmised. If this is true, then any
ideas about how to change these processes must deal with the
nature of recursive, cybernetic systems.
We have already cited Ashby in describing the change processes
involved in entities that have what he called "bimodal feedback
7
mechanisms/ As he said, these mechanisms two types of
entail
corrective action. First order changes are the minor fluctuations
from state to state within the limits for behavior that are already set.
Second order changes have to do with resetting the rules for those
limits and usually require a transformation, the discontinuous
change described in Chapter 10.
An example of first order fluctuations in a family would be a
mother who knows she can serve any main dish for dinner as long
as it is not fish. Or a child may know that he can come in any time
after school as long as it is not later than six o'clock.
Second order change applies to any situation in which the usual
range of behaviors is no longer applicable because of developments

197
 Foundations of Family Therapy
in the outer field or inside the system itself. book Change,
In their
Watzlawick, Weakland, and Fisch furnish an apt metaphor for the
two kinds of change by citing the difference between pressing the
gas pedal and changing gears when driving an automobile. 1 In a
family, a second order change might be set off by any major shift
in the rules governing one or more relationships in the family. These
can be associated with unlooked for repercussions, as when a fifty-
year-old woman for the first time in her life defied her ninety-year-
old mother-in-law by insisting on her right to invite a group of
friends for lunch. The older woman entered the hospital with a
serious illness four days later. Or in a couple relationship, a man
who regularly comes home immediately after work to please his
wife suddenly starts coming home at unpredictable times, or a child
starts to disobey his curfew and stay out all night. In both cases the
change is from a complementary to a more symmetrical relationship

on the part of the person whose behavior has changed. But basically
what is at stake is a possible reorganization of the structure of the
family.
Second order changes may also be part of the natural evolution
of a family through time. To take the example of the mother and
food, she may find that lately nothing she serves pleases her thir-
teen-year-old son, who prefers a snack when he comes home from
school and is never hungry for supper. good guess that the
It is a
problem is not really food; the problem is that the rules which
govern their relationship are being challenged. The boy is getting to
be more independent, more defiant. But his mother still expects him
to accept her choice of what he is to eat. It is a familiar struggle that
erupts when children reach adolescence. The mother reacts to the
defiance by mixed resentment at his behavior and anxiety that he
will not get proper nourishment. She intensifies her efforts to make
him eat, and he escalates his refusal to comply.
This is a good example of a problem which develops into what
Watzlawick, Weakland, and Fisch call the "game without end/' In
their language, the solution has become the problem, because attempts at
first order change are being applied when second order change is

needed. The setting for the mother-son relationship has been by the
nature of things complementary, or unequal; now it is, also by the
nature of things, moving toward a more symmetrical or equal one.

198
BREAKING THE SYMPTOMATIC CYCLE
Some families, after a period of confusion, arguments, attempts at
compromise, will take the leap by themselves. Mother will decide
to let the child eat what he wants, or he will go on doing so and she
will stop fighting him. Or they may reach some other compromise
that gives him more autonomy.
In other cases, the relationship setting may be hard to shift, per-
haps because the father never speaks up to the mother, and in a way
depends on the boy to put her in her place. Or the mother may feel
she is losing her baby, the child she turns to when the father is

absent or works late or otherwise distances himself. The escalation

of first order solutions produces a problem that makes the family
suffer more and more, and they finally take the son to a therapist
on the grounds that they can't control him. The therapist's business
is to find out whether, indeed, this is a problem that calls for a

second order change or total reorganization. As the authors of Change

tell us, he will do well to check out the family's previous solutions,
to be sure there is not some obvious piece of advice that would set
everything right. Perhaps the mother has suddenly gone on a gour-
met cooking binge and the son hates spicy foods. If this is worked
out, and peace descends, one can assume that no major structural
change was indicated.
But if the issue really has to do with the range of allowable
behaviors between mother and son, the food is only a symbol of a
A second order change has to be negotiated. The thera-
larger battle.
pist might make a direct attempt to make this change by asking the
mother if she would allow the boy to choose his own diet, at least
for two days out of the week. Again, if this solution is accepted no
more therapy is indicated; the family is amenable to an outside push
to solve the problem. But this is not always true. The mother may
resist any suggestion of this kind because it will mean that the boy
will "win," and she is angry at his attitude of disrespect. This kind
of reaction usually indicates that the therapist has a self-reinforcing
cycleon his hands: a behavior (the mother's reaction) that feeds the
problem (the son's defiance).
The therapist must now look closely to see evidence of this type
of recursive cycle. He notices that the father feels less upset than
the mother about the boy's eating problem. It turns out that the
father has a long-standing struggle with his wife over his crude

199
 Foundations of Family Therapy
table manners. He is not willing to take a stand on his own issue,
and neither he willing to take sides with the boy against his wife,
is

except when the battle between them gets to extreme proportions;

then he has to go to her defense. One could stake out the cycle this
way: The more the father one-down to the mother, the more
feels
the boy acts defiant; the more the boy acts defiant, the more the
mother tries to control him; the more she tries to control him, the
more one-down she feels; the more one-down she feels, the more
the father will come in to help her; the more the father comes in to
help her, the less defiant the boy is; the less defiant the boy is, the
more the mother resumes her domination of the father and the —
whole sequence starts again.

One simple move would be to arrange that the father take his
wife to a fancy restaurant one night a week, rewarding her for all

she has to put up with on other nights and leaving the ungrateful
son to forage for himself. If this suggestion is adopted, the boy will
be taken out of his place in the sequence, at least for that night, and
the couple will have to deal directly with each other. This may force
the problem between them to the surface —or it may be the small
shove that pushes the whole family toward a leap, and they may
return to the next therapy session with the boy no longer a problem
and the couple beginning to rediscover the possibilities of their own
relationship.
Sometimes, of course, neither direct advice nor a structural move
is sufficient to combat the rigidity of the cycle. Here is where moves
that prescribe the symptom or situation come in. These usually
consist of suggestions that fly in the face of common sense. Instead
of trying to eradicate the problem as the family wants him to do,
the therapist begins to point out aspects which might make them
less anxious to have that happen. If the problems were to vanish
overnight, what would the consequences be? Could Father deal
with Mother's energies if they were released from the combat she
is in with the son? Who would bring spirit into the home, which

is otherwise rather lifeless?

In our example, the therapist might reframe the mother's preoc-

cupation with her son's diet as a normal maternal concern that
usually intensifies just at the time a boy is ready to grow up. The

son's defiance is defined as his unconscious wish to attract her

200
BREAKING THE SYMPTOMATIC CYCLE
concern to him, because in reality this new independence is very
threatening to him. The might prescribe a ritual that will
therapist
symbolize the closeness both of them are about to give up. Mother
during the next week is to twice make the kinds of meals she used
to give him when he was small, sitting by him while he eats, maybe
even cutting up his food. She might get out his silver baby cup, just
to give the proper ceremonial touch. Father would have to eat the
same meal, and she would have the right to dictate father's table
manners. After all, he will want to be a good role model for his
growing son. Even humorously suggested, this task usually hits at
—
both sides of the dilemma the very real difficulty all three are
having in saying goodbye to their old positions, even though on the
surface they wish to be rid of them.
Ideally, the reaction will be recoil on the part of all, not only
against the therapist's slightly absurd assignment, but against the
prescription of maternal control over both father and son, which is
pushed far beyond the limits the family, including Mother, is will-
ing to tolerate. This should effectively disrupt the cycle. The after-
math is, of course, unpredictable. What one hopes is that at least one
stabilizing link in the"game without end" will break, and force the
family into the "sweat box." They may then take the desired leap
toward a new integration. Alternatively, some other stumbling
block to achieving this goal will come into view, such as the parents'
difficulties with facing a life together alone. This new problem may
need to be dealt with in a move to a new therapeutic stage. Alterna-
tively, it may simply resolve itself, once the boy is out of the way.
Of course family therapy includes many more maneuvers than
breaking the chain of behaviors in a self-sustaining loop, but thera-
pists who address themselves to relieving specific complaints do
seem to be looking for that problem behavior is a
cycle. If the
disruptive one, it usually stands out clearly and the cycle it is em-
bedded in is not hard to find. This makes the question of how and
where to intervene much easier than when the problem behavior is
a chronically pervasive condition, like many psychosomatic ill-
nesses or communicational disorders associated with psychosis.
Here the clinician may have to work to find the cycle, and it appears
most readily if he focuses on management of the condition rather
than the condition itself. This will usually cause what Stanton and

201
 Foundations of Family Therapy
Schwartz called the "mirror-image disagreement" between the par-
ents to emerge, and the problem can be redefined as one of helping
them to get together so that they can make the child behave respon-
sibly in spite of his "illness."

Breaking a Cycle in the Room

Bateson, as we have seen, used a cybernetic framework to de-

scribe the recursive, cyclical patterns he considered characteristic of
many relationship sequences, especially those dealt with by psy-
chotherapy. His most down-to-earth analogy, described in Chapter
2, was the steam engine with the governor and two weighted arms.
It is important, however, not to take this analogy of a servomech-
anism one cannot equate human mutual-causal processes
literally;

with mechanical ones. Dell, in his previously mentioned article on

evolutionary feedback, points out that in most living systems,
behaviors may occur in recursive or self-reinforcing sequences, but
they never come back to exactly the same spot. They are not like
self-stabilizing mechanisms such as a house thermostat or biological
mechanisms such as the body's arrangement for calibrating its own
temperature. A spiral rather than a cycle is a more useful image, as
this allows for constant fluctuations and change no matter how
stuck such a sequence may be. 2
Thus, when we say "symptomatic cycle," we must take care not
to think that this is anything more than an imprecise analogy for
what goes on in families and other human groups. Although we
have for simplicity's sake picked out only one cycle associated with
a symptom, we are always in these cases dealing with many in-
teracting cycles and loops. We must also allow for the possibility of
change, because as long as the sequence is moving forward in time,
it may seem constricted and unchanging, but there is always the

possibility that one small fluctuation will lead to a larger amplifica-

tion which will set off a leap to a new place.
Keeping these ideas in mind, let us take the case of an asthmatic
child whose asthma seems patently due to emotional stresses origi-

202
BREAKING THE SYMPTOMATIC CYCLE
nating within the family. Assuming for the sake of argument that
(among other things) the child's symptom is part of a closeness-
distance dance between the parents, a self-stabilizing chain can be
described: The more the parents distance, the more the child
wheezes: the more the child wheezes, the more the parents unite;

the more the parents unite, the less the child wheezes; the less the
child wheezes, the more the parents distance; ad infinitum, or until
some factor intervenes to break the cycle.
Watzlawick, Weakland, and Fisch have eloquently described this
type of feedback loop in Change. They define a whole class of psy-
chotherapy problems as deriving from incorrect solutions. This is a
simplified way of describing the self-reinforcing sequence shown
above. In almost any problem, when the common-sense effort to

eradicate it fails, it is probably because the solution itself is part of

what keeps the problem in business. Change uses the example of a
wife who complains that her husband is not open enough and never
tells her anything; the husband reacts to her by clamming up even
more, which leads her to even greater suspicion and more question-
ing,which leads to even more reticence on his part; until the out-

come is a case of pathological jealousy on the part of the wife.

What the authors do not do is to put in the element of "less" that
would keep this cycle from turning into a runaway. Perhaps, peri-
odically, the situation reaches such a pitch that the husband or wife
"blows up," frightening both of them, but at least creating some
willingness to achieve more effective communication. The "less"
here might be that the husband would stop, temporarily, his taci-
turn behavior. One would assume that the wife's suspiciousness
would in turn descend to a more acceptable level.
Perhaps the clearest and most dramatic example of a sequence
that contains a symptom, and of a therapist's successful effort to
derail it, is in a transcript of a family session reported in Minuchin's
Psychosomatic Families. 3 As usual in his first interviews with families
of anorectics, Minuchin arranges for lunch to be served and ob-
serves the Kaplan family's interaction around eating. In this case he
finds that the family will not be moved easily from their rigid
patterns; hence, he assumes that the anorectic daughter's symptom
will not be moved easily either. We will go through this session
carefully, because it not only illuminates a "family dance" but de-

203
 Foundations of Family Therapy
monstrates how the therapist can use his influence to provoke a
runaway, or positive feedback loop, which destabilizes the family
and forces it to change.
This is the first time Minuchin has met with the family. A
brother, seventeen, is present along with the parents. An older
daughter and son are out of the house and live far away. The
youngest girl, fifteen, has wasted away to about seventy-eight
pounds. She is losing one or two ounces a day in the hospital, where
she is staying at the time of the interview, despite a behavior
modification program that usually results in weight gain for such
children. The family seems very well motivated, eager to please the
therapist, and the girl is apparently quite docile, hardly uttering a
word, although she merely toys with the food on her plate. When
the rest of the group have finished the meal, the therapist decides
to ask the parents to make the daughter finish her meal. He then
leaves the room to watch from behind a screen.
The dialogue that ensues can be summarized as follows: Mother,
who also appears mild and unassertive, asks the girl if she wants to
eat the rest of the mother's sandwich. The girl, given her choice,
naturally answers, "No." The rather overbearing father comes in
with a demand that the girl eat. The girl objects, saying that the
dietician has said she can eat what she likes.
As if to join her daughter against the too strong position of the
father, the mother tells the girl that she ought to be able to under-
stand why she has to eat. Mother consistently "reasons" with
daughter, placing responsibility for choice with the girl, rather than
saying as the father does, "Eat because I say so." One feels in this
behavior a hint that the mother's coalition with her daughter is so
important that she will not jeopardize it by forcing the girl to do
anything against her will. It also seems to add strength to the girl's
ability to stand up to her father. Perhaps he now feels threatened;
at any he abruptly interrupts his wife, saying to the girl, "Just
rate,

eat!" This seems to annoy the mother, who turns directly to him for
the first time and blocks him, saying, "Let her finish eating."
Now the scene begins to escalate. The mother repeats her argu-
ment that the girl ought to realize how important it is for her to eat,
and the daughter continues to protest, saying she does not like the
food. Father comes in again, bent on forcing the issue. The girl states

204
BREAKING THE SYMPTOMATIC CYCLE
that she won't eat the food even if it is shoved down her throat. Her
voice now pitched very high. Mother asks, in a soft reasonable
is

voice, why she hasn't ordered something she likes. The girl calms
down. The parents now take turns, with Father insisting, "Eat!" and
Mother asking questions like, "Why are you losing weight?" The
father's stance defines the relationship as unequal, complementary;
the mother's defines it as equal, symmetrical. Father is forceful and
threatening, Mother gentle and comforting. At the same time, how-
ever, the parents are escalating symmetrically vis-a-vis each other,
and the girl is beginning to scream more and more hysterically in
response to this tightening spiral.

At this point the mother increases her pressuring tone, saying to

the "You won't have another chance" and beginning to sound
girl,

hysterical herself. The girl shrieks back. As if to balance off the

mother's shift to a more insistent position, the father suddenly turns
gentle and reasonable, saying that he cannot understand why the
girl is making so little effort to put on weight. The girl again calms
down, but the calm is short-lived, as the father quickly resumes his
domineering tone and the girl once again starts to cry. Now Mother
comes in, quietly rebuking her husband by saying, "Wait a minute,"
and then remarking to the girl, "You don't have to eat much."
Seeing the redundancy of this pattern, the therapist enters the room
to stop it.

This is example of pathogenic interaction. Studied care-

a classic
fully, it can be shown to be a recursive cycle or self-sustaining
feedback loop: The more the father threatens, the more the girl cries;

the more the girl cries, the more the mother protects; the more the
mother protects, the less the girl cries; the less the girl cries, the more
the father threatens; ad infinitum, or until something intervenes to
stop the sequence —quite possibly the girl's death, in this very seri-
ous case.
Looking closely at this configuration, we many features we
see
can place in the context of our triadic concepts. Smoked out by the
therapist's insistence that the parents push the girl to eat, we see a
familiar triangle: the high-status authority, the high-status friend,
and the low-status subordinate. Father takes the disciplinary posi-
tion, Mother the permissive one; and the girl, caught between the
two sets of directives, oscillates from one to the other.

205
 Foundations of Family Therapy
These directives amount to a splitting of the field. Our old friend
the mirror-image disagreement is clearly displayed as the parents
start to fight, using the child as the battleground. We can even see
the phenomenon of the temporary switch-about. When one parent
goes momentarily to the other side, the other takes the vacated
position. Here, when Mother begins to intensify her demands on
the girl, Father calms her down by making his voice low and tender
and taking the mother's ''reasonable
77
approach — at least for a
while.
One can see the beauty of this triangle and its cyclic dance only
if one forgets the seriousness of the problem it generates. As Minu-
chin explains, each person gives the cues for the behaviors of the
others. Daughter knows exactly when to come in, how pitiful the
tone of her voice has to be to activate Mother; Mother knows when
to back off and leave the field to Father; Father knows when to start
his harangue. Of course this all happens on a covert level, not
consciously. And we can see that this sequence is the result of an
aborted transformation, in the sense that it keeps the family from
taking a leap to the adolescent-departure stage of the family life

cycle, this girl being the youngest child.

The paradoxical injunctions, which Rabkin says usually accom-
pany the "time of a leap," are striking in the statements of each
parent. Mother says, in effect, "Eat because you want to eat, not
because I tell you to." But when the girl responds to this paradoxical
directive by asserting a clear symmetrical position: "I will eat what
I want," Father comes in and places her in a complementary position
by saying, "Eat what I tell you to," while at the same time paradoxi-
cally criticizing her for acting "like a two-year-old child."
Not only does each parent give a paradoxical injunction, but the
two types of injunctions are incompatible with each other: if the girl
controls her own eating, she is disloyal to Father: if she follows
Father's orders, she is disloyal to Mother. In a normal situation, at
some point the parents would join together and give some clear and
consistent message, whether to honor the girl's right to choose for
herself (a symmetrical definition of the relationship) or tell her to
obey orders and eat (a complementary definition). In either case, the
girl would probably eat, and this behavior would then presumably

206
BREAKING THE SYMPTOMATIC CYCLE
be rewarded. But the diabolical beauty of this cycle is that any time
one position in regard to the symptom is taken, it is invalidated or
contradicted by somebody by
taking the opposing position. If

chance the girl did start to gain weight, her mother or father would
find some way to remind her not to eat or she would find some way
to remind them to remind her.
At the same time, this push-pull, eat-don't eat, yes-no increases
the girl's agitation. One can see that she is held within a very narrow
band, with upper and lower limits that keep her perpetually under
stress. Just as Father pushes daughter to the breaking point, Mother
stops the action by blocking Father and calming the As soon girl.

as she is calm, Father reinstates the battle all over again. The girl can
never rest. She is caught in an ever-intensifying upward spiral.
From the parents' point of view, of course, their reactions are
logical. Each feels that the other's way will not work and is intent

on checking the destructive excesses of the other. Father thinks: "If

I insisted, my daughter would eat, but my wife keeps leaving it up

to her, so she will starve."Mother thinks: "My daughter would eat

if left to herself, but my husband is making her angry and rebel-

lious, so she will starve." The fight between them is not on any
conscious level. It is another example of the "game without end"
that keeps a symptom alive. Relationship patterns —especially the
mother's covert coalition with the daughter against the father —are
too strong in this family for the parents to take any other positions.
And there is another important relationship pattern: We learn dur-
ing the interview that the father's mother, to whom he is very close,
is dying of cancer. How, then, can the daughter choose the mother
over him under such circumstances? It may not be coincidental that
the grandmother's illness and the daugther's anorexia occurred at
the same time.
Hence the endless cycle that now presents itself, sharpened by the
therapist's insistence that the parents make the girl eat. In some of
Minuchin's anorectic cases the parents manage to get together
enough so that this move is sufficient to break the back of the
symptom. But in this family the parents are too conflictual.
The therapist now decides to embark on a new strategy that will
unbalance the cycle and place the family under considerable stress.

207
 Foundations of Family Therapy
To this end he removes one party at a time from the triangle. In
describing this particular technique Minuchin uses cybernetic lan-
guage rather than his usual spatial-structural terms:

The only way to separate the daughter from the parents is to break the
sequences that maintain the homeostasis. One way the therapist can do so
is to position himself in the system in such a way that the cycle cannot
4
repeat itself.

Thus, in directing the parents to continue to try to get their daughter

to eat (in this sense moving with the symptom-related behavior),
Minuchin introduces one crucial change. First Mother, then Father,
must take on the job alone. By eliminating the constraints each
parent uses to counteract the excesses of the other, the therapist
makes the behaviors in the sequence overpass their usual limits.
We now see the critical limits of the relationships in this triangle.
Mother, when allowed to go to the far edge of her position, becomes
more and more feeble, while her daughter fights her more and more
strongly. At one point Mother cries in despair, "In a mental hospital,
that's where you'll put me!" At the edge of this limit lies breakdown
and death. Father, when allowed to the far edge of his position,
abandons all restraint and tries to use force, grabbing his daughter
by the hair and attempting to cram a hot dog down her throat. At
the far edge of this limit we see the potential for violence.
At this moment Minuchin comes and calmly stops the father,
in
in effect taking on himself the function of the limits which had been
prevented from operating. He places the distraught parents together
and reframes their problem as having a daughter who is in a struggle
with them and is "stronger than you are." He next turns to the girl
and asks her to drop on the floor the hot dog she has clenched in
her hand because "it is not a very good victory." She does so. He
asks her quietly why she has such a need to defeat her parents. Is
it because she wants autonomy and they will not give it to her? She

sits very still, almost in a trance, while Minuchin speaks for her. For

the first time, it seems, the family allows Minuchin to take charge.
During this session a shift occurs. The parents have begun to
blame the daughter for misfortunes she could not possibly be re-
sponsible for, instead of treating her like a fragile doll. And the
daughter, formerly so obedient and docile, has not only defied her

208
BREAKING THE SYMPTOMATIC CYCLE
parents openly but has succeeded in humiliating them in front of

a group of professionals.
Minuchin's explanation is that he has tried, by his intervention,

to move the parents from a "detouring-benevolent" triangle in

which the daughter is perceived as "sick" to a "detouring-attack-
ing" triangle in which she is perceived as "bad." It is certain that
the parents' perception of her, as well as her behavior justifying that
perception, is changed by the end of the session. The
radically
father says, "We always spoiled her, but I think maybe we've been
too good to her." Meantime, the girl goes back with the nurse to the
ward. The session ends with Minuchin talking to the parents solely
—
about their own problems the father's business, the grandmother's
illness —
and the need for both of them to pull together during this
difficult time.

For whatever reason, the daughter immediately begins to eat, and

by the next session has gained almost enough weight to leave the
hospital. When she does go home, however, she becomes a "behav-
ior problem," refusing to eat meals with the family, which causes
a feud between mother and daughter. There is another, apparently
less dramatic shift when the father comes in some weeks later

asking, "Doctor, is it always true that when the patient improves,

the rest of the family gets sick?" It turns out that he and his wife
are having the first problems they have experienced together in the
twenty years of their marriage. There are some marital sessions as
well as family sessions.
Eventually the becomes a fairly normal teenager, with some
girl

relapses along the way, not into anorexia but involving other symp-
toms, including a suicide attempt. No claim is made that the girl is

permanently "cured," even though these other crises were handled

with relatively little difficulty. The point is that the therapy broke
a deadly pattern of constraint that had involved the girl in a problem
which absorbed other family strains at a crucial point in the life
cycle. The example illustrates the suddenness of the shift that can
occur when a therapist disrupts such a sequence and throws the
system into disequilibrium.

209
 Foundations of Family Therapy

Breaking the Cycle Through an Interpretive Approach

Another repetitive, cyclical sequence between father, mother, and

child is reported in the interview "No Man's Land," in Haley and
5
Hoffman's Techniques of Family Therapy. Charles Fulweiler is the ther-
apist. The an ineffectively domineering father,
triangle consists of
a mildly rebellious adolescent son of thirteen, and a mother who
sides with the son. Father keeps getting into an argument with son
over smoking, which both parents say they disapprove of. How-
ever, Mother will break into these escalating arguments to protest,
after which Father will back down. Eventually Father does not even
wait for her to come in; he backs down anyway. This sequence
occurs several times, with variations, during the interview. To give
some idea of the flavor, here is one of the shorter versions; the
mother has just said she is not in favor of taking away the boy's
allowance:

mr. k: (to Mike) Would it please you, I presume it would, it would

please you if I said that it was all right with me for you to smoke: Would
that make you feel better?
mike: Yes, I guess it would.
mr. k: Well, for your sake, Mike, I wish I could, but I honestly and truly
can't. I still don't think it is the right thing to do. I really don't.
mike: Well, by taking my
allowance away, you're, you're both, not
being, I'm not able to buy cigarettes, or I'm not being, I'm not able to do
anything. I can't go to the show, I can't . . .

mr. k: Now, Mike, whenever you have come to me in the last two or
three weeks and asked me for money for some specific thing, I have given
it to you.
mike: Those things I had to do.
6
mr. k: You're right. You're right.

As in Minuchin's family, an example of a sequence

this cycle is
that seems to direct attention away from more threatening issues in
the family. Here at least one such issue involves the marriage. If the
parents began to struggle with each other directly, their own rela-
tionship might be imperiled. One can see that the stakes are high
in favor of some arrangement that will allow the parents to express

210
BREAKING THE SYMPTOMATIC CYCLE
their disagreement but will limit the disagreement to the topic of the
boy's behavior.
As usual, this arrangement means that one parent will take a
positive or neutral attitude and the other a more negative attitude
toward the behavior. The disagreement may be hidden beneath the
parents' concern for the symptomatic child, as in the case of the
anorectic girl, or it may appear as a united antagonism toward the
child, but a persistent effort to get each parent to describe his behav-
ior toward the symptom will usually cause the disagreement to
surface. Most family therapy with a child includes an attempt to
uncover this disagreement and to refocus it as an issue between the
parents. In this family, the disagreement is not so secret as in Minu-
chin's family, but what marital issues are involved is not clear. A
surface reading of the case gives the impression that the marital
dyad is mildly one-up/one-down, with the husband trying to take
an authoritarian stance toward his not-too-compliant wife. We find
out that he maintains considerable emotional distance from her, a
situation she reinforces by investing herself in other relationships,
like a girl friend of whom the husband is jealous, and, of course, the
son. The wife's weaker position in the marital subsystem is counter-
balanced by her strong position in the parental subsystem, where
she has the son as an ally. The son's involvement is able to affect
both the power difference in the marriage and the closeness-dis-
tance axis, so that neither variable is driven too far from its usual
limit.

What has apparently happened to upset the family balance is the

advent of a major transition point: the adolescence of the son. He
is entering a period where he can realistically challenge his father.
On the other hand, the normal pulls of growing up tend to detach
him from his mother and female ways. It is logical that the behavior
that brought the family to therapy was breaking into a tobacco store
and stealing cigarettes and some change. The action also relates to
the major issue between the father and son: smoking. In the repeti-
tive argument over smoking, the problem the boy faces is the mir-
ror-image disagreement between his parents. If he does not smoke
(isnot rebellious), he supports his father against his mother; if he
does smoke (is rebellious), he supports his mother against his father.

211
 Foundations of Family Therapy
The therapist knew it was important to disrupt this sequence. The
methods Fulweiler used were mostly blocking maneuvers, sup-
ported by interpretations. A special feature of his style is to enter
and leave the room without warning. Fulweiser took advantage of
his entrances to inhibit the sequence we have described at strategic
points and turn it step by step into a different direction. He used his
first entrance to cue the mother to be more explicit in her defense

of the son; the second entrance to pick at the disagreement between

Mother and Father; the third entrance to back up the father's au-
thority in the face of the mother-son combine; and the next two or
three to stop the father from playing the role of pathetic victim,
while at thesame time clarifying the mother's part in rendering the
father ineffective. The interview ended with Fulweiler pointing out
that nobody was to blame, since all these events had their roots in
the past. After about nine sessions with the threesome and twenty-
odd with the couple, the father came in announcing a major change
for the better, but pinned the cause to an experience they had in
going to hear Billy Graham. Fulweiler said he was pleased when the
family gave credit to an outside source because they then could own
their own change rather than owing it to the therapist.
Haley identifies standard roles for this type of triangle: the "ove-
rinvolved parent," who seems extremely concerned with the child's
problem and is fighting it actively, and the "peripheral parent," who
seems more neutral and stands apart. 7 Often, of course, the overin-
volved parent is in a submerged struggle with the peripheral parent,
but the peripheral parent is often peripheral in regard to marital
issues too, and will not hazard an open fight. The child seems to act
as his secret agent in successfully defying the other parent. Once
clue that indicates that the child has been fronting for the peripheral
parent occurs when marital issues come and the overin-
to the fore
volved parent criticizes the spouse in the same terms he or she used
to describe the defects of the child.
Assuming that this kind of triangle is operating in the above
example, one sees the son offering his disruptive behavior like
sticky bait to the overinvolved parent, the father. The resulting
argument escalates to the point that the peripheral parent, the
mother, is triggered to step in. The overinvolved parent is at times

the outsider in the triangle, and in this instance much of the father's

212
BREAKING THE SYMPTOMATIC CYCLE
inappropriate anger and sense of helplessness may stem from this

fact. In the closeness-distance area, the wife maintains the upper

hand by having an affair to go to, so to speak, as she moves closer
to her son to make up for her relative weakness in the marital
relationship. In the same way, the husband, by periodically parad-
ing his own helplessness, uses the magic power of complementarity
to glue his wife back to him. Obviously, if a therapist breaks this

sequence, he then has to help the couple with both the closeness-
distance problem and the one-up/one-down problem, so that a
third party no longer gets drawn in as a monitor.

A Parsimonious Technique

If a clinician is which the

successful in identifying the sequence of
symptom is a vital part, a very small change can presumably be
pinpointed accurately enough to have a wide-reaching effect. Wat-
zlawick, Weakland, and Fisch represent an extremely parsimonious
school of family therapy in this respect. This economy of technique
was clearly shown at a demonstration by Dr. Richard Fisch, who
interviewed a couple at the Center for Family Learning in New
Rochelle, New York, in 1976. The problem the family reported on
was embedded in much the same type of sequence we have been
talking about, but the therapist's intervention strategies were com-
pletely different.
In this instance the family consisted of the father and mother,
both in their thirties, a girl boy of six. The children
of nine, and a
were not present. Dr. Fisch asked what the problem was, and the
parents told him they were having difficulty controlling the girl,
who was willful, self-centered, obstinate, and disobedient. She was
so unpleasant that she had only one friend, and even that friend she
treated badly. Mother would get into daily battles with this child
over issues like drinking her orange juice at breakfast. Mother might
win, but these were Pyrrhic victories, as Mother's nerves would
then be shattered. When the father was home in the evening, the
struggle would get to such a pitch that he would come out of his

213
 Foundations of Family Therapy
study and browbeat the girl into obedience. The parents described
their own relationship as close and loving, and stated that they
knew the fault could not be in their parenting, since their younger
boy was as delightful as his sister was difficult. Fisch offered one
small suggestion. He remarked one reason the parents were
that
unable to combat the child's behavior might be because they had
become too predictable. If they were to confuse her by doing some-
thing odd and unexpected, they might have more success. He then
told the father that the next time he came out of his study to scold
his daughter, he should simply give her a penny. If she asked him
why, he was to say, "Because I felt like it," and go back to his room.
This intervention may seem clever but inconsequential until we
examine the nature of the sequence it is designed to interrupt. The
pattern Haley describes is apparent: one parent who is overinvolved
with the symptom of the child, and the other parent less involved
but often secretly allied with the child. In the Fulweiler example, the
less involved parent could acknowledge the coalition. But here,
perhaps because of the need to keep up the appearance of a united
front despite a hint of covert disagreements, the father never chal-
lenged his wife in favor of his daughter. He could go no further than
to say he did not have as negative a view of her as his wife, since
he and the child both liked music and he enjoyed taking her to the
opera. Thus the task given to the father altered the sequence in an
important way: It undermined the myth of the child-monster that
seemed to keep the spouses tightly joined.
At the same time the therapist anticipated with the couple some
of the consequences of altering this sequence. There might be some
side-effects, he said, if their daughter, despite the hopelessness of
the case, did improve. First, the "good" child might begin to seem
less than perfect once the "bad" child no longer offered such a total

contrast. The father agreed that the boy was indeed somewhat
immature for his age, signaling another matter on which he differed
with his wife. Second, the therapist said, the mother might miss the
intensity of her feelings for her daughter, which were a product of
her motherly concern and a proof that she really did love her. The
mother accepted both the positive relabeling of her hostility and the
idea that she might find the change a hard one. Third, the therapist
warned, the mother might at some point have to restrain her hus-
BREAKING THE SYMPTOMATIC CYCLE
—
band who was in fact rather mild and rational from losing his —
temper at his daughter and perhaps being too harsh. If one thinks
about it, these consequences alter or reverse the value of nearly
every relationship reported on in that family.
But the most interesting event of the session came after the thera-
pist agreed with the mother that the most that could be hoped for
was that she would react less strongly to her impossible daughter.
The father then stated that he was not about to give up hope that
easily; he, for one, had higher expectations of his little girl. The
mirror-image disagreement, out of sight till that moment, had come
to the fore.
The results of this interview are not known, but they are not
crucial to an examination of the sequence we are studying and the
methods used to break it up. As we shall see later, the Palo Alto
group specializes in small changes, and even in their tasks they stick
to "small change." The between the limited nature of the
contrast
intervention and the many points of interaction it touched make
this a good illustration of therapeutic economy.

When the Cycle Includes Larger Systems

Haley redefines the problem of working with hospitalized adoles-

cent schizophrenics as one of helping them leave home. The process
which takes these young adults from home to hospital and back
again is merely another version of a homeostatic cycle that monitors
the parents' relationship. Nevertheless, breaking a sequence that
includes not only the family but other social systems is an unusually
complex operation and requires a series of maneuvers which may
take time.
Haley documents this process in a training tape based on a case
in which he was the supervisor and Sam Scott the therapist, while
both were working at the Philadelphia Child Guidance Clinic. The
tape is entitled "Leaving Home," and involves a twenty-four-year-
old deaf man who had been shuttled between home and hospital for
eight years. The sequence was predictable: After he came back

21
 Foundations of Family Therapy
home, he would become threatening and abusive; he would then be
moved out to an apartment; after that, he would get on drugs and
go out and cause trouble in the community. The police would find
him, the parents would hospitalize him, and the whole cycle would
start again.

Haley conceptualizes the therapeutic task as one of breaking up

one part is blocked, the whole will have to change. The
this cycle. If
therapist who was working on this case could use sign language, and
therefore could communicate both with the young man and with his
parents, who did not use sign language. The
first intervention he

made was change the type of institution the son would go to the
to
next time he was picked up by the police. He had the parents sign
a paper, in the presence of the son, saying that the next time he got
in trouble they would have the police put him in jail. This changed
the consequences not only for the son but for the parents. They
could still maintain some control over him if he were in a hospital
—visiting him frequently, for instance. In jail this would no longer
be possible. In addition, his actions were reframed from behavior
that could not be controlled to behavior he was held accountable
for.

Another change was to interrupt the cycle at the point when the
tension between the parents increased. When their son maintained
good behavior for a while, irritation between the parents would
begin to surface, and he would get into fights with his mother.
Mother and son were in a kind of sticky bond. She would protect
him from his father, who would beat him up if he misbehaved; on
the other hand, she would push him away from her, insisting that
he go out of the house and not make so many demands on her. To
some extent the mother could ignite an outburst from her son just
by this way of treating him.
The therapist, during a period when the deaf son was living at
home, prolonged the time of good behavior beyond its usual limits
by betting him a sum of money that he would not get into trouble
with the police during the next two weeks. The son stayed out of
trouble and won the bet. As a result of this period of calm, the father
went on a trip he had long been planning. During his absence the
mother angered her son by trying to make him spend time out of
the house. He became violent, threatening her with a knife and his

216
BREAKING THE SYMPTOMATIC CYCLE
sister with a baseball At the next family therapy session the
bat.

therapist showed up with a knife and a bat and laid them on the
floor. Although the patient acted as if he were not responsible for

what he had done to his mother and sister, the therapist questioned
him so intensely that he grabbed the bat and threatened the thera-
pist. The father took the bat away from him. The therapist then

insisted, and had the parents insist, that their son was not allowed
to use violent threats and weapons to intimidate people. He could
argue or criticize, but he could not use knives or bats. This move
blocked the patient's part in the sequence, labeling it as intolerable
behavior and making the other parties in the sequence join to resist

it.

The next shift was to have the son move into an apartment and
continue to live there. In the past, he had usually moved away only
when he and his parents were angry with each other, and no matter
where he lived, the parents treated him as a handicapped person
who could not be expected to take care of himself alone. The thera-
pist laid some groundwork by insisting that the son do his own
laundry while he was still at home, and contribute housekeeping
money from his welfare check. He was also expected to repay debts.
Thus the self-fulfilling prophecy that labeled him incompetent was
cut into in an active way. After the scene with the bat, the young
man moved out once more. He got into trouble with the police again
and was briefly hospitalized, but this time his parents stayed out of
it and he handled the situation on his own. According to the thera-

pist, this was the last time he got into trouble, and, more important,

the last time he was hospitalized.

In this chapter we have addressed ourselves to some of the ways

therapists of quite different persuasions block or alter a rigid pattern
or cycle that accompanies a problem and that seems to monitor the
possibilities of change. The next question we must deal with is a
tricky one. The most experienced therapists seem to recog-
fact that
nize and deal with the same type of configuration does not mean
that they base their therapeutic ideology or techniques on the same
premise, or that they would even agree with this analysis of their
work. Accordingly, we shall devote the next chapters to a more
detailed analysis of the work of the family therapy pioneers and to

217
 Foundations of Family Therapy
the emerging schools of therapy that have begun to dominate the
family field. We shall try to describe more fully the confusingly
different approaches of various therapists and schools, and to relate
their theories to their techniques.
In addition — and this is an even more serious concern —our model
has become small, constricted, and mechanistic. To speak of a trian-
gle of two parents and a child, or the recursive cycle that may be
played out in the sequence of behaviors between these persons, is

too reductive. We must go back to the booming, buzzing confusion

again, to shake up this position and bring in richness, depth, and
breadth. An examination of the work of clinicians who pioneered
the family and of those who built on the work of these early
field,

figures, will move

us toward variety and the possibility of envision-
ing new and more complex models in our examination of systems
and systems change.

218
Chapter 12

Family Therapy and

the Great Originals

The Family Movement Comes of Age

The family movement in therapy resembles the Protestant move-

ment in religion. It follows on the heels of a highly organized body
of ideas and practice which has a well-recognized founding father,
Sigmund Freud. Despite multiple heresies and schisms, psychoanal-
ysis has formed the basis for a mental health establishment. Some
pioneers in the family therapy field have mounted a revisionistic
assault, almost amounting to revolution, against the ideas of the
Freudian establishment, and this revolution has produced a host of
rival messiahs, gurus, and sects, all claiming primacy but none
finding legitimacy.
How, then, to comment on the differences and similarities among
the major approaches to family therapy which have developed in
the United States? Since Madanes and Haley have covered the large
territory of transpersonal therapies, 1 our next chapters will focus on
five major approaches within the field of family therapy: the histori-
cal, the ecological, the structural, the strategic, and the systemic

219
 Foundations of Family Therapy
(leaving out the approaches to family therapy which have been
claimed in the name of previously existing models such as behavior
therapy, Gestalt therapy, and other schools that arose indepen-
dently of the family therapy movement but might technically be
seen as interpersonal).
Before dealing with the established schools, however, we must
focus on a few pioneering figures whose contributions are of great
value and yet who fall into no schools. They can fairly be called the
Great Originals. Virginia Satir, the late Nathan Ackerman, the late
Don Jackson, the late Milton Erickson, and Carl Whitaker are
among those who cannot be categorized. In this chapter, we will
attempt to describe the approaches to therapy and the therapeutic
ideas of these inimitable souls.

Satir and the Family of Angels

Satir's place in the family movement is extraordinary and unique,

even though she has transcended the confines of family therapy to
join the wider scope of the Human Education Movement. In 1963,
Satir was conducting a family therapy demonstration project at the
Mental Research Institute in Palo Alto. I had been asked to help edit
her first book, Conjoint Family Therapy, and was amazed and enchanted
by the power of her presence with families. 2 Even more impressive
was the accuracy with which she seemed to discern the features of
7
that elusive creature she called a "dysfunctional family system/
When she would say, "I always judge therapy by the pronouns,"
she was referring to the tendency in such families for everybody to
say "we" rather than "I," a common attribute of "consensus-sensi-
tive" or "enmeshed" families. This explains her interest in exposing
"discrepancies" in communication; her emphasis on helping people
accept the "differentnesses" between them; her formulas for block-
ing repetitive sequences that end with one person taking some
standard role like victim, martyr, scapegoat, rescuer. She had pun-
gent phrases for such situations: "Did you ever see the cause of
death on a death certificate that you said 'no' to somebody?" Or she

220
FAMILY THERAPY AND THE GREAT ORIGINALS
might ask, of a family that resolutely refused to acknowledge that
anybody disapproved of anybody else, "Well, I go on the principle
that no humans are angels." Finally, she had an uncanny ability to
pull the label off an "identified patient."
There is a related tactic —more a stance, actually—for which Satir

has become justly famous: her ability to take the most negative
problem or situation and turn it into something positive. An exam-
ple is a first interview with the family of an adolescent boy, the son
of a local minister, who had impregnated two of his female class-
mates. The seating of the family vividly expressed their sense of
shame. The parents and siblings sat at one side of the room and the
boy, head down, sat in the opposite corner. He was a handsome
blond young man, and he had on the whitest and tightest pair of
jeans imaginable; if his manner seemed repentant, his dress and
body did not. Watching the interview, I felt that there was no way
for the youngster to get out of his extremely difficult spot, and no
way for the therapist to dislodge him. I underestimated Satir, who,
after introducing herself to the family, said to the boy, "Well, your
father has told me on the phone, and I just
a lot about the situation
want to say before we begin that we know one thing for sure: we
know you have good seed." The boy's head shot up, his back
straightened, and he looked with amazement at Satir as she turned
to the mother and asked in a bright, crisp voice, "Could you start
by telling us your perception?" Her strength seems to lie in her
ability to join with people not in terms of anger, blaming, hostility,
but in terms of disappointment, pain, and hope.
However, one were to classify her in terms of her work as a
if

family therapist, one would have to place her as a master of the art
of disentangling people from the mystifying communicational traps
that are a particular hallmark of families with a psychotic member.
The best demonstration of this aspect of her work that I know of,
outside of a live interview, is "A Family of Angels," a first session
with the family of a psychotic boy, to be found in Haley and
Hoffman's Techniques of Family Therapy. 3 The patient, seventeen, has
just had a psychotic break. Early in the interview, Satir asks family
members how they show disapproval of each other. She is told that
in their family nobody ever disagrees with anybody else. However,
the parents admit that in spite of what the mother terms "an amaz-

221
 Foundations of Family Therapy
ing relationship," she and her husband occasionally have cross feel-
ings toward each other. When asked how she shows such feelings
to her husband, the mother says that she would hurt him with her
silence. The father explains that he never gets angry but lets time
heal things over. Both agree that it is merely a temporary lack of
communication when this happens.
Satir comfortingly supports this idea and goes on to remark that
the patient, too, seems to feel that he cannot communicate with his
parents. By indirection, Satir implies that there may also be dis-
guised anger between the parents and their son. It turns out that the
boy's breakdown took place while the family was staying in a hotel
abroad and his parents had confined him to his room. Satir turns to
the parents and asks sharply, "Now why did you lock him up?" The
parents quickly reply that the door was not actually locked, but it

becomes clear that they had prevented the boy from leaving the
room.
The boy then explains that the reason he wanted to run away was
because he was afraid that his parents would lock him up in a
hospital. Satir asks the parents if they wanted to punish the boy.
They deny this vehemently, saying that their action was merely
protective. Satir remarks, "So this was a way you wanted to protect
him," and then, "But Gary [the son] didn't feel that way about it."
In this way, Satir validates the boy's experience of concealed anger
and punishment which is always going on between people in his
family, and now between his parents and himself.
Before the family can react to this sudden confrontation, Satir
switches to the past, finding out that the parents were first cousins
who had married against the protests of the father's mother. Never-
theless, this mother has lived nearby for twenty years and is a
constant visitor. The family portrays her as extremely domineering
and the mother says she can never say "no" to her. Satir asks the
mother what would happen if she told her mother-in-law "no." It
turns out that the mother has always been afraid to do anything to
push the mother-in-law away because her husband was so close to
his mother, even though the two women have carried on an under-
ground battle for years.
Here Gary is able to say, "And I have had a dilemma, which side
to pick. Should I take my mother's side or my grandmother's side?

222
FAMILY THERAPY AND THE GREAT ORIGINALS
And I've been waiting and waiting." By this time, the fact that there
are open wars, and sides to take, has been established. In the rela-
tionships between Gary and his parents, and the parents and the
mother-in-law, unmentionable subjects are at last coming to light.
In Satir's view, this clarification of communication is part of what
will free the psychotic from his position of understanding the bu-
ried messages and having to respond to them, yet having to deny
that he understands or responds to them or that they even exist.
Satir's major concern has always been with the individual and it

is probably this concern that stimulated her interest in the human

potential movement. In the past decade, she has shifted more and
more from her initial focus on families to working with huge groups
in a spellbinding, almost religious way. She has become a prophet
of love and joy in what is now billed as the "Satir Experience." But
before she entered that world fully, she left as a legacy to family
therapy not only her writing but many therapists whom she trained
and who have continued to extend her insights into families, and
her unique approach to working with them.

The Irreverent Artistry of Nathan Ackerman

The late Nathan Ackerman is the most important family figure to

come out of the psychiatric establishment of the Northeast. While
using psychodynamic formulations to describe his work, he never-
theless created an art of psychotherapy that crashed through every
known tradition. On the East Coast his name was synonymous with
family therapy for many 1930s with an
years. Starting in the late
article on the family as a bio-social-emotional unit, he pioneered

seeing families clinically during the 1950s. 4 But he did not merely
sit and talk with people in a family, transferring psychodynamic

techniques to individuals in the family setting. He worked with

families as a bullfighter works with were
a bull. His demonstrations
famous and an almost shocking
for their theatrical ebullience, wit,
intrusiveness into private areas of personal and family life.
Most of the work he did for public display was consultative, with

223
 Foundations of Family Therapy
other therapists carrying the case. Even the film In and Out of Psychosis,
which presents a long-term treatment case, consists of excerpts
from the only two sessions he had with the family, an initial inter-
view and an interview later in the course of treatment. Even so, an
analysis of any one of his interviews shows that he had an uncanny
nose for "the thing in the bushes" — the thing the family therapist
is out to change —and an extraordinary to use his own pres-
ability
ence to induce change.
Perhaps his best single analysis of a treatment case was contained
in his and Paul Franklin's article "Family Dynamics and the Revers-
ibility of Delusional Formation." 5 This paper includes part of the
filmed material that was the and Out of Psychosis." It
basis for "In
is the story of a girl who believed she lived on a planet called

"Queendom," peopled by creatures who were like females but re-

produced by fission; no men were allowed. Ackerman shows, both
in his description of family dynamics and by the treatment strategy,
how this delusion was actually an allegory for the skewed structure
of the family. The father appeared remote and fond of intellectual-
izing; the mother was bland and self-righteous. Both were domi-
nated by the rigid tyranny of the mother's mother, who lived with
them. Father was in effect divided from any real contact from
Mother by the powerful inhibitory effect of the grandmother's
presence. This Queen Mother, too, seemed to say: No Men
Allowed.
The parents nevertheless presented themselves as united and har-
monious, especially in their concern for their psychotic daughter,
whom they felt was a victim of a too vivid imagination. Only when
Ackerman began to challenge the rule of the grandmother, help the
father to assert his right to a real relationship with his wife, and
lessen the mother's dependency on the grandmother, was the
daughter's delusion broken. At the same time, the conflicts between
the parents came to the fore, and the grandmother was sent to live
with other relatives. These changes were the price of the daughter's
recovery, which took about a year of regular family therapy
sessions.
Ackerman had the foresight to film many of his interviews, and
he published some of them in book form. Examining these records
we have the opportunity to look closely at his actual performances.

224
FAMILY THERAPY AND THE GREAT ORIGINALS
Despite the psychodynamic phrasings, an analysis of the transac-
tions in evenone session convinces one that Ackerman was strug-
gling toward what was later to be known as the "structural" ap-
proach to family therapy, an approach that links symptoms to
dysfunctional family structures. no accident that the inventor
It is

of that school, Salvador Minuchin, was first introduced to family

therapy through Ackerman in the early 1960s, when Minuchin was
invited to join Ackerman in his work with adolescent boys. One
imagines that the imprint of the older man's work stayed with
Minuchin, who restlessly searched until he found a language, gram-
mar, and conceptual framework that would explain his own modifi-
cations and extensions of this radical therapeutic form.
In defense of the thesis that Ackerman worked in an essentially
structural way, let us examine one of his most irreverent interviews,
an excerpt that opens Treating the Troubled Family. 6 In this session

Ackerman plunges into the here-and-now, ignoring the family's

ostensible reason for coming to therapy in favor of tracking the
relationship sequences connected with the presenting symptom.
Tactics consist of blocking behaviors and sequences by a quip, a
quick movement of hand or posture, or an interpretation of nonver-
These are not only diagnostic but fall into the category of
bal cues.
what Minuchin would call "restructuring moves," pushing the
configuration of the body politic toward a more normal Ack-
state.

erman tends to take the stance of a benevolent if wary pugilist, and

is not afraid to lock horns with family members, although at other
times he will turn into a most seductive agent provocateur.

This segment is the first part of the second interview with a

family that came for treatment because of serious fighting between
the eleven-year-old daughter and the sixteen-year-old son. The girl

had recently threatened to stab her brother with a kitchen knife.

The son had a long history of temper outbursts and was doing
poorly at school. The father, in his forties, was a businessman, the
mother a teacher.
The father sighs as he sits down and Ackerman immediately asks
him why he is sighing, refusing to be put off by the father's insis-
tence that he is tired and suggesting that maybe he has some reason
to sigh. An attempt to get the son to comment on why Father is
sighing is interrupted by Mother, who breaks in to announce that

225
 Foundations of Family Therapy
she has been keeping a notebook of all family transgressions during
the week. She shows the notebook, explaining that she keeps anec-
dotal records of the children in the school where she teaches, and
decided to do the same with her family. She is a sharply assertive
woman in her forties, in contrast to the father, who seems mild and
complacent. The therapist reacts with an ironically amused com-
ment: "You come armed with a notebook/' and extends the meta-
phor by saying, "Fire away!"
Now follows a maneuver that undercuts the mother's authority
while never directly challenging it. In spite of his invitation to
Mother Ackerman picks up on the nonverbal behaviors
to proceed,
of other family members as they react to Mother's threatened ex-
pose. He turns to the father and says, "You're picking your fingers."
This diverts Mother, who cannot resist commenting on Father's
many nervous habits. The son comes in to defend his father and
challenges her about her own habits. All family members start to
talk at once, but the therapist clears the way for the son, who tells

the true story about Mother: She belches! Mother calmly confirms
this habit, and admits that the person in whose face she belches
most is Father's. Now when she tries to go back to the notebook,
the therapist continues to ask about the belching. The mother gig-
gles; it's a habit she doesn't think should be taken seriously. She
suggests that her notebook contains items about the marital rela-
tionship that will show her husband in an unfavorable light. The
therapist is deaf to her effort to unloose these damaging facts and
keeps focused on the belching, which the mother says mostly occurs
when she isdown. Father says that her habit of belching in
lying
his face upsets him a great deal.
At this moment, when it looks as if there may be some tense
issues coming out between the parents, the children start a rescue
operation. The daughter taunts her brother about a lipstick mark on
his neck, indicating that he has been intimate with his girl friend.
The son gets furious. The therapist comments that this interruption
has occurred just when the parents were going to talk about their
love life. Father, who sounds as if he is beginning to enjoy this

session, describes how it feels when you want to kiss a woman and
she belches in your face; you need a gas mask. The daughter again
tries to intrude, only to have the therapist ask her to move her seat

226
FAMILY THERAPY AND THE GREAT ORIGINALS
so that Mom
and Dad can talk together. He points out that the
children really do know about their parents' marital situation, even
though the son states his unqualified disgust at the topic. He wants
to leave. The therapist draws a generation line by remarking that the
boy may be afraid that if he intrudes on his parents' love life, his
parents and the therapist may invade his. He tells the boy that he
would like him to stay, but that he can leave if things get too
intolerable. The daughter, in a coy, teasing voice, says she is scared
her brother will want to kill her after they go home for telling about
his girl friend. Ackerman responds to this third rescue attempt by
labeling the girl's preoccupation with the girl friend as jealousy,
putting a little circle around the siblings. The son gets angry and
leaves. The daughter soon follows. The children are clearly mirror-
ing their parents' quarrels, while at the same time acting in a diver-
sionary manner.
All this is an adroit way of restructuring the family to induce
change. Ackerman has quickly disarmed the mother, who came in
ready to blast her husband on the subject of his sexual inadequacies,
and has exposed an area where the father could be the accuser. Very
quickly Father is up, Mother is down. We note that even though
Ackerman is not always decorous or polite, the mother does not get
angry; it is the children who get upset and try to fight Ackerman
off by inserting the symptom (their fights). Ackerman reframes this
intrusion as an example of a lovers' quarrel —
exactly as he will later
do with a quarrel between husband and wife. The children are
pushed out of their parents' bed, as Ackerman puts it, and placed
on their own side of the generation line. Their reaction is to leave,
which they presumably would not do if their anxiety about their
parents were not somewhat relieved.
Once they have gone, the therapist gets into the marriage bed
himself, listening to all the lurid details of the wife's disappointed
sex life. The wife complains that her husband is not romantic and
attentive; in particular, she resents having to be the one who attends
to contraception by using a diaphragm. Throughout, Ackerman uses
a style of blunt, playful vulgarity, which detoxifies issues by push-
ing them to the edge of absurdity. The belching is blown up to such
heroic proportions that the wife starts to laugh. Her husband's
rather nervous attempt to respect her is brushed aside by the thera-

227
 Foundations of Family Therapy
pist, and he encouraged to be bold and breezy himself. Ackerman
is

teases the wife away from her imperious, commanding position by

going back to her early married days when she was "green and
innocent" and her husband was a sexy man of the world. In recent
years, as Ackerman finds out, the husband has lost interest in sex.
Is that because his wife is not a good "piece"? Is she a "dog"? The
wife is enjoying the exchange, in spite of (or perhaps because of)
the impolite language.
Ackerman then with the wife, asking her, "How
starts a flirtation
can you cooperate with me so beautifully, so juicy-like and with . . .

[your husband] you don't cooperate?" He tells her what a pleasing

woman she is to him. At the same time he has been challenging the
husband to make demands on her, get her off her "fat ass." Un-
believably, this grim couple is beginning to laugh, to appreciate each
other,and the positions they came in with are far more balanced.
The wife seems softer, the husband firmer. In fact, they have
achieved something like parity in a remarkably short time. The
notebook has long since been forgotten. The children return and
arrangements are made for the next meeting.
This analysis gives an idea of Ackerman's provocative style, but
also indicates an approach to family therapy that can be thought of
as essentially politicaland organizational. Although Ackerman is
not as explicit on Minuchin, he is clearly revising the
this point as
configuration of relationships in the room and pushing them toward
a more "normative" state.

Whitaker and the Therapy of the Absurd

An equally provocative therapist is Carl Whitaker, whose work

seems calculated to shock, amaze, enchant, and confuse. Whitaker,
who calls himself a therapist of the absurd, specializes in pushing
the unthinkable to the edge of the unimaginable. He will suddenly
suggest to a young "psychotic" woman that she sit on her father-in-
law's lap, announcing, "Incest is better than love." If asked about
his reasons for such statements, he is likely to answer, "To please

228
FAMILY THERAPY AND THE GREAT ORIGINALS
myself. If I don't get something out of therapy for myself, I know
it's not going anywhere."
Whitaker uses many techniques that, as he puts it, push over the
leaning tower of Pisa. In one interview Whitaker turned to an ex-
pressionless young man who had recently made a serious suicide
attempt and said, in the presence of his family and his slightly upset
therapist: "Next time you try that you should go first class take —
someone with you, like your therapist." For the first time during the
interview, the boy's eyes snap awake. He continues to watch Whi-
taker, who casually begins to describe a new invention he has been
working on which will take the messiness out of suicide —a human
Dispos-all, "like that thing in the sink only bigger." By the end of
the session he has not only reached the boy but activated the boy's
father, the other one in the family who has given up on life and is
in despair.Whitaker says, "My tactic has become a kind of tongue-
in-cheek put-on, an induced chaos now called a positive feedback
— that is, we augment the pathology until the symptoms self-
destruct."
One of Whitaker's techniques is to spread the problem around:

We She gets one point for that. Now

started with Mother's alcoholism.
we've uncovered Dad's gruesome collecting disease, so he gets two points,
John's revealed his school phobia and bared Henry's delinquency. Mary,
do you plan to destroy yourself by being the family heroine and the nurse
to every patient in the entire hospital? . . .

Jim, if Mary tried to kill herself because she sensed you'd like her dead,
do you suspect that if the gang of you would help her, she might convert
to killing you? 7

If all Whitaker will prescribe an impasse in such a way

else fails,
that the family will be very reluctant to let him win. In one case he
came into an interview as a consultant and finally said,

Well,it's been a dismal hour. Your therapist has worked hard to help you

tochange but to no avail. He even asked for this consultation, to see if he

was too blind or too feeble and failed you in that way. I'm convinced that
I can't help either. It looks hopeless. I guess it'll just stay this way for ten
more years. Maybe this is the best you can do and that's all right even
though it is discouraging. I doubt if it will get worse and that's some
consolation. 8

229
 Foundations of Family Therapy
about therapy, Whitaker focuses a great deal on the
In writing
area that comes under the heading of taking control. Whitaker
counts on achieving some of his most powerful effects from what
he calls the backlash of the encounter. He very effectively uses
indifference, gentle ridicule, boredom, even urging a family not to
come into treatment, as a "negative come-on." Whitaker feels that
he must win at establishing the rules of the game that will be played
in therapy, and he will not even admit that therapy has begun until
this stage is reached.
He is famous for his colorful tactics in achieving these
justly
victories. When one couple called for an appointment, each was
having an affair, but they wished to discuss the future of their
marriage. They objected to seeing Whitaker with a cotherapist be-
cause of the double price. Whitaker gave the couple a choice: he and
his cotherapist would consent to see the couple: (1) under a contract
specifying no sexual relationships during therapy, "to potentiate
the affective relationships with the two therapists"; or else (2) the
two therapists would see the couple together with their respective
lovers and the spouses of the lovers, in a group of six. The couple
declined both offers, but called six months later to say that they had
gone back to being a pair. 9
The delicacy of Whitaker's dance through an interview has to be
seen to be appreciated. In a videotaped consultation Whitaker did
for a Philadelphia Child Guidance Clinic family, the symptomatic
child was a boy with a soiling problem who came in with his mother
and father and a baby brother. Whitaker pushed his way into the
good graces of the family by sitting down on the floor with the baby
and playing with him. In one of the nicest recorded examples of
nonverbal reframing of symptomatic behavior, Whitaker noticed
that the baby was straining and grunting while having a bowel
movement, and he strained and grunted along with him.
Whitaker's manner with the family was equally casual. This
turned out to be a family in which the men sat on their anger.
Neither the father nor the boy found it easy to stand up to the
mother, who was something of a drill sergeant, chunky and brisk,
and quite annoyed with her problem son. She launched into a loud
diatribe about him at one point, and Whitaker, who was sitting at
the other side of the room, got up and sat next to the boy. Address-

230
FAMILY THERAPY AND THE GREAT ORIGINALS
ing the mother, he began to compliment her on the "fire in her
eyes/' As a result of this move, her manner immediately softened.
The fatherwas sitting next to her like a large, plump Buddha. He
held the baby part of the time and fed him his bottle. He was as
placid as she was sharp. Whitaker reframed the marital skew by
saying that it was lucky a fireball like Mother hadn't married an-
other fireball, otherwise there would be continual explosions. He
declared that he and his own wife shared the same lucky situation,
since she was also a fireball and he was sort of a blob. He added that
two blobs shouldn't marry either.
Operations of this sort were constructed so deftly that most of
them did not reach the family's awareness. By the end of the inter-
view, however, the boy seemed noticeably less depressed than
when he came in, and the mother had responded to Whitaker by
becoming milder and the husband by livening up. The family's
reaction to the interview was very positive, although they remem-
bered very little about it afterward except "that nice doctor." One
could assume that this reaction was a natural response not only to
Whitaker's appeal but to a more normative structuring of family
relationships.
The interview was notable in that Whitaker did not call up his
big guns; the familywas easily persuaded to move with him and
"therapy of the absurd" was not needed unless one takes the —
position that a professional man of mature years sitting on the floor
and grunting along with a defecating baby is absurd.
A more intensive, session-by-session account of ongoing coth-
erapy with Whitaker is given in Napier's The Family Crucible, one of
the best introductions to the inside experience of being a family
therapist. 10
Whitaker has said that he must be "captured" by the
To this end he plays hard to get, setting
patient for therapy to work.
up hurdles and obstacles to the therapeutic relationship like a
spoiled and haughty courtesan. Implied in Whitaker's therapy is a
Zen-like theory of change. As he says, "Psychotherapy of the ab-
surd can be a deliberate effort to break the old patterns of thought
and behavior. At one point, we called this tactic the creation of
"n
'process koans.' It would be hard to think of a better term.

231
 Foundations of Family Therapy

The Uncommon Therapy of Milton Erickson

Although Erickson is known primarily as an experimental hypno-

tist, it is due in part to Haley's great interest in his work that he has

been included as one of the grandfathers of family therapy. Readers

will find Haley's account of Erickson's therapeutic miracles in Un-
common Therapy both fascinating and mystifying. 12
Perhaps the most outstanding feature of his art is in the broad
category of "encouraging the resistance/' This common hypnotic
technique has become the basis for the development of the paradox-
ical directive, a hallmark of the strategic school. Historically, then,
itmakes sense to emphasize this aspect of Erickson's work, even
though it is by no means representative of the whole of it.
A vivid example of Erickson's way of encouraging a symptom
while subtly introducing change occurs in the case of a young man
about to enter military service. 13 His problem was that he could
urinate only through an eight- or ten-inch metal or wooden tube.
Erickson induced a trance and suggested that he get a twelve-inch
tube of bamboo and substitute it for his other equipment. He was
to hold the bamboo with his thumb and forefinger and his other
three fingers flexed around his penis, alternating his right hand with
his left. At the same was to try to feel the passing of the
time, he
urine through the tube. He was also told to mark the tube off in
quarter inches, and the suggestion was made that he might begin to
consider shortening the tube but that he must not feel compelled to
do so but let the decision happen of itself; he was to concentrate
instead on what day of the week he might choose to shorten it.
Finally, he was told that his army physical examination would be
postponed but that he would be called up for a second examination
in about three weeks and would probably be accepted.
The young man did get a bamboo tube and used it as instructed.
After a week, he decided to saw an inch off, then (for some reason
on a Thursday) two inches, and so on, until he had a ring of bamboo
only a quarter of an inch wide. Soon afterward he realized that his
three fingers around his penis constituted a tube, and he discarded
the bamboo ring. Erickson writes, "He [held the tube] with both the

232
FAMILY THERAPY AND THE GREAT ORIGINALS
right hand and left hand and even experimented by extending the
little finger and then realized that he could urinate freely without
14
resorting to any special measure." The tube had become the penis
itself.

This is as nice an intervention as one can find, even in Ericksonian

annals. The patient was not asked to change, instead he was directed
to continue to urinate through a tube. But the idea of change was
most certainly introduced: the tube could be longer, but it could
then also be shorter, and it could also be of a different material. The
wording of the suggestions given during the trance insured that the
young man would "of himself" gradually abandon the artificial
tube. And the idea of a successful outcome was contained in the
message that his next examination would probably be successful.
The patient reported with some bewilderment at what he found
himself doing (he had been told to have amnesia for the trance,
although he remembered some of the suggestions), but began to
have hope that he would begin to break the habit he had
a sense of
been carrying for so long. Although the habit seemed to have origi-
nated from a humiliating experience when, as a small boy, he was
caught urinating through a knothole in a fence, there was no at-
tempt to "work through" this initial incident or to use it as part of
the treatment. The charm of the story, of course, is that Erickson did
not cure the young man of urinating through a tube at all; he simply
helped him to substitute a natural tube for the artificial one he had
been using.
Reading. accounts of Erickson's work, one is struck by how little

attention he pays not only to the past but to the relationship context
of the problem. In the cure of an enuretic young man also about to
go into the army, the anxiety about having a wet bed was trans-
ferred, using hypnotic suggestion, to "pseudoanxieties." 15 For in-
stance, Erickson directed him to stay three days in a hotel in a
strange city and, while remembering how distressed he would be
when mother found a wet bed, to suddenly think that it would
his
be an amazing joke on him if, after all his agonizing, the maid in the
hotel should discover a dry bed. What shame and embarrassment he
would feel when the maid found the dry bed. In addition, he was
told that on checking out of the hotel he would find himself in a
conflict as to which set of grandparents to visit first, since he was

233
 Foundations of Family Therapy
making good-bye visits. One consequence of his enuresis had been
that he could never visit relatives. Although one might guess that
the enuresis protected him from any conflict of loyalty regarding
sides of the family, Erickson never mentions this possible aspect of
hissymptom. He dealt with it by suggesting that the
indirectly
young man would resolve his obsession with the problem of which
relatives to visit first by making the visit to the first set shorter than
the visit to the second, and that, once he got there, his visits would
be pleasant and comfortable. The young man experienced all the
obsessional thoughts suggested, and found himself shamed by hav-
ing a dry bed in the hotel. He also reported feeling "crazy" and
confused about which relatives to visit, but that the visits had gone
well.
Erickson describes his goal as trying to substitute other anxieties
for the enuresis while still addressing the central anxiety: the young
man's shame at his mother's finding a wet bed. The tie with the
mother was not explored in any way; Erickson gives it a minor place
in the story by saying that when he was accepted by his draft board,
his only worry was his concern about his mother's adjustment to his
going into the army.
Unfortunately, the immense interest in Erickson has focused
mostly on his inimitable talent. One can read through Haley's book
on Erickson, or peruse Erickson's own articles, and appreciate the
amazing ideas and the incredible outcomes but be no wiser as to
how one might replicate such work. Richard Bandler and John
Grinder's more recent microanalysis of Erickson's sessions with cli-
ents (speech patterns, use of tonality, pauses, metaphors, and so on)
16
is no more helpful. Here, too, one studies the ingenious subtleties
laid out for the neophyte and feels that these things are simply not
replicable. The art of therapy is still the art of the shaman, the high
priest, the medicine man. For all the fascination with Erickson's

work, and for all the brilliant studies that have been made of it, it
remains something that only an extraordinary individual can
achieve if initiated into the mysteries by an extraordinary teacher.
Another unfortunate by-product is that none of these analyses of
Erickson's work lead to a further understanding of "what is to be
changed," but only to a refinement of the art of persuasion. The unit
of treatment is narrowly defined as "therapist plus problem." Thus

234
FAMILY THERAPY AND THE GREAT ORIGINALS
schools of therapy associated with what eventually emerged as the

strategicapproach deal with theories of persuasion or behavior

change, but do not bother overmuch with the shape or pattern of
that which is to be changed. With an Ericksonian therapist there is
no such thing as a problem, only something defined by somebody
as a problem. Change the definition, the perception that "creates"

the problem to a different one, and the problem will no longer exist.
So we are back to dazzling sleight of hand, elegant magicianship,
and the triumph of the arcane.

Jackson and the Therapeutic Double Bind

Don Jackson, Mental Research Institute,

like his colleagues at the

was interested in a hitherto undescribable phenomenon: the recur-

sively shifting, yet interlocking, sequences of behavior in families
which correlated somehow with a symptom. He sensed that altering
one element in the pattern might alter others and, he hoped, the
symptom as well. Thus he was moving in a holistic, or what we
might call now a systemic, direction. There are few examples of
Jackson's work during an entire family interview, but the chapter
"The and Hoffman's Techniques of Family
Eternal Triangle" in Haley
17
Therapy shows him is critiqued by Jackson himself.
at his best and
In addition, Jackson shared the Bateson group's fascination with the
double bind and wondered whether there might be such a thing as
a homeopathic use of the double bind or a "therapeutic double
bind."
An example of what Jackson thought of as a double bind is
described in a study by Jackson and Weakland of the case of an
adolescent boy whose parents were afraid he might be homosex-
ual. 18 The young man tried to assert his manliness by staying out
late one evening with a group of friends and was greeted on his
return with a marked display of worry by his mother. As a result,
he began to come home early. His mother then threw out a new
worry: that he was not popular enough. Although uncertain which
way to go, the boy decided the next night to come home early, only

235
 Foundations of Family Therapy
to find that the mother had left him a note telling him that she was
out and would be home late. The authors say, "Here was an oppor-
tunity to break the bind they were in, but it is in the nature of these
reverberating cyclic sequences that he could not pull out." So he
started dinner for his errant mother, and when she came home he
covertly agreed to hide from his father the fact that he had cooked
the meal, with homosexual implications. As the authors note:
its

"And so the cycle was perpetuated." In other words the solution to

the original "simple bind" was punished, not rewarded.
Jackson's accounts of working with families with schizophrenics
emphasize constructing a therapeutic double bind as a major strat-
egy for change. The term "prescribing the symptom" had first been
used by the Bateson group, and Jackson was particularly ingenious
in experimenting with he was probably in-
this format. In this,
fluenced by one of his teachers, the psychoanalyst Frieda Fromm-
Reichman. In "Toward a Theory of Schizophrenia" an anecdote is
told about Dr. Fromm-Reichman's treatment of a young woman
who was in contact with powerful gods. 19 The girl told Fromm-
Reichman that God R would not let her talk to her doctor. Fromm-
Reichman replied that she herself did not believe in this god, but
that since the patient did, the patient must ask his permission to talk
to her doctor. After all, the patient had lived in the god's kingdom
for nine years and that had not helped her. So now the patient must
ask the god to let Fromm-Reichman try. This could be explained as
a therapeutic double bind, because if the patient became doubtful
about her belief in this god, she would be giving power to her
doctor. But if she went to her god with the request that since he had
failed, Fromm-Reichman must be given a chance, she again gave

power to her doctor.

In Pragmatics of Human Communication, Watzlawick, Beavin, and
Jackson discuss this type of therapeutic double bind in fuller de-
20
tail. Most of their examples are directed toward individuals or
couples and not toward larger or more complex family contexts. In
an example similar to the one described above, Jackson is shown
teaching a paranoid patient to be more suspicious. The patient was
refusing to respond to Jackson at all. Jackson said that the patient
must think he was God, and might really be God, and if he really
wanted to be treated like God, the therapist would comply. Jackson

236
FAMILY THERAPY AND THE GREAT ORIGINALS
then got down on and presented the patient with a large
his knees
hospital key, saying that since he was God he would have no need
of a key, but if he were indeed God, he deserved to have the key
more than the therapist. The patient dropped his stony demeanor,
came over to Jackson, and said, "Man, one of us is certainly crazy."
In another example, a woman with intractable headaches was told
that since her headaches were obviously not amenable to cure, the
therapist could only help her to live with her condition. The
woman, possibly to prove the therapist wrong, came back reporting
her headaches had abated. In another case, a young woman was
unable to get up in the morning to attend her college classes and was
in danger of flunking out. She was told she must set her alarm for
the proper time for getting up, but then must rest in bed until
eleven. After several days of this regime, which she found unbeara-
bly boring, she began to get up on time, and then was able to discuss
more openly some of the reasons why she was so afraid of going to
class.

Two examples in Pragmatics extend therapeutic bind interventions

to problems with couples rather than individuals. In one example
the couple consists of a compassionate caretaker wife and an alco-
holic husband. The wife is instructed to drink along with her hus-
band but keep one drink ahead of him. This reverses her usual role
of caretaker because she will become drunker than he and he will
probably end up being the caretaker. To avoid this outcome, he will
either have to drink much less or not at all. In any case, the pattern
is disrupted.
The second example is the case of a couple who argued con-
stantly. They are told that their arguing is a sign of emotional
involvement and that this apparent discord only proves how much
they love each other. As the authors say,

No matter how ridiculous the couple may consider this interpretation —or
precisely because it is so ridiculous to them —theyabout to prove
will set
to the therapist how wrong he is. This can best be done by stopping their
arguing, just to show that they are not in love. But the moment they stop
arguing, they find that they are getting along much better. 21

The authors of Pragmatics describe how therapeutic double binds

work by pointing out that in a pathogenic double bind the patient

237
 Foundations of Family Therapy
is "damned he does and damned if he doesn't. " In a therapeutic
if

double bind, since he is told not to change in a context where he

has come expecting to be helped to change, he is in a similar trap.
If he resists the injunction, he changes;
he does not change, he is
if

"choosing" not to change. Since a symptom is something which, by

7
definition, one "can't help/ he is then no longer behaving symp-
tomatically. Thus he is "changed if he does and changed if he
doesn't."
It turns out that nearly all of these illustrations refer to simple
moves with one problem and the unit is in only one case larger than
two. But in a few of Jackson's clinically oriented articles, one finds
far more attention paid to the matrix of the problem, as if it were
a hub with spokes that touched many points of a larger circumfer-
ence. One also finds a sense of therapy moving in stages, so that the
complexity pertains to time as well as to the immediate configura-
tion.

The question, posed by the therapist to the family, "What would

be the negative consequence of change?" first emerges in the MRI
22
literature in an article by Jackson and Yalom. This question chal-
lenges the family to prove that they can, despite the therapist's
doubts, live happily ever after without the presenting complaint.
However, it sometimes also happens that the predicted "negative
consequences" appear as soon as the presenting complaint clears up,
and this not only can upset the family but may unnerve the thera-
pist.Problems vanish, only to be replaced by others. As Jackson says
ruefully of one such family he treated, "I felt like a paper hanger
with bad glue, running around, never organized the way I would
23
have liked to be."
The mentioned above not only illustrates the idea of the
article

negative consequences of change but gives some idea of the multi-

ple dimensions that characterized Jackson's clinical thought and
work. Dave, twenty-four years old, had been diagnosed schizophre-
nic about five years before. Since that time he had been in and out
of hospitals, returning to live at home and take jobs, only to collapse
and be hospitalized again. The family sought therapy mainly be-
cause individual and group approaches had not worked, and when
the family therapist came on the scene, they resisted family treat-
ment as well. At that time Dave had been continuously hospitalized

238
FAMILY THERAPY AND THE GREAT ORIGINALS
for one and a half years. His family consisted of his mother, father,
and a "well" brother, who was a very polite, controlled young man
of eighteen. Eighteen sessions in family therapy had not cracked the
facade the family presented of three very caring people whose only
problem in life was Dave, their crazy son. The parents, was noted, it

seemed to behave like a single person, so tight was the bond be-
tween them. Attempts to gain a portrait of the family only elicited
assertions of "happiness, cooperation, love, and inexorable financial
success." At one point, the authors note, Dave pounded on the table
and shouted, "My God, I come from a perfect family." Mother
asked, "Dear, have we said anything that wasn't true?" Dave, per-
ceptive even in defeat, replied, "No, but now I see what a goof-ball
I really must be."
Jackson decided to counter the family's position by placing the
question to them: What problems might arise for this family if Dave
improved? This is a therapeutic double bind, or, as the Bateson
group also called it, a therapuetic paradox:

It is a paradox in which the family as presently organized "can't win." The

question encourages problems in a framework of help . . . the consultant
takes advantage of the family members' view of themselves as helpful
individuals and implies they would be uncooperative if they did not pro-
vide some difficulties to discuss with him. Backed as it is by an expert, the
question is heavily weighted to evoke at least token answers. Yet any
indication of family difficulties which might be caused by Dave's remission
can be amplified as barriers to his recovery and, hopefully, force the family
to consider, at some level, that they must change before recovery is

possible. 24

The family — like most families placed in this position —reacted

with incredulity to the question. The father, however, finally admit-
ted that if his son got well enough to come home, he would be
embarrassed by some of his son's public behaviors. The mother was
openly upset by the father's failure of support, and the first sign of
a rift in the marriage appeared. In the ensuing conversation Dave
helpfully suggested that he might fall in love and want to get
married. It would be problem for him to introduce the girl of his
a
choice to his parents. His mother countered by saying she would be
delighted, but then qualified her statement by saying, "Of course,
I would always hope it would be the right one." Mainly, the parents

239
 Foundations of Family Therapy
said, they agreed with Dave's many doctors that he ought to become
independent, in which case all their problems would be gone.
Dave then came in with another helpful remark: He might be-
come so independent that he wouldn't even want to see them, and
as he began to leave home, "I see associations with my family more
7
or less going down the drain/ To which Mother could not resist a
little rejoinder: "Well, other people's don't." Then Dave brought up

still another "problem": If he got more successful than his father,

how would his father feel about it? Although the father said, "I'd
be thrilled," the nervous laughter that erupted spontaneously in-
dicated some discomfort.
At this point the consultant asked the brother how he felt about
the patient's weekends with the family, after which Dave usually
became quite agitated and The brother admitted to being
upset.
nervous because he never knew what mood Dave would be in or
how to handle him. The consultant pointed out that it was as if
Dave were being asked to bear the whole burden of the family's
solicitousness. Dave thereupon burst out with a quite amazing anal-
ysis of his position in the family, and reframed his own symptom
positively within the total family context: "Well, it's just the story
that I'm the sick one in the family and so this gives everybody else
a ... a chance to be a good Joe and pick up Dave's spirits whether
Dave's spirits are down or not." 25
Here the consultant decided to use a second tactic to shift the
family system, this time an injunction to another member to change.
He asked Charles, the "good" son, if he could help his brother by
becoming more of a problem during the time Dave was away from
home. Charles asked, "You mean rebel against my parents?" and the
therapist said, "I mean you would be more of a problem in the sense
that you would get a little more honest about some of the things
that trouble you, or some of the uncertainties you may have, or
whatever you don't share with your parents now because you don't
want to bug them." Charles, being a good son, agreed.
This apparently innocuous intervention caused havoc. Father
began the next meeting with the announcement that he would like
to be the problem for a change. When asked how he might go about
doing this, he replied that he might come home late from work some
night without telling his wife. It turned out that a covert rule in the

240
FAMILY THERAPY AND THE GREAT ORIGINALS
family was that everyone had to "sign out" with mother, who
would become worried and upset unless she knew exactly where
everybody was at all times. In the following session the mother
came in very depressed and astonished the family with an account
of her own history: her mother, now dead, had been a severe asth-
matic who became addicted to narcotics; the daughter had feared
that her father would remarry and bring in a stepmother who would
not love her; later, the mother had known that her first husband was
unfaithful to her; and so on and so on. Father continued to bring
up things that worried him: his lack of confidence in himself and
feelings of failure as a parent. He told Dave, "If you think I'm
superhuman, that I don't have feelings or hurts or problems, that
nothing inside bothers me, then I've got news for you."
This news apparently made an impression on Dave, for after this
session he went out and got the first job he had found for himself
in years.Although he lost it because he lacked the appropriate skills,
his family was supportive and he got another at which he did well.
Gradually he moved to disengage himself from the family, and with
their encouragement he got an apartment for himself, continued to
get better jobs and support himself, and a year and a half later was
completely on his own. Although therapy did not stop then, an
important step had been made, and in subsequent sessions Dave
found better and different means of handling the many ways in
which his family and he were continually binding and rebinding
each other.
After Jackson's death in 1968, his colleagues at the Mental Re-
search Institute (mainly John Weakland, Paul Watzlawick and
Richard Fisch) continued to work with and expand these ideas, both
in terms of theory and clinical practice. In the next chapters we will
move on to discuss the establishment of schools, starting with the
approaches that seem most congruent with the major positions of
psychodynamic theory, though by no means identical. I have called
these approaches "historical," because of their emphasis on the past
and their multigenerational view of family pathology and theory of
change.

241
Chapter 13

Historically Oriented
Approaches to
Family Therapy

The Lingering Influence of Psychodynamic Thought

Although the family therapists we will discuss in this section are

not, in any formal or literal sense, associated with psychoanalytic
theory or practice, there is one group of therapeutic approaches that
seems to be more acceptable to the psychodynamic establishment
than others. Even when the unit worked with is different, the theory
of change (abreaction of repressed material) may be very similar; or
the goals (to achieve an individuated self) may be similar; or the
techniques (exploring feelings, gaining insight, "working through"
past events) may be similar. We might say that the central strand
linking these approaches is that they aim at disentangling the indi-
vidual from the family web (or disentangling all family members),
and therefore the individual patient is still the major focus of
concern.
The other approaches, which we have called ecological, struc-

242
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY

tural, strategic, and systemic, do not take the final blossoming of the

individual as their objective, but move to change the context of

whatever problem is presented (in most cases the family, or the
family plus other systems), in the expectation that once this is
achieved, individuals will blossom on their own. Let us start with
the practitioner who, in many respects, has focused most on
clearly

the individual within the multigenerational family context: Murray

Bowen.

Bowen and the Differentiated Self

The more psychodynamically oriented family therapists believe

that one must get at historical or causal factors to relieve a symptom
or achieve change. Family therapy versions of the psychoanalytic
concepts of insight, catharsis, and abreaction seem to be the major
avenues of change, and a mature objectivity is, as with most
Freudian therapies, the desired end result.

Bowen's approach is the most influential of these more historical

therapies. 1 Since by his own say-so he does not emphasize symp-
toms or problems, he may fairly be placed in the larger movement
that includes individual-oriented "growth therapies." His method
of coaching family members to go back to their families of origin
offers a path for achieving personal individuation and autonomy,
albeit via a family route. Many followers have found that his ap-
proach does relieve symptoms and problems, and his theory of the
multigenerational transmission of emotional illness has laid the
conceptual groundwork for an important school of family therapy.
In explaining the emergence of emotional illness in a family mem-
ber, Bowen suggests that it has its origin in the difficulty previous
family members have had from the core family. This
in separating
difficulty is mitigated, if not solved, by involving (or as he calls it,
"triangling in") a person from the next generation. As this process
unfolds from generation to generation, family members' inability to
individuate intensifies until one or more children exhibit the ex-
treme case of undifferentiation known as "symbiosis," which keeps

243
 Foundations of Family Therapy
them stuck forever and the family stuck forever
in the family,
around them. It is a kind of repetition compulsion applied to the
generations, except that each generation pushes more of its trouble
onto the next.
Bowenian family therapy is designed to identify the patterns
originating in the past that have such a hold on people in the
present, and to help people unlock themselves. Thus he emphasizes
searching out clues from living members of the extended family,
especiallyfrom older generations, to trace a pattern, and if possible
alter it. To do
this, he uses the genogram, a visual diagram of the

family tree going back in time and extending collaterally, with an

individual or a couple as the focal point.
Bowen's theory of change resembles Freud's dictum: Where id
—
was, there ego shall be if one substitutes the darkly primitive
condition of "fusion" (which Bowen equates with being ruled by
one's emotions) for "id," and the objective condition of "differen-
tiation" (defined as an ability to remain detached yet connected
within one's own family) for "ego."
Critics say that this makes therapy too intellectual a process. But
people who have been coached to revisit or reconnect themselves
with family members, to change some aspect of their interaction
with those persons, often report an enormous impact, both on their
own lives and on the lives of other members in the family. If the
family is, as Bowen huge interconnected web, the repercus-
feels, a

sions in one corner may be felt in another corner far away, if not
across the entire web. Even remembrances of relationships with
parents or relatives now dead can be influenced or changed, at least
in the mind of the searcher, with helpful results.
The model for this procedure was first presented in a paper
Bowen read during a symposium in 1967; it was later revised and
published in a book based on that symposium. 2 In this paper Bowen
describes his own family of origin, a large extended kin group going
back many generations which dominated a small southern town.
Bowen tells of deliberately intruding into most of the dominant
triangles in the immediate family by means of an astonishing strat-
egy. He sent off letters that told various relatives of the unpleasant
gossip that others were circulating about them. These letters ended
with endearing salutations such as "Your Meddlesome Brother" or

244
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY
"Your Strategic Son/' and announced an impending visit. Bowen
then showed up as heralded, to deal with the predictably indignant
reactions of his kin.
The on the family was dramatic. It loosened up many
effect
—
closed-off relationships and once the original fury against Bowen
—
himself had died down it created a climate of better feelings all
round. The effect on the symposium was equally dramatic. Bowen
had prepared a theoretical talk, scrapped it at the last moment, and
instead gave a blow-by-blow account of his incredible journey the —
first time, in the experience of most of his audience, that a practi-

tioner had ever attempted to change and influence his own family
or described such a literally "convention-shattering" procedure to
an august body of colleagues.
This experiment of Bowen's development of
set the stage for the

an entirely new therapy process. The commonest form of Bowenian

therapy is the quest for the differentiation of self. A trainee is not
considered fully developed until he has been "coached" in differen-
tiating himself from his own family of origin —a process that Bowen
says can take twelve years. There is a resemblance here to the
mandatory training analysis that a neophyte psychoanalyst has to
undergo before being given full responsibility for patients. And the
goal —to produce a person who is free of crippling entanglements
with family relationships, past and present, and can therefore get on
in a more unfettered way with his own life— is certainly close to the
goals of psychoanalysis.
A side benefit of Bowen's training process is that it allows a
family therapy induction to take place in a facility where no clinical
families are available. The trainee can use his or her own family to
practice on, or else the group of trainees role-plays one trainee's
family, taking the various parts of family members and being di-
rected by either the trainee or the teacher. A weakness may be that
the trainee learns only how to coach similar individuals in achieving
a "differentiated self," rather than to do family therapy per se. This
is also an advantage, however. The coaching process can be used as
a therapy format when the client is a young single person pursuing
a career (like social work) in a city far from home.
Many essays have appeared in family therapy journals describing
trips back to families of origin that resulted in extraordinary shifts,

24 5
 Foundations of Family Therapy
intended and unintended, affecting an entire kinship group. These
stories have the flavor of testimonials, and emphasize the appear-
ance of more positive and meaningful relationships and a general
opening up of lines of communication. The approach is especially
appealing to individuals from families in which strains in kinship
ties arehandled by emotional distancing, use of secrecy, and
pseudocommunication, as with the powerful extended clans one
finds in the South and Midwest, and families from ethnic groups,
such as the Irish.
Bowenian therapy with families is an adaptation of Bowen's pref-
erence for coaching individuals. Even when Bowen is dealing with
a couple, he insists on communications being channeled through
him, to lower the anxiety and irrationality that he believes foster the
reactivity of pathological family relationships. This technique in-
creases the power and leverage of the therapist. It also has the effect
of setting up a therapy of two dyads simultaneously, since each
person interacts only with the therapist. It is a curiously one-to-one
model, in spite of its obvious impact on a couple system, and may
betray Bowen's fidelity to a more individual-oriented stance. It is

and for his emphasis on the importance of

for reasons such as this,
objectivity and the supremacy of the rational, that he has been
placed in the psychodynamic group. Although the content of his
work is different, many formal aspects are similar, including the
training implications, the longevity of the process, and the final goal

of an autonomous self.

As for dealing with whole families with children, Bowen clearly

excludes this from the center of his repertoire but his disciples do
not.Such practitioners as Philip Guerin, Elizabeth Carter, and
Monica Orfanides have translated Bowen's theory and practice into
an intriguing multigenerational family therapy. They take Bowen's
cool Euclideanmodel and map out the key triangles attached to the
problem or presenting complaint. These triangles may go back sev-
eral generations, even when the identified patient is a child in the
present one. Here the use of the genogram is a standard tool. With
its help the therapist identifies for the family the reactive processes
which link key triangles in self-repeating chains. Typical interven-
// ,
tions are designed to block triangulation/ to cool off escalations,
to restate heated matters more objectively, to detoxify dangerous

246
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY
issues, or toexpose and nullify the effects of buried secrets. In
general, Bowen therapists try to lessen the extreme emotionality
which they feel is the amniotic fluid that nourishes symptomatic
behaviors.
These goals are often achieved by straightforward interpreta-
tions. For instance, in a family therapy case described by Guerin and
Guerin, the wife is complaining that her husband makes hypercriti-
cal remarks to her when they are in public. 3 The wife says this
upsets her and she starts to escalate emotionally, unlike her hus-
band, who and collected. The therapist may reframe the
stays calm
wife's emotionality as no less a choice than her husband's self-
control. This is the side of Bowenian therapy which works through
making people aware of their own steps in the reactive dance, and
in this sense is a training in objectivity about the self.
Bowen's concept of "reversals," however, comes closer to one of
the hallmarks of what has come to be known as strategic therapy.
To get one person in a self-reinforcing relationship process to re-
verse his usual reaction to the predictable response of the other is

to inhibit or break the mutual-causal cycle. In the Guerins' example,

the therapist also asked the husband if he could agree not to try to
rescue his wife when she behaved foolishly in public, instead of
feeling so responsible for her. The husband responded by backing
off. This intervention example of positive reframing the
is a nice —
husband's behavior was not condemned but described with adjec-
tives of praise. This makes the intervention a nonconfronting one
and is apt to get a cooperative response.
Reversals sometimes work better if only one of the parties in such
a pattern knows about the plan. Coaching one spouse or family
member in a reversal has the added power of placing the therapist
and that family member in a secret pact and reinforces compliance
with the directive. Carter and Orfanides describe how a Catholic
woman, who had angered her extremely religious mother when she
married a Protestant, was helped to regain closeness with this
offended mother. 4 Instead of getting into the usual quarrel over
religious issues, the woman was instructed to write her mother and
tell her how much she appreciated her strong faith and envied the
inner peace it brought her, and furthermore that she felt lonely and
cut off from the family because of her marriage. The mother re-

247
 Foundations of Family Therapy
sponded with unexpected warmth, confessing how much she
missed her daughter and voicing her doubts about her own faith.
As a result the tie between them was reestablished.
although Bowenian therapists do not speak about using
Finally,
—
paradoxical techniques Bowen's framework having no place for
this concept —
they do not hesitate to use them when resistance in
a family makes them necessary. In the Guerins' family case the
presenting complaint was around the inappropriate social behavior
of the daughter. The father criticized the daughter, and also criti-
cized the mother for bringing the girl up badly, while the mother
seemed ashamed and paralyzed, not happy with the child's behavior
but feeling somehow responsible. Guerin started by asking the fa-
ther to increase his efforts to make his wife improve the behavior
of the daughter. Predictably, this task failed, as everybody backed
off from intensifying the pattern. The therapist was then able to
give a straightforward instruction: the mother was told to get more
involved outside the family while the father was asked to step into
the mother's place. The girl began to behave more appropriately and
stopped being a problem. But the husband became anxious about
what his wife was doing outside the home, now that he didn't have
her under his surveillance, and therapy shifted to a marital issue.
This double maneuver is not only paradoxical but also a common
way of "rocking the system": pushing a stuck pattern so hard that
the participants react against the push and then are quite amenable
to a direct suggestion for change, in this case a simple structural one.
A major difference between Bowen's approach and structural and
strategic approaches is that therapy does not stop when problems
go away. The girl in the case above improved and the couple began
to get along better, but at this point the job of the Bowen therapist
has just begun. Now it becomes time for each of the spouses to fix
on their dysfunctional ties to their respective families of origin. This
is the final stage of reaching the elusive "differentiated self." As
Guerin writes, "The therapist must provide individual family mem-
bers a degree of emotional freedom from their reactive triggers. That
way they won't continually be in a responsive position, caught up
in the reactive flow of the family process and behaving like predict-
5
able robots."
This is clearly where Bowenian therapy comes closest to the

248
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY
psychodynamic model, in that the desired result is a mature, auton-
omous self for every adult family member. Therapists who take a
problem-oriented, shorter-term approach feel that this elongation
of therapy can be exploitive, resembling Freud's "therapy intermi-
nable/' In this case, however, the wife had never been able to
respond sexually to her husband and was able to do so only after
revisiting her family of origin, discovering that her aunt had really
been a surrogate mother to her, and reestablishing intimacy with the
aunt. The sexual improvement seemed to happen "of itself," since
it was not being focused on, but Guerin cites this kind of change
as the side benefit of a shift in thebonds that hold individuals in
the grip of a powerful relationship system of which they are not
aware. Certainly it is true that a problem may remain frozen until
patterns connected with the original laying down of the problem are
changed. But let it be understood that one is still dealing with an
addiction in the present. Bowen's use of history suggests strongly
that it is not the revisiting of the past but the redoing of the present
that counts.

The Theory of Repression and Family Therapy

Other family therapists who have emphasized the past, like Nor-
man Paul, seem to have picked up echoes of Freud's theory of
repression, applying it to the family unit instead of the individual.
The idea isone goes back to some event in the past which
that if

has been closed off and relivesit, uncovers it, "abreacts" it, the

symptom presumably attached to it will disappear. This is much the

same as saying that if an individual goes back to the repressed
traumatic incident underlying a symptom and works it through, all

will be well. In Paul's concept of "unresolved mourning process,"

a death or loss in the family that was not properly mourned at the
time is exhumed, so to speak, and the whole family goes through
the ritual in a symbolic way. 6 Paul is usually able to find some
important loss or death to work with in any family where there is

a symptom.

249
 Foundations of Family Therapy
Other historically oriented therapists emphasize uncovering fam-
ily secrets —
and hidden skeletons the fact that a child was adopted,
or a great-aunt went insane. The presumption is that once the
dreaded item is made public, it will lose its fearfulness, and the
symptom which served as a cover will wither away.
A related group of family therapists believes that getting out
feelings,whether of anger or grief, is a way of "abreacting" family
or individual issues which are embedded in affect that is buried or
covert. This is, of course, very close to the psychoanalytic theory of
repression as an explanation for symptoms. Critics who take a sys-
tems view complain that this is an oversimple approach. Simply
airing a feeling does not always extinguish the symptom that pre-
sumably arose to hide or mask it. Family therapists in particular
quickly found that helping a family member "let it all hang out" in
the bosom of the family either subjected the person to later reprisals
or, by transferring blame or inducing guilt, reinforced the sequence
that perpetuated the original problem.
Yet another school of therapy that emphasizes the past-in-the-
present is represented by the work of James Framo. Framo has
adapted Fairbairn's theory of objects relations, and insists that Fair-
bairn's "introjects" (memories or imprints of parents and other sig-

nificant figures still deeply influencing the patient) be asked to come

in person to the patient's or family's therapy sessions. For this
reason Framo insists on including as much of the extended family
as possible, and will sometimes create a large tribal event, finding
7
this a way to break up old, repetitious relationship patterns.
Family sculpting —pioneered and developed by such therapists as
David Kantor, Fred and Bunny Duhl, Virginia Satir, and Peggy Papp
— is still another way of influencing family structures. Sculpting is

something form of psychodrama in which people re-create the

like a
family, usually to elicit major coalition formations and homeostatic
sequences, so that old patterns can be perceived and played out
differently. It is useful in training, because the training group can
take the parts of a trainee's family members, who are of course not
present. It can also be used by members of a family in therapy as
a geospatial metaphor for various aspects of a relationship system:
closeness/distance; splits and alignments; the experience of being
one up or one down in reference to another — all aspects not usually

250
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY
elicitedby verbal reports and often powerful in their effects on
7
family members appreciation of situations that had been hidden
—
from view. Papp's article on sculpting in family therapy or, as she
prefers to call it, "choreography" —explains the uses of this tech-
8
nique.

Ivan Nagy and the Family Ledger

There is a difference between transposing elements of a psycho-

dynamic point of view to the family setting and using information
from past generations strategically to add power to an intervention.
Like Bowen, Ivan Boszormenyi-Nagy does both. Much of the time
he operates from a framework that has a strongly psychoanalytic
flavor, but at other times, seemingly inadvertently, he uses data
from the past to construct multigenerational paradoxical interven-
tions. His most interesting contribution, however, is a rich and
poetic metaphor of families as a multigenerational account book.
Nagy, in the book Invisible Loyalties, speaks of a "family ledger,"
which consists of a multigenerational system of obligations incurred
and debts being repaid over time. 9 No matter when an injustice
occurred, there would be a retributive move at some future point,
although not necessarily by the original debtor. Problems, in Nagy's
view, arise when this justice comes too slowly or is insufficient, and
then what he calls "the chain of displaced retributions" occurs. A
symptom might be the signal that there has been too great an
accumulation of injustices. To deal only with the symptom, without
looking at the history of the problem in terms of the family account
books, would be a grave error.
One example he and his coauthor, Geraldine Sparks, cite is the
case of a tense, irritable nine-year-old girl brought in for therapy.
Itseems that she has been raised by her mother's parents, because
her mother became psychotic and is now living in a hospital. In the
course of family therapy it turns out that the grandmother, at the
age of fourteen, was sent to live with her own mother's parents after
her step-father tried to rape her. Her mother, siding with the step-

251
 Foundations of Family Therapy
father, refused to believe her and sent her away. Later the grand-
mother made a marriage fraught with tension and unhappiness.
Nagy remarks:

It is easy to see how an unsettled account between herself, her mother and

her step-father will have to be "taken out" on her marriage. The resulting
helplessly hostile and frightening atmosphere of the home must have been
reflected in the child's desperate call for attention at school. 10

Thus the done to the grandmother emerges as

injustice a symptom
in a child two generations later (and, one might add, a disabling
psychosis one generation later).

Another example would be the common pattern in which a

mother who was angry at being rejected by her own mother would
compensate by offering total devotion to a daughter of her own. The
daughter, in the language of balance of payments, is asked to rees-
tablish family justice by giving the mother all that her own mother
did not give her. If this child grows up with an unexplained negative

feeling toward the "loving" mother, she can then be made to see
how her mother has used her to make up for her own deprivation,
and can perhaps forgive her mother. Or, alternatively, the mother
can be made aware of how she has unwittingly asked her daughter
to make up for the grandmother's deficiencies, and perhaps change
her expectations of the child.
Nagy does not perceive these patterns negatively but points out
that they may have a profound function in sustaining the family.
He pictures the family as a group of people caught in an ever-
unfolding web of obligations that acts to keep the family or individ-
ual members from harm. Family members impose their own primi-
tive obligation ties through what Nagy calls, comparing his concept
to Wynne's "pseudomutuality," a "counterautonomous superego."
Individual interests are thus sacrificed to the survival of the group.
Confused bickering between parents may be keeping a mediating
daughter close to home, and she, in turn, may be keeping the mar-
riage together. The sacrifice of a child who is symbiotically attached
to a mother who was herself emotionally deprived is a way of
redressing that old wrong. may also be
It a way of keeping the
impaired mother from collapse. Or a child of harsh parents may
push his unspoken bitterness to them onto his wife, thus loyally

252
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY
preserving his relationship to them at the expense of his marriage.
Nagy does not condemn complex accounting system as long as
this

it ultimately balances out, and above all if role obligations are not

so frozen that they do not allow a more just order to be periodically

established.
Elegant as this logic is, it selects only what Elizabeth Carter has
called the "vertical stressors" —the linear chain of events cascading
down the generations. that A causes B causes C
The assumption is

till at last we end up with Z. Something happened in the past that

set off compensatory behaviors and eventuated in a symptom in the
present. The horizontal picture in the here and now is virtually
ignored, and this limits the richness of contextual clues, which tell
the therapist what, in the present, is recursively maintaining the
problem and vice versa.
Thus the therapeutic approach that grows out of this linear,
causal explanation of pathology is close to a psychodynamic ap-
proach. Nagy says,
First, the chain of injustices that led to the
particular symptom in the present must be explored. The therapist
is seen as a benevolent moralist, creating an atmosphere in which
it is possible for people to face their own emotional debts or injus-
tices, and, once they have this insight, correct them. This is made
easier if they can see that they themselves were victims, and that
the way they are acting by wrongs previously done to
is dictated
them, as when it turns out that a tyrannical father was himself
treated harshly as a child. This father can then more easily obtain
forgiveness from the people he in turn has been so harsh to. Or two
spouses who bitterly attack each other may redirect their anger
toward their families of origin, once it is pointed out that their anger
at each other is their way of loyally refraining from criticizing their
own At the same time, the newly designated "victims" are
parents.
not thereby allowed to seek revenge. Forgiveness is the key to this
therapy, which works only when the reactive blaming and hurting
—
process "the chain of displace d injustices" is halted — -

Une cannot help noticing the usefulness of this approach as a

therapeutic rationale very close to what we will later describe as
strategic or systemic approaches, with their emphasis on positive
reframing as a tool for change. If one can make a patient believe that
his hatred of his wife is only displaced hatred toward his mother,

253
 Foundations of Family Therapy
he is likely to end up behaving more pleasantly to his wife. She, in
turn, may reciprocate in a benevolent fashion. Thus their cycle of
recriminations can be broken. Similarly, if spouses redirect their
enmities toward a mother-in-law, this can create a bond; with fur-
ther therapy may grow into one that does not need a mother-in-
it

lawto cement it. And the face-saving implications of this displace-

ment are wonderful; if a mother can point to her own unhappy
childhood as the reason she has been unable to show more love to
her offspring, she can say to herself, "Oh, it was not because I am
a bad person, but because I had an unfortunate history; now I can
behave better without having to thereby acknowledge that I have
7
been in the wrong/
There are obvious limitations to taking the somewhat straightfor-
ward and moralistic stance Nagy seems to prefer. The therapist may
be tempted to act like a wise rabbi or priest, and there are many
patients and families who know how to use this against him. Some
resistant families are experts at games of pseudo-atonement, having
used them as weapons for years. Another possibility is that this
approach may enforce negative self-awareness and perpetuate guilt,
blame, and other unwanted behaviors.
Nevertheless, Nagy seems to be one of the few clinician-writers
(with the exception of Helm Stierlin, who has been much influenced
by him) who have redefined symptomatic behavior as evidence of
family loyalty and indicative of a sacrifice of individual growth to
the interests of the group. How different this is from using a nega-
tive language that defines symptoms as dysfunctional or indicating
7
a dysfunctional family. In spite of his use of words like "scapegoat/
7
"victim/ "injustice
77
— all part of the vocabulary of therapists who
have simply substituted "bad or sick family" for "bad or sick pa-
tient
77
—Nagy is edging out of this linear position and coming close
to a more circular epistemology.
In his descriptions ofone or two cases Nagy seems almost to be
using a system-wide therapeutic double bind. In one of these cases
a teenage boy was on drugs and was getting into minor but poten-
tially serious trouble. It turned out that the parents had separated,

then the older children had gone, and this boy was the last child left

at home with the depressed, obese mother. Nagy says:

254
HISTORICALLY ORIENTED APPROACHES TO FAMILY THERAPY
While on the overt level this boy was conducting an irresponsible pleas-
ure-determined on the family loyalty level he performed a valuable
life,

sacrifice for the whole family. ... In fact, the self-destructive pattern of
his living served as an assurance that as the last member he is not capable
of leaving mother. 11

After the therapist had made the loyalty aspects of the boy's behav-
ior visible to the boy and his family, a major shift apparently took
place. The boy got off drugs and got a job. The mother temporarily
and
lost hers, for a period really did depend on the son, but then
moved ahead in her profession.
The hint of a move toward a concept of circular causality rather
than the historical linear approach also appears in a discussion of the
futility of siding with a presumed scapegoat. Nagy notes that the
therapist will often be rejected in such an effort by the scapegoat
as well as the rest of the family, since the scapegoatis, as Nagy puts

it, game as everybody else. A better way to

just as addicted to the
handle such a situation, Nagy says, is to congratulate the scapegoat
on his position as the "winner," in that he is able to make everyone
else feel guilty, and to sympathize with the rest of the family for
their uneviable position as underdogs.
These interventions, with their paradoxical flavor, are described
out of context. They do not main thrust of Nagy's work
reflect the

and seem to arise as an almost accidental consequence of his empha-

sis on seeing symptomatology in terms of loyalty to the family.

Nagy's theory of change is a basically historical one of cause and

effect proceeding down the generations. To get to a less linear posi-
tion, we must move on to those pioneers who made a sharper break
with the therapeutic establishment: the ecological, structural, stra-
tegic, and systemic schools.

255
Chapter 14

Ecological, Structural,

and Strategic Approaches

The Ecological Model

In this chapter, we shall start by examining the group of systems

therapists who flourished during the period of the late 1960s, when
there was money for community programs and for treating the
psychosocial problems of the poor. In 1962, Salvador Minuchin,
together with E. H. Auerswald and Charles King, got a research
project funded and work with families of delinquent boys
to study
at Wiltwyck School. Minuchin's project, reported on in Families of the

Slums, was more than just another research study. 1 One might say
that if the Bateson research project became a magnetic center for
talent and ideas on the West Coast in the 1950s, the Wiltwyck
project provided a similar climate on the East Coast in the 1960s.
Even though Minuchin led the project, the people he recruited
represented a diverse and brilliant array of talent. Gathered together
were researchers and clinicians like E. H. Auerswald, Richard Rab-
kin, and Braulio Montalvo, to mention only a few. Most of these
people continued to contribute original ideas and to seed new pro-
jects long after the Wiltwyck project ended in 1965.

256
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
Rabkin and Montalvo should first be singled out for the intense
and poetic vision each contributed to the field of community psy-
chotherapy. Rabkin left Wiltwyck and eventually went into private
practice in New York, but wrote in 1970 a brilliant treatise on what
he called "social psychiatry": Inner and Outer Space. 2 To date, there is
not a better metapsychology for the family systems movement.
Montalvo went with Minuchin to the Philadelphia Child Guidance
Clinic in the late 1960s. There he created a series of exquisitely

crafted training tapes, analyzing the contextualmeanings of behav-

iors in family interviews, many of them with minority families.
Some of these can still be seen at the Philadelphia Child Guidance
Clinic.
Of all Minuchin's collaborators, E. H. Auerswald took perhaps
the most interest in using a systems approach to transform the
structure of community psychiatry programs in public settings. He
left Wiltwyck to create a unique "Applied Behavioral Sciences Pro-

gram" at Gouverneur Health Services on New York's Lower East

Side. His aim was to construct a new kind of health service which
would respect the total context of the problems experienced by the
poor population served by Gouverneur. Here he had some support,
since the head of Gouverneur was an innovative public health offi-
cial, the late Howard Brown, who had just turned Gouverneur from

a municipal hospital with a record so bad that it was locally referred

to as The Morgue to an attractive and well-planned ambulatory care
clinic.

Auerswald's "ecological systems" approach, as he called it, was

directed at the total field of a problem, including other profession-
als, extended family, community figures, institutions like welfare,
and all the overlapping influences and forces that a therapist work-
ing with poor families would have to contend with. His essay "In-
terdisciplinary versus Ecological Approach" 3 best summarizes his
position. This paper attacks the idea that to create a fully rounded
health care system it is enough to put together a team of profession-
als What is needed, said
each grounded in a different discipline.
Auerswald, is a new kind of health professional who takes a holistic,
"systems" view of the problem. The Batesonian aversion to "chop-
ping up the ecology" is well illustrated by this article and its central
thesis. Two companion articles — Lynn Hoffman and Lorence Long's

257
 Foundations of Family Therapy
"A Systems Dilemma" and Emery Hetrick and Lynn Hoffman's
'The Broome Street Network" — show the application of the multi-
vectored ecological model to the treatment of crisis situations com-
bining biological, psychological, social, and environmental factors. 4
In creating health care formats that would carry out his ideas,
Auerwald proposed such unheard-of activities as convening confer-
ences in which every member of a multiproblem family, plus at-
tached professionals, would all gather in one room to work out a
plan for coordinating all services relating to that particular family.
His Family Health Unit, set up to serve the immediate neighbor-
hood of Gouverneur, was made up of professionals from the entire
health spectrum and a representative from the local Department of
Social Services as well.
Auerswald was also one of the first to advocate what I call "out-
door," as opposed to "indoor," therapy; he insisted that a commu-
nity psychiatry unit should not only be responsive on a temporal
basis (twenty-four hours a day) but on a spatial basis as well. If the
contextual integrity of a problem was to be heeded, the limitations
imposed by the time and space requirements of health professionals
had to be done away with. To this end Auerswald created a Mobile
Crisis Unit, a team of mental health professionals who operated out
of a van, and who would go from home to school to hospital to
court, as needed, wherever trouble arose.
Auerswald left New York in the early 1970s to head a small
mental health center on the island of Maui, but his ideas are now
incorporated into many health services in the form of crisis units
and quick-response teams. In addition, the need for mapping out
the dimensions of mental health problems in their total ecological
context has begun to be accepted in major institutions, schools, and
training programs.
A piece of research that particularly influenced the course of
community psychiatry in the late 1960s was Donald Langsley and
David Kaplan's project to study brief family crisis therapy at the
Colorado Psychiatric Hospital in 1964. 5 This research was a crucial
piece of evidence backing the trend toward crisis intervention in
working with poor populations. It not only compared the effects of
brief family treatment with routine hospitalization in the case of
acute psychiatric problems, but it had a well-constructed research

258
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
design. Family or "systems" therapy had made many claims and had
a growing number of advocates, but outcome studies providing hard
data were scarce. In this sense, the Denver project was a milestone
piece of work.
The design of the project was simple. Selected at random, one of
every five persons who appeared emergency room of the
at the

Colorado Psychiatric Hospital and was judged by the resident psy-

chiatrist to need hospitalization was sent to the Family Treatment
Unit. This Unit consisted of Frank Pittman, a psychiatrist; Kalman
Flomenhaft, a social worker; and Carol de Young, a public health
nurse. The Unit treated each patient with his or her family on a
brief, outpatient basis. The other patients who were hospitalized

according to the usual procedures constituted a natural control

group. In all, the experiment included 36 pilot cases, 150 experi-
mental cases, and 150 control cases. The groups were found to be
well-matched, and the only requirement for admission to the exper-
imental group was that the patient be between the ages of sixteen
and sixty and be living with at least one relative in the Denver
Metropolitan area.
The approach was essentially pragmatic. The aim of the team was
to get the patient back to his or her previous level of functioning
and the family past the immediate crisis that brought them in. The
whole family was convened, along with any other persons or help-
ers involved, for the first meeting. Occasionally an over-night stay
at the hospital was recommended, but in general the patient went

home with the family that first day. An attempt was made to under-
stand the reasons for the and to mobilize resources within the
crisis

family, or block pressures that seemed to have intensified the crisis.

Medication might be prescribed, but it might be given out to the
whole family, not just the patient. Considerable direct pressure
might be used to get a patient back on his or her feet. A woman who
was unable to function might find the public health nurse on the
team out at her home, standing over her while she mopped the
kitchen floor. A home visit was routinely scheduled within the first
thirty-six hours. Surprisingly few office visits were needed in most
cases, and two and a half weeks was the average length of time the
Family Crisis Unit was intensively involved in a case. It should be
added that external resources from the community were routinely

259
 Foundations of Family Therapy
brought in (visiting nurse services, vocational rehabilitation, and so
forth) to continue the work of the Crisis Unit in whatever area was
most essential. At the earliest possible point the unitwould back
out, always with the proviso that they were available to the family
if the crisis flared up again. Usually, subsequent crises were handled
with a minimum of involvement, often only a phone call or two.
The most important finding of the project was that acute cases
could be treated equally well, if not better, with a brief crisis ap-
proach. Patients did not waste time being "sick" in the hospital but
started to function almost immediately. Patients who did go into the
hospital not only took that much longer to get back to normal, but
became addicted to hospitalization. Thus their rate of recidivism
was much higher than with the experimental group, most of whom
did not use the hospital at all Other natural be-
for future crises.
nefits, of course, were the obvious savings in time, money, and

personnel. This finding may have contributed to the fact that soon
after the book describing the success of brief family treatment ap-
peared, the unit was disbanded and the Colorado Psychiatric Hospi-
tal resumed hospitalizing all acute cases.

An illustration of the rather unconventional methods used by the

Family Crisis Unit has been described anecdotally by Frank Pitt-
man, the psychiatrist of the team. Pittman recounts a time when the
team went out to the house of a woman who had discovered her
husband was having an affair. The team found her lying in an
apparently catatonic state on the kitchen floor. All efforts to get her
to get up were to no avail. Pittman looked out the screen door to
the back yard and saw a wet cocker spaniel whining to get in. He
had also noticed that the woman was wearing a very pretty
housecoat. Pittman opened the screen door, whereupon the muddy
dog rushed to its owner and began to lick her face and climb all over
her. The woman immediately got up and offered to make the team
coffee. It is for this reason that Pittman sometimes characterized
theirapproach as "wet cocker spaniel therapy." A very nice presen-
tation of their work on a case can be found in Haley and Hoffman's
interview with the Denver team, "Cleaning House." 6
Another center that pioneered in developing the crisis interven-
tion model was Bronx State Hospital in New York. Family therapy
had been introduced as part of the training program at Bronx State

260
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
by Zwerling during the 1950s, but its Golden Age came dur-
Israel

ing the 1960s. During this time, the Family Study Section was
started at Bronx State by Andrew Ferber, with the collaboration of
clinician-researchers like Chris Beels, Marilyn Mendelsohn, Nor-
man Ackerman, Thomas Fogarty, Philip Guerin, and many others,
making up a distinguished, if peripatetic, faculty. As well as doing
research and offering training in family therapy, this group ex-
tended the use of crisis techniques to the turbulent psychiatric
problems of families of the South Bronx.
A profound influence and presence at Bronx State during this

time was the late behavioral scientist Albert Scheflen. Working with
anthropologist Ray Birdwhistle in Philadelphia in the 1950s,
Scheflen had helped to invent the field of "kinesics": the microstudy
of human communication patterns in social settings. During his stay
at Bronx State, Scheflen inspired many of the family therapists who
learned from him to analyze family sessions as well as other behav-
ioral events, producing some striking documentation of the hidden
patterns that shape communication. Scheflen's own focus at the
time was on collecting data by videotape which would allow him
to compare the use of space and territory in the homes of families
from different ethnic groups in the community. Indirectly, he
helped train a generation of clinicians dealing with poor, minority
families to think and work like anthropologists, broadening the
family field to include a growing concern with issues of ethnicity.
At the time when researchers like the above were developing new
methods for working with poor communities, other practitioners
—
were experimenting with natural systems tribal networks or
—
neighborhood configurations which seemed more logical arrange-
ments for helping the poor and the isolated than artificial groups.
Ross Speck and Carolyn Attneave are the pioneers of this branch
of therapy, which goes outward from the nuclear family into the
larger groupings around it. Speck and Attneave became known for
forming huge community or kinship networks that met ceremoni-
ally on a regular basis to deal with problems such as a symbiotically
attached mother and child, and that often continued to meet after
therapy stopped. 7
Minuchin worked with the ecological model for a brief time but
abandoned it to concentrate more narrowly on child problems

261
 Foundations of Family Therapy
within the nuclear family and to work out the "structural" approach
for which he is now widely known. Harry Aponte, who had joined
Minuchin in the early years of his directorship of the Philadelphia
Child Guidance Clinic, stuck to the families of the very poor as his
main concern. He found the combination of an ecological frame-
work with a structural model peculiarly effective with these fami-
lies. His essay describing an eco-structural approach to a school-
family problem and his article on "underorganization" in the poor
family are unique descriptions of a way of applying structural ther-
apy to fields that include systems other than the family. 8 The con-
cept of "context replication," in which dynamics in the home are
repeated in the child's situation at school, is dramatically illustrated
in the school-family article.
Aponte is especially good at conceptualizing the layering of con-
texts and is sensitive to dilemmas that include systems other than
the family. The "ecological" therapist must be prepared to deal with
the larger professional scene: doctors giving medication, the use of
hospitalization by families, and the by other systems
role played
such as schools. It would be a mistake to assume that any therapy
can safely be "indoor therapy" and to abandon the field-oriented
ecosystems model, whatever the presenting problem may be. At the
very least, the therapist himself brings a wider ecology to the family
as soon as he steps into the case.

1 The Structural Approach

Minuchin's approach is best represented in Families and Family Ther-

apy, a book that has become a classic. 9 Minuchin has a clear method
and with that method. He also has striking proof
a theory consistent
that his methods work with severe problems of childhood, as
demonstrated by his research with families of psychosomatic chil-
dren. 10
Minuchin's normative model for a family that is functioning well
is especially useful. According to him, an appropriately organized

family will have clearly marked boundaries. The marital subsystem

262
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
will have closed boundaries to protect the privacy of the spouses.
The parental subsystem will have clear boundaries between it and
the children, but not so impenetrable as to limit the access necessary
for good parenting. The sibling subsystem will have its own bound-
aries and will be organized hierarchically, so that children are given
tasks and privileges consonant with sex and age as determined by
the family's culture. Finally, the boundary around the nuclear fam-
ily will also be respected, although this is dependent on cultural,

social, and economic factors. The extent to which kin or agents of

larger social institutions are allowed in varies greatly.

With this model in mind, the therapist then has the task of noting
{

the angle of deviance between it and the family that comes in thej
door. Therapy, from a structural point of view, consists of redesign-
ing family organization so that it will approximate this normative

model more closely. For instance, a functional family will have a

clear generation line. This means that if mother and daughter are
acting like siblings, the therapist may put the mother in charge of
the daughter's activities for a week. Similarly, one finds a fair degree
of individuation in a family that is working well. If the boundary
that delineates an individual is not being respected, the structural
therapist may ask each person to think and speak only for himself.
Or, since in a functional family the marital subsystem and the
parental subsystem have distinct boundaries, the therapist who
finds that a couple spend all their time parenting may ask them to
go away together without the children.
The process seems very logical and very simple. It is as though
one began by saying, "What are the organizational characteristics
of a family in which things go well and family members do not have
problems?" and, when somebody does have a problem, noting
which of these characteristics are absent and changing the family
accordingly. The assumption is, of course, that a symptom is a^
product of a dysfunctional family system, and that if the family
organization becomes more "normal," the symptom will automati-
cally disappear. If this were one's central theory, one would not
have to worry overmuch about the particularities of the symptom,
its history, its present effect on other people, or any other specific

detail. One would merely scrutinize the way the family was orga-

nized (Does everybody go through mother? Is father treated like a

263
 Foundations of Family Therapy
child? Is oldest daughter acting like mother?) and shift it from a less
normative to a more normative state.

In practice, thisapproach works. There are detractors who say

that Minuchin's idea of what is normal is biased and does not
account for families of other classes and cultures. On the contrary,
his model is flexible enough to include the alternative ways in
which poor or ethnically different family structures can be orga-
nized, and respects these ways insofar as they work for the well-
being of the individuals in those families.
One further point should be made about Minuchin's conceptual
framework. It owes much to systems theory, yet it leans very little

on the cybernetic paradigm that has been stressed so much in these

my view is one of the identifying facets of the
pages, and that in
family therapy movement. Only occasionally, as in the case of the
anorectic girl described in Chapter 11, does Minuchin speak of
"cycles" or creating a "runaway." For the most part his language
seems to derive from organization theory and role theory, drawing
heavily on spatial metaphors like boundary, mapping, territory,
structure, role.

C Of major importance therapeutically is Minuchin's inclusion of

the therapist as an active intruder, changing the family field by his
very presence. Schools of therapy that emphasize getting informa-
tion or delving into history miss the fact that the focus on content
can obscure for the therapist extremely important matters: To
whom does he speak? Who is allowed to speak? Whom does he
elevate? Whom Which persons does he bring
does he challenge?
together? Which does he push apart? With whom does he make a
coalition? With whom does he not? It is by such moves that the
therapist begins to restructure the relationship system in the family,
and to alter the context that supposedly nourishes the symptom.
Minuchin's method of "mapping" the psychopolitical terrain of
a family saves a therapist much time since the nature of the family's
organization gives a structural therapist the clues he needs to deter-
mine which directions to go in revising patterns of relationship in
the family. In Chapter 5 of Families and Family Therapy, Minuchin
shows his own method of mapping family groups, indicating im-
portant factors such as membership in coalitions, the nature of
boundaries, and how the subsystems are structured. 11 By delineat-

264
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
ing the form these aspects take in a family that comes in for treat-

ment, and by revising the map as treatment progresses, Minuchin

gives us a graphic documenting the stages of therapy.
method for
To demystify Minuchin's expertise, it may be useful to summa-
rize a commentary on one of Minuchin's most elegant interviews

which was published as 'The Open Door: An Interview with the

Family of an Anorectic Child/' 12 The article is a step-by-step analy-
sis of the therapist-family interaction throughout the interview.
This is the first time Minuchin has met the family and also the last,

because he is a consultant here and will be handing the case to

another therapist. Minuchin has spent some time getting to know
the family: a mother and father in their forties; the thirteen-year-
old anorectic girl, Laura; the twelve-year-old sister, Jill; and the
eight-year-old brother, Steven. He has found out that the girl

started dieting while away at camp summer, and has been

that
losing weight ever since. However, she has not yet had to be
hospitalized.
In a move to assess family interaction around the symptom,
Minuchin challenges the father's statement that family members
never disagree by asking how the parents deal with Laura when she
does not want to eat. The mirror-image disagreement starts to sur-
face. Father, the more domineering parent, pushes food on the girl
and gives up only reluctantly. Mother says she tries to push and
then stops, because she sees it upsets Laura. In this pattern the
"one-down" parent covertly supports the symptom that so success-
fully eludes the authority of the "one-up" parent. Minuchin's re-
quest that family members enact behaviors elicited by the symptom
is a typically structural move. The therapist is not satisfied with a

report of what goes on at home; he wants to see the sequence with

his own eyes.
Next Minuchin goes to the twelve-year-old sister, Jill, and asks
what happens when she disagrees with her father. She replies that
her father gets angry when he asks her to comb his hair and she
refuses. For some reason she is able to be more open and forthright
than her sister. Minuchin thinks she means that her father objects
to her not combing her hair. He goes over and musses her hair and
asks her to play out the scene. The father tells Minuchin he has
misunderstood, and describes a nightly ritual that consists of him-

265
 Foundations of Family Therapy
7
selfand the children lying together on the parents bed while the
children comb the father's hair or massage his legs or rub his back.
Father says that he sometimes does these things for them, but that
he and the older daughter ''have not done so much for each other"
during the past year. Minuchin finds out that Mother is usually
busy at these times, folding laundry or washing the dishes. When
asked if she rubs Father's back too, she says yes in a doubtful way,
but when further asked if she ever shuts the bedroom door and
sends the children out, she responds, "Never!" In fact, she says, the
doors are hardly ever shut to anyone's room.
With this piece of information, the contextual circumstances of
the symptom seem clear enough. Laura has been close to
girl's her
father; she has in a way been a present from Mother to Father, the
wife having felt it necessary to remain somewhat distant from her
husband. However, as the girl nears adolescence, the dictates of
nature and society enforce a movement toward more autonomy. At
the same time the implications of this move are threatening because
it will disrupt the parents' relationship. By becoming anorectic the
girl remains close to home and very close to her mother. At the same
time she is still available to Father but not sexually available, since
many of her sexual characteristics are inhibited or disappear. Fi-
nally, by refusing to eat, she asserts herself in a desperate caricature
of adolescent rebellion. This symptom, like all others, is a perfect
artifact of nature, with something in it for everyone.
Minuchin's response to the father-children closeness described
by Father is to disrupt He does not point it out or challenge it,
it.

but interferes with it by the way he directs personal exchanges with

and among all The father is clearly the dominant
family members.
figure in the family and Minuchin never confronts him directly.
Instead he carefully reframes the sensual activities of hair brushing
and back rubbing as "nurturance" between Dad and the kids, say-
ing, "Dad is a cuddler, he likes people to be close to him."
Then, once he has joined with the father, who relaxes visibly,
Minuchin moves to the all-important boundary issue, and begins to
talk about the open-door policy Mother has described. He asks
Laura whether she ever closes the door of her bedroom. When she
says she does, he asks whether people knock before they come in.
Does the sister knock? Yes. Does Mom knock? Yes. Does Dad

266
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
knock? Laura says yes, but her tone is uncertain. Minuchin says,
"You are doubting," and she agrees that sometimes he does and
sometimes he doesn't. Minuchin asks if she would like him to knock
at her door before entering. Very softly she says, "Yes." Minuchin
now asks whether she ever tells her father that she would like him
to knock. She says no. He asks her if it would bother her if she asked
her father to knock at her door. She says she doesn't think so.

Here, in one of those shifts that make his work seem like delicate

choreography, Minuchin challenges her statement, siding with the

father: "Ihave the feeling that it would bother Dad because he is
a very loving Dad that likes always to have people respond to him
—
and he respond to people to the children, certainly." Minuchin has
moved to support the thing that is not normal, the kind of behavior
that presumably reinforces the symptom. Why? Perhaps because he
is going to ask the girl to do something she does not often do, which
is to tell her father that she wants privacy. He knows that for this
to be possible, he must back the father; otherwise the girl's loyalty
to him will make it hard for her to take a position against him. He
says to Laura, "Ask Dad if it will bother him if you asked him to
knock at your door." Laura, in an almost inaudible voice, does so.

The father says, "Probably so," and adds, "because I like to have
all the doors open."
Jill, the forthright sister, now interrupts to say that he doesn't
knock because he can't stand closed doors and will open them if
they are. Minuchin asks the other children if they, too, sometimes
want their doors closed and get answers in the affirmative. Even
though the father has not given verbal consent for Laura to close her
door, Minuchin does not push further. He has engineered a very
gentle confrontation between Laura and her father on the issue of
privacy, and has helped the other children, who can state the same
position more strongly, to back her. That is all, yet that is much.
Moving from the Father-Laura dyad, Minuchin now takes on
the Mother-Father dyad. The mother is equally involved in the
behaviors helping to sustain the problem, although it is the father
who plays the most visible role. The mother is compliant to her
husband in one way, but by being unavailable to him for inti-
macy, she is not at all compliant. To help free Laura, Minuchin
will have to give Mother as well as daughter some different ways

267
 Foundations of Family Therapy
to oppose Father. So far, the mother has mainly used the denial of
intimacy and a covert mutinous coalition with the children, espe-
cially Laura. After much work, Minuchin finally gets the mother
to stand up to the father on an issue of her own she objects to —
his phone calls at dinner, especially as the phone wire goes right
across her neck. The couple falls into an arguing sequence typical
of an apparently one-up/one-down pair; the mother makes a fee-
ble complaint, the husband interrupts, then begins a monologue
that collapses into silence. Minuchin finally has to get out his big
guns, and starts repositioning people and moving chairs about
until he has finally placed himself between Father and Mother.
Once he has blocked Father's access to Mother, he carries on a
cheerful conversation with Mother about her loneliness. He goes
back to the picture of the father and the children on the bed, say-
ing, "Your children are not employing you and your husband is
—
not employing you what kind of corners are there where you
are?" For the first time in the interview, the father (who has been
desperately trying to break in on this flirtation) capitulates to
Minuchin, saying, "You have a point." Minuchin goes on a while
longer in his tete-a-tete with the wife, and then stands up and
gives his chair to the father, saying, "I want you to come back to
your wife." This is one of the clearest examples of a structural,
in-the-room, rebalancing of a couple that I know of. It is an im-

portant move because it is clear that no conflicts will be allowed to

arise, much less be resolved, between them, until the timid wife
feels that she has gained some parity with her overbearing hus-
band. And until that time, Laura will probably have to remain
anorectic.
During the last act of this drama, food is brought in. Minuchin
focuses solely on Laura and her eating- even though the subject of
food is never brought up. Instead, Minuchin launches into a conver-
sation with Laura about her age. One of his major contributions to
family therapy has been to point out the confusion of the normal
sibling hierarchy in a family that is not working well. Here he
remarks to Laura that her younger sister seems more like a twin, or

even older, and that the parents treat the two of them alike. He asks
Laura if she likes being treated as if she were the same age as Jill.
He says, "Maybe you are telling them that you are twelve, and

268
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
maybe you are telling them that you are ten, and maybe we are just
making a concession in treating you as if you are twelve. But some-
thing wrong here."
is

Minuchin sits by Laura sandwiches are brought in and

as the
continues to direct a rapid stream of conversation at her while she
eats —
Does she buy her own clothes? Can she say when she can go
to bed? —and finds out that she is allowed all these minor freedoms.
Minuchin presents her with an interpretation —her not eating is the
only way she has to rebel in this permissive household where by
being given so much autonomy she actually has so little. But he is

clearly concentrating on her sandwhich, which she keeps toying

with. He times his bites with hers, at the same time, paradoxically,
directing her not to eat:

At the point which you are fourteen, Laura, you will eat without any
at
problems. But I think it is good that you are not eating now because I think
that this is the only area in your family in which you have a say-so. And
at fourteen you'll need to have a say-so in another way. And you know,
7
at this point, that's the only way in which you say, "No/

The parents are not allowed to intrude on Laura during this time.
Minuchin suggests that maybe Daddy doesn't want his big girl to
grow up. He asks what will happen if she grows up, and she says
softly, "I don't know get married?" Minuchin says, "You will
. . .

then be interested in rubbing the back of someone else. Dad, what

will happen then? Maybe you will need to have Connie [his wife]
rubbing your back." At this moment, in an apparent accident, the
father spills his drink onto his lap and the wife helps to clean him
up.
The session ends with Laura finishing her food and Minuchin
giving the family some "homework." He extends the metaphor of
the open door, with its implications of intrusiveness, by telling
Laura that she is to close her bedroom door for two hours a day,
during which time her parents must knock if they wish to come in.
Jill is to keep her door open, since she will not earn Laura's privilege

until the following year. The little boy, who has been staying up late
and sleeping with Jill, is told to obey his mother and go to bed when
she tells him to. Turning to the parents, Minuchin tells them that
they are to close the door of their bedroom every night from nine

269
 Foundations of Family Therapy
to ten and watch television together without the children. The
session ends.
This was only a consultation interview. The rest of the case was
handled by another therapist, who brought it to a good conclusion.
The importance of the interview described is that it contained the
blueprint for the eventual restructuring of family relationships, and
gave every person the temporary experience of living in a normally
organized group. When the family first came in, the children were
overtly presented as father's bosom companions, but they were
covertly allied with mother. This seesaw arrangement of child-adult
alliances kept the parents in uneasy balance. By the end of the
interview, however, the children were disengaged from the parental
struggle; the maritaldyad was cordoned off and evened out; and the
sibling hierarchy had been revised, creating a set of stairs up which
the oldest daughter would, one hoped, march on her way out.
A serious deficiency is that Minuchin's theory of change contains
no provision for so-called paradoxical techniques. If asked, he often
says that he does not use these methods, although, as in the inter-
view just described, he comes very close to doing so. A more obvi-
ous example is reported in a profile on Minuchin written by Janet
Malcolm for the New Yorker. She quotes Minuchin as telling the
parents in a family with a girl who had been hospitalized for a
psychotic break:

I am concerned that when you leave here today your daughter will go crazy
again. And think the reason she will do it is to save your marriage.
I

. . . Yvonne, I suggest that you go quite crazy today, so that your parents
can become concerned about you. Then things will be O.K. between them.
. You're a good daughter, and if you see a danger, go crazy. 13
. .

Other gifted practitioners who have worked under Minuchin

Braulio Montalvo and Harry Aponte, for instance use symptom —
prescription and paradoxical interventions in many inventive and
subtle ways. It is a genuine limitation that although Minuchin's
theory most eloquent about family systems and family structure,
is

it does not contain a comprehensive enough theory of change to

cover the area misnamed "resistance," and the moves which deal
most successfully with it, especially in cases of what Minuchin
would call "enmeshed" families.

27o
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
Another difficulty with Minuchin's approach is that it sounds
simple but is hard to teach. Minuchin works with analogic behav-
iors so much that his trainees must see many families before they
can begin to recognize the invisible patterns that an experienced
structural therapist knows at a glance. It is not much help to rely
on one's rational faculties when working structurally, any more
than it is by reading about it
helpful to learn ballet or watching it.
Suffice it to say that to be a good structural therapist requires much
experience, and extensive live supervision by a master. — J

The Strategic Approach

Haley first coined the term "strategic" to describe any therapy in

which the clinician actively designs interventions to fit the problem.
The term has become identified with the work of Weakland, Wat-
zlawick, and Fisch as represented in the article, "Brief Therapy:
Focused Problem Resolution," and the book, Change: Principles of Prob-
14
lem Formation and Problem Resolution.
These therapists say that they are not interested in family struc-
ture or the family system. Unlike Minuchin, who starts at an ab-
stract level and works in, they start at the most specific level and
work out. Consequently they have a very clear procedure for the
initial interview, much like the questioning a detective might do to
solve a mystery. What is the problem? Who did what the last time
it happened? When is it likely to occur? When did it first appear?
This extreme interest in the details of the symptom is misleading
if one assumes that the symptom is the only thing the therapist is

interested in. Using the model of the self-reinforcing sequence, this

group assumes the symptom is being maintained by the very behav-
iors that seek to suppress it —
by the "solution." As we have seen,
a careful analysis of these behaviors will indeed show that at the
same time the problem is being attacked by behaviors opposing it,

it is being supported covertly by eliciting behaviors. The therapist

is looking for this cycle or sequence. With the wife whose constant
jealous questioning of her husband only reinforces his reticence,

271
 Foundations of Family Therapy
which only reinforces her jealousy, ad infinitum, the strategic thera-
pist will seek to find a way to interrupt or block this vicious circle.
It may be enough simply to point out to the wife how her behavior
is achieving the opposite of what she wishes to happen. But the
therapistassumes that if the problem were that easy to solve, it
would probably not appear in his or her office. Therefore the thera-
pist looks for a more tactful way
change or disrupt the sequence.
to
It is in this light that one must understand the strategic therapist's
emphasis on reframing, the technique by which the therapist res-
tates a situation so that it is perceived in a new way. The strategic
therapistmight persuade the wife that if she suddenly stopped her
constant questioning, or if she even became mute for a week, this
might make her more mysterious in the eyes of her husband. With
some women this idea alone might make enough impact to cause
them to change.
On the other hand, the behaviors may be so entrenched that the
wife literally cannot stop herself. Getting the husband to be more
open may not succeed either. The next move will probably be in the
direction of encouraging rather than trying to stop the jealous be-
havior. The therapist might tell the wife that her husband seems
strong and self-sufficient but that he is really a shy, dependent
person who is unable to ask openly for her attention and concern.
Since he cannot ask for a more direct confirmation, her jealousy is

for him a proof of love. Therefore she should redouble her jealousy.
This directive may well produce a recoil. Not only will the wife feel
somewhat reluctant to continue her supervisory behaviors, espe-
cially if the therapist asks her to intensify them, but her husband
may not like the implication that he is a shy, dependent person.
Both parties may collude against the task, and announce the follow-
ing week that they did not follow it, but that their relationship has
nevertheless improved.
At this point, if the couple does not come in with a new problem,
the strategic therapist feels that his work is finished. He has not
tried to look at the context of the problem any more than was
needed to resolve it. He has not inquired into the history of the
marriage, or the extended family, or their children, or the back-
ground and early childhood of each spouse; nor has he made a guess
about the meaning this jealousy may have in the larger context of

272
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
the family. It may be that when the husband's father died, he found
himself faced with a lonely, demanding mother, but that it would
be unthinkable for him or his wife to push away this suffering
person. The jealousy may be the only way the wife is able to
communicate to her husband that she wishes she had the time and
attention from him that she used to have, at the same time that it
insures a distance between them that protects the husband's loyalty
to his mother. But this is not a piece of information, or a surmise,

that will necessarily interest the strategic therapist.

Nor will he concern himself with other dysfunctional behaviors
in the family if they are not presented as problems. The couple may
be taking their six-year-old daughter into bed with them when she
has nightmares, but if they do not complain about this, the strategic
therapist does not investigate this habit or suggest a change. And
he does not assume that he must work with a marriage when a child
who seems to be mediating the relationship between the spouses,
and consequently becomes a problem, gets better. The partners may
choose to bring their marriage in, and make a new contract over it,
but the strategic therapist does not push himself in where he is not
asked. In the world of therapy, this person is a minimalist.
As amore extensive example of the brief, strategic approach, we
might review highlights from a case that the Palo Alto group pre-
sents during seminars and workshops. The therapist in this instance
is Paul Watzlawick. The family consists of Mother, Father, a sexu-

ally acting-out fifteen-year-old girl and three younger siblings, two

girls and a boy. The girl has run away from home and seems headed

for a career as a juvenile delinquent. Therapy consists of five ses-

sions, mostly with the parents. The therapist does not include the
younger children in therapy.
In the first session the therapist meets first with the parents. They
describe their daughter as boy-crazy, argumentative, flighty, angry,
and impossible to control. They present themselves as ground down
by the constant fights and arguments. The therapist, taking advan-
tage of their frustration, asks if there is any way they could give her
a taste of her own medicine
—
"grate on her nerves the way she
grates on theirs." They say they would love that. This is an impor-
tant response for the therapist to note, as they might also have said,
"Oh, no, poor child. We couldn't do that." But these parents are

273
 Foundations of Family Therapy
willing to go along with the therapist's suggestion. He insinuates
that they think of a way to be unreasonable instead of reasonable.
He suggests, for example, that when she asks to go out in the
evening they say, "No." When she protests, instead of trying to
justify their position, they are to say something absurd like, "Be-
cause it's Friday." The parents like this idea, but the therapist re-
strains them, saying, "Don't try anything like this yet, just think
about it."

He then meets with the girl alone. As with the parents, he works
upon the girl's self-interest, as she sees it. When he asks what she
would like to see changed for her own advantage, she says she is
tired of all the fighting. The therapist remarks that she has gotten
herself into an extremely powerful position, and that her parents
have become quite helpless to deal with her. The best way for her
to maintain her position, he says, is simply to persist and go even
further; if they deny her something, she to ask them for their
is

reasons, and keep hammering at them until they finally give up. He
adds in a musing tone of voice that there is always a price; she may
be in a chronic state of rage, because at first the fighting will get

worse; she may even end up at times in Juvenile Hall, but she can
get used to that; the most important thing is to pursue her advantage
because then she will win. He then says that he will continue to see
her parents, because his task will be to teach them how to live with
this situation.
He never sees the girl again. The next four interviews are with the
parents alone. When he sees them, he asks what they have been able
to come up with. The father can easily think up frustrating answers
to the girl'sdemands, but the mother turns out to be the more
benevolent parent. She says she feels helpless. The therapist, far
from challenging this position, goes with it. He argues that they
should change from a position of strength to one of weakness,
because if one is helpless, there are all kinds of things one can get
away with. For instance, if the daughter is bolting down her dinner
before going out on a date, the mother might do a stupid thing like
spilling a glass of milk on her, and then apologize, saying, "Oh, I'm
so upset, I've been so depressed lately; I do all kinds of stupid
things." If comes home later than she is supposed to, they
the girl

can lock the doors and put the lights out; then, after having made

274
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
her wait a long time, the mother can go out and say, "I'm sorry I

left you out in the cold; I do the silliest things these days."
In this way the therapist captures the resistance of the mother,
shown as reluctance to fight her daughter directly; the father does
not have to be pushed. In general, the rule seems to be that when
the child is perceived as "bad," the parents will be encouraged to
seize power directly; if the child is perceived as "sick," the goal will
be the same, but the tactic will be labeled as benevolent, and re-
framed as "strong medicine." If the parents are scapegoating the
child, the therapist may push an extreme,
their vengefulness to
suggesting not just temporary placement but permanent placement,
and commenting perhaps on the terrible conditions in facilities of
this kind, over which parents will have little control. With a surly
youngster, the therapist might comment on how cooperative he is
being in helping his parents prove he is a rotten kid. And so forth.
In the case we are describing, the parents came in for the second
session saying that they were carrying out the therapist's sugges-
tions with success. Instead of arguing with the daughter, the father
was responding to every request with, 'Til think about it," which
Mother simply agreed with her and did not
frustrated her terribly.
argue either. The daughter was more and more furious because they
wouldn't fight with her. The therapist pushes on, suggesting that
during the following week they take an even more powerful posi-
tion of extreme helplessness, and telling the mother to say that
something upsetting came out in the session that she cannot tell
about but that has made her very depressed. Watzlawick, a master
of reframing, says, "You should give her that creative doubt and
insecurity that a youngster needs in order to grow up."
The mother then mentions that the girl's sixteenth birthday is

coming up, and the daughter has asked for a pair of boots which cost
about thirty-two dollars. It turns out that the mother is irritated

over her daughter's old, worn-out bras. The girl refuses to buy new
ones (she has a large cup size and her bras cost eight dollars apiece)
and never washes out the few she has. The therapist suggests that
the parentsbuy her four new bras for her birthday they will cost —
exactly the same as the boots she is hoping for and when —
she opens her present and expresses disappointment, the mother
must act genuinely upset. The suggestion was acted upon and the

275
 Foundations of Family Therapy
parents reported with delight on their daughter's baffled reaction.
By the fourth session the parents are conniving together, plotting
how to get the best of this infuriating daughter, and are mystifying
her completely. The girl undergoes a miraculous personality change.
The parents can hardly recognize her as the same person. She has
become a pleasant, cooperative child, with a normal life and inter-
ests. She has taken up sewing (after the mother "stupidly" sewed

the back of a dress to the front in response to the daughter's request

that she mend it); she is eating dinner like one of the family; one
day she even bought her mother a box of chocolates.
During the last session, the therapist (as is customary with this
group) expresses concern about things going too well. He warns the
parents that the outcome of their success with their daughter will
almost certainly be that she will become a delightful child; then it

may become difficult for them to let her grow up. So it might be a
good idea for them to reinstate the old situation so that they will
not be too unhappy when she does leave. He asks them to imagine
how they might have their first relapse into their old pattern. This,
predictably, only reinforces the change which has taken place, and
after a three-month follow-up it turns out that the parents have
begun go out together (which they couldn't do when the girl was
to
acting so badly); her grades have jumped from Ds to Cs and Bs, and
she seems altogether a much happier person.
A structural therapist would say that the change came about
because the therapist got the parents to unite and take control of the
girl's behavior, where previously mother was covertly supporting
her against father and subverting the generation line. The strategic
therapist would agree but would ascribe the change to the thera-
pist's ability to reframe the situation so that the parents would

behave differently with the girl. It is not necessary to change all

behaviors in a self-reinforcing cycle to remove the problem, and not

essential to have all family members present to bring about change.
The strategic school focuses on the problem as the unit to be
attacked, not the family. Thus, unlike the structural school, strategic
therapists do not worry about seeing all members of a household
together. They even prefer to see individuals or family subgroups
separately, maximizing change by setting one group or person se-
cretly against the other. One might also say that where the struc-

276
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
tural therapists actively repattern relationships in the room, the
strategic therapists are strangely inactive in the room. For them the
key to change is the art with which they can reframe the client's
perception of the context of his behavior. They use the analogy of
salesmanship in teaching their method, and actually send students
out to watch how car salesmen persuade customers to buy a prod-
uct. The point is to change the perceived "reality" of the client so

that different behaviors become possible.

An example cited by the authors of Change is the case of a man
with a stammer who was taking a job as a salesman. His under-
standable belief that his speech problem would interfere with his
become a good salesman was challenged by the idea that,
ability to
farfrom being a liability, his defect would be an asset. People
always pay more attention to someone who has trouble talking, in
contrast to the way they often turn off in response to a fast-talking
huckster. Therefore this man was encouraged to increase his stam-
mering as a way of becoming a better salesman. This is an example
of the use of positive reframing in connection with prescribing the
symptom. Clearly, in their clinical work the strategic therapists use
a variety of therapeutic double binds and a variety of benevolent
rationales for making them seem palatable.
For strategic therapists, the Art of Therapy becomes the Art of
Rhetoric, and strategic therapists indeed have the bad reputation
that the Sophists had in ancient Greece. It does not matter, our Palo
Alto friends say, whether we believe the ingenious rationale we give
the client to make him change his ways; as long as he changes them,
our job is done. This position has been objected to by more tradi-
tional therapists, who feel that the use of such currency debases the
profession. Charges of "manipulation" and "social engineering" are
heard in the land, and are cheerfully accepted by the strategic peo-
ple. They claim only to be simple craftsmen, solving people's prob-

lems in the most expedient (and least expensive) way.

And in truth, this is the strength of their position. They have a
narrower focus than other types of therapists. Thus, like the strict
behavior therapists, they have a good chance of achieving what they
set out to do. Oddly enough, because
of this they also have a better
chance of accomplishing more. Although they disclaim interest in
the family as a "system," they work systemically, and quietly hope,

277
 Foundations of Family Therapy
if not expect, that a small change in an important family relation-
ship will have a domino effect on other relationships: a marriage
may "of itself" get better after a child improves, simply because for
the first time in years, the parents will stop thinking about the child
all the time, rediscover each other, and start having a good time.
The contribution of the strategic school has been to create an
elegant and parsimonious model for change. The group's procedure
for tracking the behaviors around the problem is an invaluable
clinical tool, all the more so because it is based on such a clear

understanding of the self-perpetuating sequence of behaviors

around a symptom. Their use of paradoxical techniques to counter
this kind of problem has been another invaluable tool. They take
a strong therapeutic position in constantly challenging the family's
wish to change, and insisting on
minimal powers to help. This
their
not only maintains the therapist's control but draws on the opposi-
tional qualities of most families with psychiatric symptoms, almost
obliging them to change by resisting the injunction not to.
The strategic approach offers its own hazards for beginners, pro-
viding a deceptively simple formula for identifying what to change
and how to change it. It is really not enough to ask about the
problem, find out what solution is being tried, and then interrupt
or reverse that solution. These are the shortcuts of master therapists
who have an understanding of the complexity of the processes they
are dealing with, and who have evolved an intuitive methodology
for redirecting these processes. They can well say that they do not
have to bother with the structure of the family they know it by—
heart. In the same way, the structural therapist can elect to ignore
the particularities of the symptom or the behaviors sustaining it; he
knows very well how to recognize a symptomatic cycle and to break
it.

We
might conclude by saying that if the structuralists need to
admit to their knowledge of process, the strategists need to admit
to their knowledge of form.

278
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES

Haley's Problem -Solving Approach

Although in some respects Haley should be placed with the

strategic school, he is really a bridging figure between the strategic

and structural positions. Haley presents his strategic side most

on the clinical work of Milton Erickson. In
clearly in his writings
Uncommon Therapy, Haley not only coined the term "strategic ther-
apy" but attempted to create a model for therapy based on hyp-
notic techniques. 15 Many of these techniques are unobtrusive
ways of managing resistance. For instance, there is the technique
of "providing an illusion of alternatives": Would you like to go
into a trance now or later? The question of whether or not to go
into a trance, which is what the subject is really resisting, is

bypassed, and the subject has an illusion of choice. A therapist

might thus say to a resistant family: Would you like us to make a
home visit on Thursday or Friday? The question of whether or not
to visit is bypassed. An elaboration of this tactic is called "provid-
ing a worse alternative," and it consists of setting up two choices,
one of which is so dreadful or difficult that the client either thinks
up a different but equally effective solution on his own, or else
goes along with the less bad idea. In one case that a friend being
supervised by Haley described to me, the problem was a school-
phobic child, but a home visit showed that a psychotic sister was
living in the home, as well as a senile grandmother. The family
had resisted efforts by other agencies to get the child back to
school. Haley adopted the tactic known as the "Devil's Pact," in
which the therapist tells the family that he has a sure solution to
the problem, but the family must agree to do it before it is dis-
closed. The family fought the bargain but finally agreed. The ther-
apist then told them that as long as the child remained home from
school, the parents must disconnect every TV set in the house. As
not only the child but the psychotic sister and the grandmother
constantly watched TV, the family rejected this idea. They left,
angry with the therapist and threatening not to come back. How-
ever, some weeks later they called and told the therapist that they
had in fact disconnected the child's TV set, and had also en-

279
 Foundations of Family Therapy
rolled him in a school towhich they planned to drag him bodily.
In writing about strategic therapy, Haley stays mainly with pro-
cess language. After he decided to join Minuchin in Philadelphia
and started to develop his own clinical work, he also moved to a
different conceptual universe. He began to downplay the use of
hypnotic techniques and paradoxical directives (although by no
means abandoning his sense of their importance) and concentrated
on a more organizational model for therapy. Using his knowledge
of hierarchy and coalitions, he evolved his own method for dis-
rupting or changing abnormal family structures, as well as attend-
ing to the triadic configurations that accompany them.
What is peculiar is that the two worlds depicted in Haley's Uncom-
—
mon Therapy (1973) and his Problem -Solving Therapy (1977) the book
that came out of the Philadelphia years —are so far apart. It is as
though Haley makes a strange leap, from one side of Bateson's
zigzag to the other, from process to form. Most of Haley's career,
in fact, can be seen as an oscillation from one side of this zigzag to
the other, from the early microstudies of schizophrenic communica-
tion to the research on coalitions in families, to the development of
a strategic model for therapy, to an interest in a more structural
model, and so forth.
In Problem-Solving Therapy, Haley makes the point very strongly
that one must identify sequences of behavior that circle around a
problem, not just concentrate on the problem alone. 16 Here, of
course, he is joined by the strategic therapists Watzlawick, Weak-
land, and Fisch. But where they point out that most "problems"
consist of self-reinforcing cycles, Haley describes these cycles in
terms of family organization, laying out "problem sequences"
which may involve a mother, father, and child; or a grandmother,
mother, and child; or the therapist, parent, and child; or parental
child, mother, and child. Unlike the Palo Alto group, Haley thinks
of therapy in terms of a step-by-step change in the way the fam-
ily is organized, so that it goes from one type of abnormal organi-

zation to another type before a more normal organization is finally

achieved. By then, presumably, the symptom is no longer neces-
sary.
These two ideas, of tracking organizational sequences in assessing
the problem, and going through stages in the process of changing

280
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
it, are perhaps Haley'smost distinctive contributions to theory of
therapy. His emphasis on appropriate hierarchical lines, although
not his particular invention, should also be mentioned. This empha-
extends to an awareness that therapists and other professionals
sis

may be contributing to organizational abnormality by crossing

these lines in their efforts to help.
A nice example of a way of dealing with the latter problem comes
from Peggy Penn, who studied with Haley and his wife, Cloe Ma-
danes, in 1978 at the Family Therapy Institute in Washington. A
mother came into therapy because of her retarded daughter's rock-
ing behavior at school. This behavior made the teacher angry with
the child and her family, and the mother felt especially helpless
with this teacher. Madanes, the supervisor, had the teacher come
into the session and helped the mother show the teacher how to
"supervise" the child's rocking at school. This put the mother hie-
rarchically above not only the child but also the teacher, just as the
therapist was above the mother, and the supervisor above her. All
status levels were thus respected.
Haley has included in Problem-Solving Therapy the transcript of one
of the minidocumentaries he made with Braulio Montalvo while
both were at the Philadelphia Child Guidance Clinic. This video-
tape, called "A Modern Little Hans," contains a clear expression of
the idea of therapeutic stages; it is also an outstanding example of
therapeutic art. In this case of a six-year-old boy with a phobia
about dogs, previous attempts to treat the problem with individual
psychotherapy had been unsuccessful. The strategy devised by the
supervisor (Haley) for the therapist (Mariano Barragan) to carry out
was a charming one: to ask the boy to find a puppy that was afraid
of humans and to "cure" it. The treatment was a metaphor exactly
approximating the problem, but placing the boy in a reverse posi-
tion, so that by following the therapist's instructions, he would
have to stop being afraid of dogs. And indeed, in teaching the puppy
not to fear a boy, the boy conquered his own fear.
The strategy was also designed to accomplish a structural shift in
the organization of the family. Haley, as we remember, had ob-
served that when a child in a family had a problem, one parent
would seem very upset and would be alternately exasperated and
forgiving while the other parent would seem far less concerned.

28l
 Foundations of Family Therapy
"Haley's Triad" is as common a constellation in families with prob-
lem children as the Big Dipper is in northern skies.
Haley considers it important to interrupt this formation or to shift
it, and describes several ways of doing so. One is to disrupt the more

intense parent-child dyad by encouraging the overinvolved parent

to redouble the involvement in the hope that this will bring about
a recoil. A second way is to focus on the parental dyad and tease
out the parents' differences about the child's behavior. In so doing
the therapist inserts himself into the triangle with the parents, re-
placing the child, who is often serving as a covert battleground for
marital issues anyway. The third way is to enter through the periph-
eral parent's relationship with the child. This can be done either by
making this parent the disciplinarian, thus disturbing the covert
alignment with the child, or by giving the child and parent a task
to do together, making the alignment overt. This last tactic can,
however, have the effect of distancing the overinvolved parent and
unbalancing the marriage.
A simple step-stage model for therapy with two parents (or any
executive dyad) and a child became a cornerstone for Haley's think-
ing about therapy. It has resurfaced in Leaving Home, a book that
contains Haley's more recent ideas for dealing with what he calls

"crazy young people" — adolescents undergoing a first psychotic

17
break. Here Haley recommends that the parents be encouraged to
on the adolescent's behavior: the approach is one version
set limits
of "going through the parental dyad." If the parents do set limits,
the child will usually improve. If they cannot, the therapist engages
them with himself in a struggle to resolve their differences over the
adolescent child's behavior. During this process, Haley observes,
they are often metaphorically dealing with marital differences. In
this case too, once the child is disengaged from the struggle, he or
she will improve.
Haley is an artist at making the complex seem simple. His recipes
for creating shifts in triads in easy, geometric sequence have helped
many clinicians extricate themselves from mounds of useless data
and therapeutic sprawl. His cookbook-style outlines for therapy
display a sound respect for principles of good organization and
introduce the idea that persuasiveness alone is not the whole story

282
ECOLOGICAL, STRUCTURAL, AND STRATEGIC APPROACHES
in bringing about change. would be interesting, though, if Haley's
It

concern with structure and organization were to double back to

include his original fascination with strategic maneuvers, especially
those that have been subsumed under the rubric of paradoxical
interventions. One hopes, in fact, that he will take another "strange
leap."

283
Chapter 15

The
Systemic Model

The Quiet Revolution in Milan

In 1968, the year Jackson died, the ideas of the Bateson group
leaped across an ocean and took root in Italian soil. Mara Selvini
Palazzoli, a child analyst, had been working for many years with
anorectic children. Discouraged by her results, and impressed by the
family therapy literature that was coming out of Palo Alto, she
decided to discard all elements of psychoanalytic thought and adopt
a purely systems orientation.
In thatkey year, she organized the Institute for Family Studies in
Milan. After an initial process of winnowing, the group narrowed
down to four psychiatrists: Luigi Boscolo, Giuliana Prata, Gian-
franco Cecchin, and Selvini herself. This group, working together
over some ten years, developed a family systems approach that they
used not only with families of anorectics, but with families of chil-
dren who had serious emotional disorders.
Selvini's first book, Self-Starvation, published in 1974 in the United
States, documents her therapeutic trajectory. It is only the last part

284
THE SYSTEMIC MODEL
that describes her shiftfrom the analytic model to the circular,
cybernetic epistemology of the Bateson group and to working with
families. 1 A second book, Paradox and Counterparadox, published here
in 1978, was written by the Milan Associates (as they now call
themselves), and is currently the most comprehensive description of
2
their work and methods.
The Milan Associates, although influenced by the Palo Alto
group, evolved in quite a different direction, creating a form unique
and distinct enough to be considered a school of its own. In Europe,
where the approach has stirred up much interest, the term "sys-
temic" is used to describe it. From the beginning, the group used an
unusual format. They work (or did, when Paradox and Counterparadox
was published) as a foursome, with a man and a woman therapist
in the room with the family and a man and a woman behind a
one-way screen.* Periodically the observers may summon one of
the therapists out of the room to offer a suggestion or ask for more
information. Toward the end of the session, the therapists break for
a consultation with the observers, during which time all four share
opinions and come up with an intervention or a recommendation.
This may be a ritual, a task, or a prescription. It comes from the
whole team and is shared with every member of the family. If it is
in letter form, each person in the family is given or sent a copy.
Sometimes, if an important family member has failed to come to a
session, he or she will be sent a copy of a letter that may address
his absence.
From the beginning this group has attempted to prevent their
approach from being based on factors of personality or charisma.
For this reason, they change partners from family to family. Ther-
apy begins with the first and much attention is paid to
phone call,

details like who made the call, tone of voice, and attempts to control
the conditions of treatment. The entire household is required to be
present in a first session. In later sessions, the team may decide to
see different units. Information sought during the initial phone call
is minimal: who called; who is in the immediate family or house-

hold; who referred the family; what is the problem; and of course

* More recently, only one therapist conducts the interview, and one to three may be behind
the screen.

285
 Foundations of Family Therapy
items like address and date of call. Information must also be gotten
from the referring professional.
Before each session the team meets to discuss the previous session
or, in the case of a first interview, to examine the family intake. The
sessions last about an hour, and during that time the team not only
asks for information but carefully notes nonverbal communications.
The team discussion is conducted in a special room; and at the end,
the two therapists rejoin the family to present the recommendations
of the team.
Treatment usually consists of about ten sessions at monthly in-
tervals or longer. This practice first began in order to accommodate
families who it was then decided that this
lived far away, but
relatively long time lapse between sessions was favorable for ther-
apy with families with psychotic members. In an important article,
"Why a Long Interval Between Sessions," Selvini connects this
practice to the nature of families with schizophrenics, and to their
resemblance to the too richly cross-joined systems of Ashby, al-
ready described in these pages. 3 Each family has its own time span
for processing a complex set of information: and the more richly
joined the system, the longer the time it will take for this process
to come to rest.
Calls and attempts to schedule earlier sessions are handled by the
team as responses trying to undo the effects of a given intervention.
They are treated with careand respect in the sense that if a family
goes into crisis after a session, the team will be especially careful to
avoid any move that might stabilize the system and negate potential
for change. Thus they will tend not to give in to requests for extra
sessions and will respond calmly to reports of dire emergencies, in
the belief that this is the best possible indication that change is
taking place.
Obviously and good team
this stance requires nerves of steel
support. On one husband was
occasion a wife called to say that her
so depressed that he was threatening to cut off his penis, and she
asked for an earlier session. The team, sensing that the wife was
making a bid to control the treatment and that the husband was not
in any imminent danger, told her that this extreme anxiety on her
part was a predictable reaction, foreseen by the team, but that the
session would be held as scheduled.

286
THE SYSTEMIC MODEL
The Milan Associates call this treatment a "long, brief therapy/'
because the number of hours with the family is small but the length
of time needed for family reorganization can be very long. Each
session is videotaped, and notes are made of each session. Follow-up
studies have not been routine in the past but are planned for the
future.

The Counterparadox

A question that was raised in Palo Alto in the 1950s, and has been
haunting the family field ever since, was how to use the discovery
of the part played by double-level communications in the family of
the schizophrenic. The Bateson group, experimenting with "thera-
peutic double binds/' reasoned that it would have to use the same
type of paradoxical communication with the family that the family
itself was using. The Milan Associates, taking the same position,
elaborated on the idea of the therapeutic double bind, called by
them a "counterparadox,"and used it as the cornerstone for an
intricate, elegant, and logical methodology for change. In Paradox and

Counterparadox the Milan group states:

As far as paradoxes are concerned, we can say that our research has shown
how the family in schizophrenic transaction sustains its game through an
intricacy of paradoxes which can only be undone by counterparadoxes in
the context of therapy. 4

A good portion of this book is devoted to a discussion of the ideas

of Bateson, Haley, Watzlawick, Weakland, and other contributors
to what is rightly described as not just a movement in the field of
mental health but a much larger epistemological shift necessitating
a new approach to human behavior and a new language for describ-
ing it. Perhaps more than any other clinician-researchers, the Milan
Associates have used this epistemological shift as a base for their
approach.
Central to their thinking is the Batesonian concept of circular
causality that has been discussed in previous chapters. Along with

287
 Foundations of Family Therapy
this goes a conscious distrust of being caught in the traps of "linear
thinking" — the illusion, peculiar to our Aristotelian heritage, that
there is a historical causality in which A causes B, which then causes
C, and so on. These traps contribute to — are even part of —the
dilemmas the clinician must deal with, and at the same time aug-
ment his most frequent clinical mistakes. A familiar example would
be the position taken by the family therapist who prides himself on
realizing that the child is the victim of a "dysfunctional" family
system. The therapist sympathizes with the child as the scapegoat
for the unexpressed hostility between the parents and immediately
tries to pronounce the child blameless and moves to the dysfunc-

tional marriage as the "real" cause of the child's problems. Not only
is this an extremely linear view, but it often provokes resistance and

reduces therapeutic effectiveness.

A systemic approach, the Milan Associates make clear, involves
abandoning these notions and realizing that the enemy the clinician
must attack is not any family member or even the malfunctioning
family itself, but what they call the family "game." The way they
describe this game recalls elements of Haley's control theory of
communication in families of schizophrenics, whereby each person
tries to win control of family rules while denying that he or she is

really doing so. Unless all can agree on the rules of the family game,
there can be no winning it, of course, and no finishing it; in an
eternal cycle, the game about the game, or the metagame, goes on
and on.
Since such games are not overt, one can only infer them from
communications that go on in the family, but here is a good exam-
ple, quoted by Selvini in Self-Starvation and related to the struggles
about leadership that are disqualified even while they are going on:

mother: I don't let her wear miniskirts because I know her father
doesn't like them.
father: I have always backed my wife up. I feel it would be wrong to
5
contradict her.

It has been the genius of the Milan team that they have devised a
method for breaking such games-without-end. First, of course, one
would have to expect that the therapist would not be exempt from
the control moves of family members. Any attempt to get the family

288
THE SYSTEMIC MODEL
to do something different will automatically call forth counter-
moves and disqualifications. Therefore, the first step in therapy
would be to establish what game the family is playing with the
therapist, and to agree with the game and encourage it. The game
is usually: Here is our burdensome, sick, or bad person, fix him or

her and relieve us, but do not make us change. The therapist knows
that to fall into the trap of trying to do this will only lead to his own
downfall.
The opposite move, then, would be symptomatic mem-
to ask the
ber to continue with the problem, rather than try to fix it. But this
is nothing new. Clinicians doing individual therapy have been using

"reverse psychology" or similar tactics for years, and the strategic

school pioneered by Watzlawick, Weakland, and Fisch has elevated
to a high art the technique of prescribing the symptom. What is
different about the Milan Associates is their insistence on prescrib-
ing not just the problem behavior or set of behaviors but the larger
configuration of relationships surrounding the problem. To under-
stand this, we must examine their concept of the "positive connota-
tion," linked closely both to the development of a systemic hypoth-
esis and to their interventions.

The Positive Connotation

The is a therapeutic device that may be one

positive connotation
of the Milan group's most original inventions. Initially they wished
to give a rationale that would be consistent with encouraging a
symptomatic behavior. Since, by taking the family into therapy,
they had implicitly agreed to help the family get rid of the problem,
it would be inconsistent simply to prescribe it without giving a good

reason. In this, they are addressing the necessity, also recognized by

the strategic group in Palo Alto, of "refraining" a situation so that
this type of intervention seems logical.
One would be to say that the symptom of the patient
possibility
was in some way required by the family; that the family "needed"
a sick person. But to do this would go against the prohibition against

289
 Foundations of Family Therapy
linear causality. It isno more correct to blame the rest of the family
and praise the sick one than vice versa. The solution to this puzzle
would be to connote positively all the behaviors in the family that
pertain to the symptom:

It thus became clear that access to the systemic model was possible only
if we were to make a positive connotation of both the symptom of the
identified patientand the symptomatic behaviors of the others, saying, for
example, that all the observable behaviors of the group as a whole ap-
peared to be inspired by the common goal of preserving the cohesion of
the family group. 6

In effect, one cannot disentangle the positive connotation from

the intervention, usually a paradoxical prescription, in which it is

embedded. The positive reframing of the symptom as it is linked to

other behaviors in the family is the core of a paradoxical prescrip-
tion. Therefore, to explain one, we have to explain both. This is not
easy to do. It is probably best simply to describe a particular exam-
ple.

This case was a consultation (first interview, really) with a family

the Milan Associates saw during a demonstration workshop. Peter,
seventeen, had been briefly hospitalized with an acute nervous col-
lapse, partly brought on, it by the use of LSD. He had been
was felt,

diagnosed schizophrenic although he seemed quite

at the hospital,
lucid in the session and kept weeping, which is not a classical sign
of schizophrenia. The therapists found out that each parent had
previously been married to an abusive and irresponsible spouse; the
mother, in fact, had been driven to thoughts of suicide before she
decided to divorce. The children came from these previous mar-
riages. Mother had an eighteen-year-old son, Anthony; the patient,
Peter; and a daughter, Sarah, fifteen. Anthony was about to leave
for college, and Peter had apparently been very close to him. The
father's two children were Linda, twelve, and Debbie, fourteen.
According to the parents, not only Peter but Debbie was a ''wor-
rier. " Linda, like Anthony and Sarah, supposedly had no problems.

Although the parents seemed to have a good marriage, the mother

was clearly the less assertive partner, and signaled her fragility by
breaking down and crying when she recounted the hideous brutali-
ties of her former husband.

290
THE SYSTEMIC MODEL
It was impending departure of the oldest son might
clear that the
be triggering some The mother had leaned consid-
of Peter's upset.
erably on Anthony, and it seemed that Sarah, the mother's third
child, was now coparenting with the stepfather. A change in the
balance between the parents was clearly in the offing.
The team's intervention was slanted to do several things: first, to
normalize Peter's role; second, to push Sarah into the child group,
drawing a proper generation line; and finally, to counteract the
effect of Mother's tears, which amounted to a message to the chil-
dren, or at least to Anthony: Don't leave. The following ritual was
suggested by the therapists, Selvini and Cecchin, in their message
to the family:

You, father, and you, mother, had a tragic and disastrous experience in
your first marriages. Each of you married the other to give a good parent
to your children. And you, children, are working very hard in the service
of your parents' wish to be perceived as good parents, and are trying hard
to help them maintain this conviction. Anthony and Sarah, also Linda, are
showing how good their parents are by their perfect behavior. But Peter
and Debbie wonder whether it would be better to be perfect or to be a
problem. If they are problems, this helps the parents even more to show
what good parents they are. Not knowing which is best, perhaps the
children should meet in a week to discuss how they should continue this
work of helping their parents, who have this understandable need because
of their tragic life. Peter, it is you who must call the meeting. If you

children feel you cannot succeed, you have to phone this clinic and ask for
help/

The reactions were immediate and revealing. Peter looked startled

but pleased on being asked to convene the children's meeting. Linda
and Debbie brightened up. Sarah seemed less happy. Mother looked
upset, and father perplexed. But it was the self-controlled Anthony
who surprised everybody by throwing his arms around Selvini and
bursting into tears, as if to say, "You finally understood the spot
we're in."
It is hard to make claims for interventions like these, especially
ifone hasn't had first-hand experience with them; and in this in-
stance, there is no follow-up. It is possible that the family never

This message was taken down verbatim by the author while watching the team interview
the family during 1979.

291
 Foundations of Family Therapy
came back. Perhaps it did not need to come back. But it is clear that
such prescriptions can touch and stir a family. An intervention of
this sort will —
sometimes only temporarily but sometimes forever
— break a fixed family pattern. At the very least it will break the
united front, or united story, a family presents to the therapist. One
member will seem angry, another puzzled, another worried, while
another may 7
say, "I understand perfectly/ In this case, reversals of
what the family presented in the interview were noticeable. The
two most worried children seemed happy, while the apparently
"strong" oldest son broke into tears. In particular, the supposed
fragility of the parents, especially the mother, was called into ques-
tion by the message.
The effect of this prescription would certainly be to shake up or
revise options for relative positions in the family. Linking Peter
with the other children erased his special status as the sick one, and
placed the "worrisome" children on the same level with the "re-
sponsible" children. Giving Peter the task of convening the meeting
made him special again, but in a positive way.
Most important, the children were enjoined to continue their jobs
of parenting their parents. It is true, as Madanes says in a recent
article on paradoxical prescriptions, that the outcome is often to
change the family structure. 7 The preceding example prescribes
what Madanes refers to in her article as the "incongruous hierar-
chy" one almost always finds in families with symptomatic mem-
bers. If there is a recoil, we may hope it will be in the direction of
more appropriate boundaries and proper status lines.
Another Milan group used here is to put the therapists
tactic the

who may be working with the family in a one-down position to the

children. This makes all the adults lower than all the children. It is

another example of prescribing an incongruous hierarchy, but the

professional context is seemed likely
included as well. In this case, it

that any therapist assigned to the family (which had been referred
to the outpatient clinic by the hospital where Peter was treated)
might unite with hospital and family to continue to see Peter as
"crazy." The message would tend to challenge any hospital/
parents/therapist agreement along that line. It would also alert a
therapist not to form a coalition with the parents, but to claim an
appropriate hierarchical position above parents and children both.

292
THE SYSTEMIC MODEL
What is beginning to be clear, however, is the importance of
"reading" the internal (and external) politics of the family. One
must study the coalitions and apparent power balances or imbal-
ances in relation to the symptomatic behavior. This is why the most
important contribution of the Milan group may not be their most
visible signature, the systemic paradox, but their detective work in
devising a hypothesis that will explain the symptom in the family

and how all the pieces fit.

The Systemic Hypothesis

In their article "Hypothesizing —Circularity—Neutrality," the

Milan Associates state that a hypothesis must be circular and rela-
tional 8 —by which they mean that it will organize all the confusing
data attached to a symptom so as to make sense in the relationship
context of the family. They
cite as an example an interview with a

divorced mother and her adolescent son. The two came for therapy
because of constant fights. At first the team entertained the notion
that the boy's behavior might, in part, be a disguised effort to bring
the natural father back into the picture. Questions along this line
fell very flat, however, so present circumstances were investigated
and a new hypothesis was devised. It turned out that the mother
was seriously dating another man, for the first time since the two-
some had begun their life together twelve years before, at the time
of the divorce. The son was also at the age when he was seeing more
friends. The mother-son couple were beginning to break up, with
consequent distress.
This time the team's message was based on a simple hypothesis:
that thetwo were undergoing a natural process of growing apart and
making new ties, a process that brought with it unavoidable grow-
ing pains but was nevertheless "irreversible." The team therefore
recommended that the couple come in, not for therapy, but for a
"few meetings," to try to slow down this painful but irreversible
process of separation.
The issue arises: Is there then one true hypothesis? Obviously

293
 Foundations of Family Therapy
some more "true" than others, as this case shows. The Milan
are
group handles this issue by citing the Oxford English Dictionary, which
defines a hypothesis as "a supposition made as a basis for reasoning,
without reference to its truth, as a starting point for an investiga-
tion." This immediately sets an intriguing framework around ther-
apy — each case becomes an experiment of its own, a real-life mys-

tery novel. But there no one "solution" to this kind of mystery.

is

One comes out of the dilemma with a Pirandellian notion of the

"truth": There are as many possibilities of the truth as there are
places to stand and look at it from.
Whether it is "workable," in the sense of a supposition on which
to base an experiment, can be judged only in retrospect and even
then inaccurately. By the time a hypothesis seems justified by the
course of events, the family will present a different configuration,
which means that the original hypothesis must be revised or even
totally scrapped. One suspects, however, that a sufficientlycomplex
hypothesis will stand the test of time and will form a core
at least

for the picture that begins to appear as family and team move
through changes together.
A hypothesis does two important things. First, the hypothesis is

useful in its "power of organization." It not only offers a rough

scaffolding on which to hang the masses of information thrown out
by a family, but can give the therapist a thread to follow in conduct-
ing an interview, thus blocking out the meaningless chatter that
consumes so much of the usual session. Second, it suggests what
meaning the symptomatic behavior has in this family at this time.
In the case of the boy and the mother, it was clearly not a sarcasm
to tell them that they needed help to slow down the process of
growing apart. The problems they were having did seem to be
intensified by their efforts at separation.
It is linear, however, to say that the hypothesis defines the "func-
tion" of thesymptom. In Self-Starvation Selvini points out that the
members of a family become "so many elements in which no one
element can be in unilateral control over the rest. It would thus be
epistemologically incorrect to say that the behavior of one person
one cannot say that a symp-
7
'causes that of another." 9 As a result,
tom is caused by the family's reactions to it, any more than vice
versa; rather, all these behaviors are circling around in a mutually

294
THE SYSTEMIC MODEL
supportive arrangement. One has to see a process in which activities
dovetail with each other as rhythmically as the inhaling and exhal-
ing of breath, or the systole and diastole of the heart.
What goes into an intervention or a prescription is never com-
pletely the same as a hypothesis. The hypothesis respects the cir-
cularity of family events as far as possible. When translated to a
prescription, a linear epistemology is unavoidably adopted. This
linear interpretation of the data presented by the family usually
reverses the family's version, introducing a new "punctuation" into
the family. The family may say: "So-and-so is to blame for our
misery by -his insensitive behavior." The team says, "We see things
differently. We see your son not as insensitive but as extremely
sensitive." What will follow is an explanation of his distressing or
destructive behavior as crucial to the welfare of someone, or as
confirming the unity of the family, or as a solution to a dilemma
brought about by some shift in the family.

Is this way, yes; in a way, no. I prefer to replace the

linear? In a
concept of paradox with that of polarity. In the I-Ching, or Book of
Changes, the meaning of each hexagram is modified by the inclusion
of an opposing possibility. In the same way, by replacing the linear
punctuation of the family with an opposite one that is equally
linear, the Milan group creates a polarity. The essence of polarity
is an interpretation that moves from one pole to another, neither
true in itself but only in combination with the other, and always
suggesting other, unforeseen possibilities which are never spelled
out. When a family responds to a reverse punctuation with a rejec-
tion not only of the punctuation but of the behaviors it describes,
and discovers a completely different way to organize relationships,
one feels that this method of therapy could indeed be called a
dialectic of polarities.

The Uses of Time

A crucial aspect of the way the Milan therapists develop a hy-

pothesis is their attention to time. They are concerned with the way

295
 Foundations of Family Therapy
a family evolves new patterns in adapting to changing circum-
stances. A behavior, no matter how senseless or destructive, is al-

ways in some sense a solution. A dilemma arose at some point in

the family trajectory when the natural processes of growth or an
accidental shift required a change in family organization. A symp-
tom can be a solution of sorts. A good hypothesis will often describe
a symptom, or any irrational behavior, as an ingenious solution to
the difficulties faced by the family on its evolutionary path.
An example is another North American family the team saw once
for consultation; the family's problem was a beautiful, promiscuous
daughter of twenty. The family consisted of the girl, her parents,
and a thirty-year-old half-brother born of the mother's liaison be-
fore she met the father.
team noted the constant twinning between
In the interview the
mother and son, who both seemed very melancholy and even wept
in tandem. By contrast, father and daughter were loud and lively;
they fought constantly but in an amiable, affectionate manner. The
father and mother were distant in the session, but the children
reported that the father would often criticize the mother, who
would then cry. The problem dated from the time of the son's return
from the Vietnam War. Quarrels and fist fights between son and
stepfather led to his banishment to an apartment below, where he
lived like a hermit. Soon afterward the daughter, then in her early
teens, began to go out with men, with the apparent knowledge of
the mother, who deplored her behavior on the grounds of safety but
insisted on knowing all details. If the father tried to scold or control
the girl, the mother protected her.
The hypothesis of the team was that the girl's behavior kept
dangerous couples apart at the same time that it prevented them
from separating. The girl's behavior distanced the girl from her
father because of the quarrels it set up; it distanced her mother from
the half-brother because her mother was so preoccupied with her;
it distanced the mother from the father because they kept fighting
about how to manage her; and it distanced the two men because the
mother's preoccupation with the girl kept the son out of a rivalry
with his stepfather that would otherwise have surfaced. The behav-
ior also kept the family stuck together in their attempts to deal with
it.

296
THE SYSTEMIC MODEL
The team's prescription addressed the evolution of the two coali-

tions, mother/son and father/daughter —defining the daughter's

behavior as a solution to the father's original position vis a vis his
stepson and wife. The prescription went as follows (this is a verba-
tim transcript of the message, taken from a videotape of the inter-
view with the family):

dr. b[osco1o]: [to father] This is about the irresponsible

our opinion . . .

behavior of D. ... we feel instead that D.'s is a very responsible behavior

— that what she has been doing for all these years that you might feel was
irresponsible behavior was extremely responsible.
And why is she responsible? What she's been doing all these years, from
the age of twelve, fifteen till now, she's been doing all this for you, L.
your daughter has been doing everything for you. And what is she doing
for you? At the age of twelve, thirteen years old, when she would psycho-
logically be getting out of the house, she felt that if she really would get
psychologically, emotionally, out of the house, she would have left an
intolerable situation at home for you, because she saw that mother and S.
were the privileged couple. Mother and S. are very close together. And D.
felt that if she got out of the house she would have left you alone. So, at

a certain point, she herself decided, "I have to do something to help my

father to be present in the house." And we have seen today that she
succeeds in making you present in the house, in having rapport in the
—
house, with your wife, with her otherwise, you would have been com-
pletely cut off from the family, you would be completely out. So we feel
that D. has been doing all this for you.
father: And that includes her sexual behavior, and all that?
dr. b.: Including everything.
father: Think so? I don't know — I find it very difficult to believe.
dr. s[elvini]: Because her affection is for you —she is separating the
privileged, first couple, mother and S., and gives you the occasion of
quarreling so warmly with her.
father: Well, if she had done it to help, she certainly did not. If I had
got my hands on her I certainly would have helped her.
mother: D. said many times, "My father never embraces me, and never
even gives me a kiss."
father: [to D.] See that, you never give me a kiss!
mother: She missed that.
father: Because she's so fresh all the time.
dr. s.: understand. He is not grateful because he has not understood
I

what she is doing for him. I understand she is very sad because Father has
not understood what she has done for him.
dr. b.: Certainly, what D. has done for her father, as we said before,

297
 Foundations of Family Therapy
is so that he can be present in the family and would not be cut off; she is
a sacrifice; she gives up her adolescence to go out and lead a certain kind
of life.

dr.s.: —
She is doing like her brother maybe forever. But we have seen
in our experience in Milan many, many, many beautiful young girls in this
situation do the same for the father. So [rising to go]. . . .

father: So you mean I'm the bad guy. [Laughs.]

daughter: Depends how you see yourself. [Gets up.]
father: I don't see myself as anything.
dr. b.: you have really been doing a lot for your father
[standing] D.,
— the you make for him so he can be present in the family
sacrifice that
—and this can go on for all your life, as we see in many cases . . .

dr. s.: [talking over] You insist on this —

not to be excluded from the
rapport between mother and S. . . .

dr. b.: ... is our conclusion. [Farewells all around.]

While Selvini is speaking of the sacrifice made by beautiful girls in

Milan, the daughter with great dignity, and sweeps out as
is rising
if quite offended. The father seems upset too, although he laughs.
The thrust of the message may have been to bring down the
7
current "privileged couple/ father and daughter, and to create a
split between them. This apparently succeeded, for the next session
began with the daughter announcing that she had obtained a volun-
teer job at a local settlement house working with teenage girls. As
she became more involved in her own life and career, the prediction
implied by the message came true. After two more sessions, the
father came in complaining that his wife and stepson were together
all the time and that he had come to blows with his stepson. He felt

so displaced and insulted that he had decided to leave home. The

Milan team's message was revived by the family's ongoing thera-
pist/ and the girl was advised to return to her previous behavior.
Naturally, she did not do so; and as both she and the son began to
become less involved with the parents, the couple began to com-
plain about the inadequacies of their marriage. The therapist dealt
with the situation with continued paradoxical prescriptions, and the
case came to a good conclusion.
It is interesting that the original hypothesis remained valid during

*The family consultation was held at the Ackerman Institute for Family Therapy, which
organized the Milan Associates' first visit to the United States in 1977. and the therapist was
Gillian Walker.

298
THE SYSTEMIC MODEL
the course of the case, even though it was added to and elaborated
upon. The girl's behavior was tied to an evolutionary impasse, when
she would normally have been leaving home. The mother's history
furnished even more evidence for a difficulty arising at that time (at
age thirteen the mother had been raped by her own stepfather and
sent away from home). The message, in effect, said to the family,
not that it was a dysfunctional family but that it had shown ingenu-
ity in solving a dilemma created by its own history. The fact that
it failed to evolve to a next-stage level is not criticized, except
indirectly, by the overzealous way in which the team expresses
admiration and praise for the ingenious solution. The hypothesis
was justified by the fact that after the daughter improved, the
family moved back to the place where the son, returning from the
war, had fought the father for possession of the mother. What
happened with the help of therapy was that the family arrived at
a more "evolved" solution, one that allowed the crucial separations
to take place. Son and daughter became more independent, while
the couple took a memorable trip to Europe together, a thing they
had never before allowed themselves to do.

The Referring Context

The Milan Associates see family and therapist as embedded in a

larger context or field, and take the entire field as the unit of treat-
ment. They pay strict what could be called the "outer
attention to
ring," the entourage of professionals and institutions that may be
heavily influencing the family in its management of the patient. If

therapy produces a crisis —which it often does before a change takes

place in very rigid family systems —the patient may appear to dete-
riorate.The change may then be aborted by the family's move to
rehospitalize the patient or find someone to dose him with massive
medication. Gillian Walker's term for a professional who takes this
role is "Dr. Homeostat," because this person acts to restabilize the
field so that the symptom remains intact.
In their article "The Problem of the Referring Person," the Milan

299
 Foundations of Family Therapy
Associates describe their efforts to counteract the influence of this
kind of person, who is often the one who referred the family to
therapy and who may have an emotional stake in the outcome of
treatment. 10 Often the team will ask such a person to attend the
session. They add: "We no longer make the mistake of advising or prescribing
" ll
the interruption of the relationship between the family and the referring person.

Instead, they simply prescribe the situation. The family is told that
it must not hazard any movement toward the change it desires
because if the symptom were no longer present, the family (or some
particular family member) would lose an important friend/ally/
comforter. Alternatively, the team will prescribe the presence of the
professional as essential to maintaining equilibrium and preventing
a premature change.
The Milan Associates are clearer than any other group about
giving priority to contextual issues of treatment, especially those
pertaining to the professional field. If the family refuses to come to
therapy at the time set, or if a member refuses to attend, this will
take priority for the team, no matter how serious a problem the
family may be presenting. They will either postpone therapy until
the family agrees to their terms, or will address the issue in the
intervention, which will usually prescribe that the family continue
to behave in that particular way to prevent change. Sometimes, as
we see, the question of the interfering professional may be the total
focus of the intervention. This stance insures freedom of movement
and is in part responsible for the extraordinary leverage the group
maintains at all times.

Circular Questioning

For conducting the session, the Milan Associates have recently

developed a format that is based on Bateson's statement that "infor-

mation is a difference/' and which they refer to as the technique of

circular questioning. The team's article "Hypothesizing —Cir-
cularity —Neutrality" contains a good description of this technique
12 augment powerfully the
and its rationale. The method seems to

JOO
THE SYSTEMIC MODEL
amount and quality of information that comes out in an interview.
The basic tenet is always to ask questions that address a difference
or define a relationship. Asking someone to comment on his parent's
marriage, or to rank family members on the basis of who has suff-
ered the most from someone's death, or to rate, on a scale from one
to ten, his mother's and then his father's anger when his sister
comes home late at night, are all "difference" questions. So are
questions that deal with before and after: asking a child by what
percentage the fights between the parents have diminished since the
older brother was hospitalized, or posing "hypothetical" questions
such as, "If you had not been born, what do you think your parents'
marriage would be like by now?" or "If your parents were to di-
vorce, which child would go to which parent?"
Using method, one notices several things. First of
this all, such
questions make people stop and think, rather than react in a stereo-
typed way. The people who are not talking also listen attentively.
Second, these questions cut into escalations and fights, not only
between family members but between a therapist and family mem-
bers. And, third, they seem to trigger more of the same kind of
"difference" thinking, which is essentially circular because it in-
troduces the idea of links made up of shifting perspectives. The
Milan Associates point out that in families in schizophrenic transac-
tion people seldom define a relationship or notice a difference, and
that this technique used alone may have a powerful effect on these
families.
The questions may have a cumulative effect. One might ask a
wife what kind of relationship the husband had with his mother,
and then ask him the same question about her and her mother. This
cross-referencing of information can be revealing and can lead to
even more revelations. In addition, the therapist can use this tech-
nique to ask quite heavy questions without the usual constraints,
since he is only getting the opinions of others. The Milan Associates
will ask even young children what their opinion is of their parents'
sex life. Since children always have an opinion, this does not in fact
make them privy to information they should not have. And despite
the horror of family therapists who have been drilled to make each
family member talk only for him or herself, my sense is that in an
indirect way these questions push people to differentiate quite as

301
 Foundations of Family Therapy
much as asking people directly to do so. For instance, the usual
response of parents and children to the question about sex seems to
reinforce the generation line, not blur it.
Another use for these questions is that they can be used to block
behaviors by simply pointing to them. If a mother has a "death
phobia/' the team may ask the father: "What do you think would
be the effect on the family if Mother died?" This puts the "worst
case" on the table and takes away from the death phobia some of
its old power to upset others. In the case of an attempted suicide,

for instance, one might ask: "If X had managed to kill herself, who
in the family would be the And so on.
last to forget her?"
In general, it seems that the Milan Associates now have an elegant
model for conducting the interview that is congruent with their
Batesonian philosophy of treatment. Both the interviewing tech-
niques and the systemic intervention at the end insert punctuations
that emphasize difference and circularity. The questions reinforce
and are reinforced by the prescriptions derived from them in a
manner that makes the entire interview an example of circularity at
a more complex level than if either technique were to be used alone.

The Importance of "Neutrality"

If one could say that there is any one stamp or signature that
characterizes the Milan Associates' approach, I would not choose
the ingenious interventions, the elegant interviewing style, or the
meticulous care with which a hypothesis is made. I would say it is

their entire stance, summed up in their concept of "neutrality."

Again, the reader is referred to the article "Hypothesizing —Cir-
cularity — Neutrality." This short article is like concentrated space

food: it contains in brief some of the group's best ideas.

"Neutrality," despite its hands-off implications, has more to do
with effectiveness in therapy than with staying cool. The team
maintains an impassive, if respectful, attitude during the interview,
in contrast to the sociability adopted by many other schools and
practitioners. However, the Milan Associates are only too aware of

302
THE SYSTEMIC MODEL
the power of families to render therapists impotent, and they place
techniques for maintaining leverage above any other pragmatic
achievement. To this end, they have adopted a number of devices
that help the therapist or the team to stay in a position from which
maximum change can be achieved.
In its simplest form, the Milan Associates describe ''neutrality" as
the ability to escape alliances with family members, to avoid moral
judgments, to resist all linear traps and entanglements. If, for in-
stance, no one member can say after a session that either therapist
sided with him or her, "neutrality" has been achieved: "The thera-
pist can be effective only to the extent that he is able to obtain and
maintain a different level (metalevel) from that of the family." 13
I would add to this many of the other devices and methods the
team uses in its approach. For instance, what strikes an observer of
the Milan Associates is their deliberate use of mystery and drama.
The family knows that people are silently watching from behind a
screen, not just as observers but as active participants. The thera-
pists in the room come and go for mysterious reasons —sometimes
on impulse, sometimes in response to a knock at the door.
In addition, team members —
even those in the room cannot be —
influenced, because they are controlled by invisible others. One is
reminded of the old-fashioned analyst with his impassive position
behind the couch. Messages and letters that emerge from behind the
screen reinforce the notion of the one-way street. The attitude of
the therapists toward family responses, always moving with them
or remaining unmoved, rather than joining in outright battle, is also
a way to remain "neutral."
This position fits with the basic tenets of strategic therapy. Unlike
the therapist who pushes and pulls the family into shape, the
"bullfighter" therapist, one could call these practitioners "hen-
house" therapists: "Ain't nobody here but us chickens." These ther-
apists take a low profile, speak softly, and carry a tiny stick. Like
Judo experts, they use the momentum of the family's own resistance
to effect change.
The surprising thing, for those who begin to work this way, is the
power of the approach. I have sometimes called it the Therapy of
the Weak, since the force seems to come from the very negation of
using force. Shakespeare, in one of his most famous sonnets, de-

303
 Foundations of Family Therapy
scribes a lover too cool even to notice his own influence on the heart
of the Bard: "He power to hurt and will do none/ That
that hath
will not do that —
which he most doth show," is to paraphrase the
embittered and lovesick Shakespeare —a much tougher customer
than your everyday activist. In fact, he has all the cards in his hand.
"Neutrality" confers on the systemic therapist the power to be
effective. But the ingredients are many: the quiet, nonreactive
stance; the circular questioning, always placing the therapist at the
metalevel; the devices that prevent the therapist from being endan-
gered by family suction (the screen, the team, the messages, the
unexplained and unexpected words and actions of the therapists);
the concern with field and context issues in order of priority; and
finally, the implacable attitude toward resistance.The team will
sooner lose a family than on change.
insist
In ending, let me cover myself by saying that by the time this
book is out, this chapter may be obsolete. The Milan Associates'
work is always in process, always changing. The group is now
moving in very different directions than they did when Paradox and
Counterparadox was published. The two men, Boscolo and Cecchin,
are teaching in Milan and traveling extensively across Europe, Can-
ada, and the United States, giving workshops. Selvini and Prata are
also teaching, mainly now in Europe, as well as continuing with
their research. The group
coming up with very different clinical
is

instruments than before, and as they are now in many ways two
teams, not one, it is to be expected that differences will develop
between them, as well as between them and their ever more numer-
ous followers.
Whatever happens, the Milan Associates have given us not only
a pragmatic expression of a truly circular epistemology, but a new,
more tuned apparatus for assessing and working with difficult
finely
families. In addition, as the next chapters will show, this method
raises many novel and interesting clinical issues. And these issues
in turn lead us back to a more rigorous look at our theoretical
frameworks and the epistemologies that support them.

304
Chapter 16

Theories About
Therapeutic Binds

The Puzzle of the Paradox

The work and Milan team have become the base

theories of the
for a new round of experimental thought about behavior and
change. It would be helpful at this point to look at some of the
theories about why counterparadoxes or therapeutic double binds
work. We have seen that in Pragmatics of Human Communication, the
term ''paradoxical intervention" is used as a synonym for the thera-
peutic double bind. 1 In addition, "prescribing the symptom" was
considered a form of paradoxical intervention and was first used by
members of the Bateson group in various papers describing both the
paradoxes people imposed on one another and those the therapist
devised to combat them.*
In a therapeutic double bind, as we have seen, the client is told
not to change in a context where he has come expecting to be helped
to change. If he resists the injunction, he changes; if he does not

"The reader is urged to go to Richard Rabkin's Strategic Psychotherapy (New York: Basic
Books, 1977) for a history of this idea, and an account of how therapists of various types
at various periods stumbled on its effectiveness and invented terms for it like "negative
practice," "reverse psychology," "paradoxical intention," and "prescribing the symptom."

305
 Foundations of Family Therapy
change, he doing what the therapist has asked him to do. In both
is

cases, the therapist remains in charge. Usually, the client's "resist-

ance" to the therapist gets the better of him and he chooses to
change.
In a subtle variation on this view, Haley bases his thinking about
therapeutic binds on analogies with game theory and an implicit
assumption that people are playing power games with one another
which they must at all costs "win." A person with a symptom
derives immense power to control his environment —especially his
relationships —by this behavior. Thus, if told to continue with the
symptom by a therapist, he is in a bind, since the only way he can
now control the therapist is by abandoning the symptom.
This argument for the success of paradoxical interventions has
been a persuasive and pervasive one. Even the Milan Associates, in
their earlier writing, seem to have adopted it. There is a central
objection to it, though, which has to do with a continual tendency

to use linear explanations positing special attributes in the individ-

ual. The rationale in question presupposes that certain kinds of
clients have an "oppositional" trait or need to control others, due
to some unknown personality component or to a family context
which brings out this trait. Such a view comes close to assuming a
motivational drive that, if cleverly harnessed by the therapist, will
push the client in the direction the therapist wants him to go.
There is another universe of explanations based on an ecological
or system-wide view. One such explanation has been developed by
psychologists Duncan Stanton and Thomas Todd and associates in
Philadelphia while working with addict families in the late 1970s.
Their technique of "ascribing noble intention" to the symptomatic
member, although it arose independently of the work of the Milan
group, is similar to the positive connotation and equally systemic in
that it symptom in the service of the family. Stanton gives
places the
credit to Nagy for the idea that symptoms are adaptive for the
family across generations, but he differs from Nagy in using this
formulation strategically. By way of explanation, Stanton proposes
a novel theory of "compression," based on the notion that dysfunc-
tional families continually oscillate from intense fusion with the
nuclear family to intense fusion with families of origin. One is

reminded of Bateson's bell buzzer analogy in that a preference for

306
THEORIES ABOUT THERAPEUTIC BINDS
one state promotes a switch to the other, in a no-exit cycle. The
therapeutic paradox, according to Stanton, intensifies this oscilla-
tion, or pushes it hard toward one pole, thus disrupting the cycle.
The resultant crisis, and the therapist's response to it, forces the
family to find new pathways. 2
If one were to take as the major unit of investigation the configu-
ration of relationships of which a problem is the most visible and
central sign, one can hypothesize a somewhat different rationale for
the success of so-called paradoxical moves. A good illustration is an
interview that was part of a brief therapy project at the Ackerman
Institute for Family Therapy.* The patient was a supposedly de-
pressed three-year-old with a harassed young mother and a preoc-
cupied, career-oriented father. The
improved quickly but then
child
it seemed that the wife was depressed. She found the pressures of

two young children and a purely domestic existence very difficult.

However, the more the therapist suggested activities that would
take this bright, well-educated woman out of the house, the more
difficult she became. Accordingly, the therapy team, stationed be-
hind a screen, called the therapist out and suggested that the thera-
pist make an about-face, telling the wife that she has convinced him
he has foolishly misjudged her nature. How could he have been so
insensitive as not to see that she is probably one of those women
who find their greatest happiness in serving others? Particularly
now, when her husband needs to put all his energy into his career
in order to help his family, it is him from
essential that she protect
the annoying distractions of domestic life. When he comes home in
the evenings, burdened down by work from the office, she must
under no circumstances let him do any chores or allow the children
to bother him, but see that he remains in his study. Even if he
should wish to come out and help her (and since he is a concerned
and loving husband, he probably will), she should resist, even to the
As
point of putting a lock on his study door. the therapist transmit-
ted this opinion to the couple, theyoung wife began to look more
and more displeased, although she sat quietly and said nothing. The
husband, by contrast, seemed very nervous and attempted to con-

*Led by Olga Silverstein and Peggy Papp, this team also included Joel Bergman, John
Clarkin, Richard Evans, Betty Lundquist, Gillian Walker, Anita Morawetz, and myself. The
intervention described was primarily the creation of Olga Silverstein.

307
 Foundations of Family Therapy
tradict the therapist, who politely refused to listen to his arguments
and ended the interview.
In the next session, the wife came in looking not at all depressed,
and gave the psychiatrist a piece of her mind for having misread her
character. She mentioned the projects that had taken her out of the
house that week, and her intention to begin taking courses at a
community college. The husband said that he had washed the
dishes every night and had even cooked the supper while his wife
went out to a concert. In fact he had discovered that he could get
enough work done at the office so as not to have to bring it home
any more. The therapist professed to find it hard to believe that he
had been so far off in his judgment and stated that he doubted that
things would continue in this way. However, he wished the couple
well and terminated therapy. A one-year follow-up found the wife
taking a graduate degree and the couple extremely happy with each
other.
To explainwhat was going on in this transaction, one could
simply say that these were two oppositional people who were strug-
gling for control of the relationship, and that the wife's tactics were
to play Cinderella in the ashes, making her husband feel guilty,
while the husband retaliated by distancing himself in work. How-
ever, one could equally well ignore individual motivations and con-
sider the relationship system alone. According to this view, the
change in the couple derives its momentum from the forces already
residing in the triangle.
First, this is a somewhat complementary couple, with the wife in
theone-down position. When the therapist pushes the wife too far
down, he overpasses one of the system's limits for relative power.
The wife may in other contexts be quite assertive, but her systemic
position vis-a-vis her husband requires compliant behavior. So she
will "resist" any effort to make her more assertive. The husband at
the same time must maintain a slightly higher position on the see-
saw, but not too high. This explains why the husband becomes so
agitated when the therapist defines him as needed to be served like
a pasha. One can almost read his mind: "There will be sheer mutiny
in the house if this happens; my wife will make me pay a thousand
times." Monitoring behaviors to keep the seesaw from too great a
slant immediately take over, with the wife refusing to play a menial

308
THEORIES ABOUT THERAPEUTIC BINDS
role and the husband desperately anxious to see that she does not.
But in their overreaction, the couple change their status to that of
a much more symmetrical couple, a change that presumably will
relieve both the wife's depression and the husband's less obvious
malaise.
Another factor that must be considered is the outrageous position
the therapist takes. Not only are both spouses eager to prevent any
attempt by the other to assume the suggested new role, but both are
irritated at the therapist who is them such absurd advice.
giving
Previously they have been covertly conflictual and unhappy with
each other; now they come in united and speaking loudly in one
voice. Never mind that the voice is saying, in essence, what an
unmentionable so-and-so this therapist is. The therapist's rejected
advice has brought them closer and has also evened out the balance
between them.
The point is that the forces that accomplished these changes were
the potentials for recoil built into the relative balance of the rela-
tionship. When the therapist tried to even out the seesaw, the
couple, answering to interior laws, could not do so. But when he
pushed the seesaw too far, it rebalanced itself, so to speak, empow-
ered by a recoil from within five relationship arcs: husband/wife;
wife/husband; wife/therapist; husband/therapist; and (finally)
couple/therapist. Since a team was also involved in this situation,
one must also add the relationship: therapy system/family system.
This activation of what could be called balancing responses in
each person and subgroup in a complicated relationship network
seems to be a more inclusive way to explain the success of this
paradoxical directive than an imputed "need not to be controlled"
within each spouse. The program the couple was caught and the
in,

rules of the overarching mechanism of this program, seemed to

govern the type of relationship this couple had worked out for itself.

Presumably the relationship, whatever its penalties, was adaptive

for this couple's particular situation.
Once the therapist has figured out the rules of the relationship,
or the behavioral redundancies which express them, he does not
need to rely on advice, persuasion, or attempts to repattern the
relationship or reeducate the couple; nor does he need to be espe-
cially charismatic or have the power of a prestigious setting or

309
 Foundations of Family Therapy

Figure 16.1
Family Balance Scale

status. Those who do not have huge reputations, important posi-

tions, or personal magnetism may yet by their wit and ingenuity
prevail if they let the energy of the relationship system, and the
energy of their relationship to that system, do the work for them.

The Metaparadox: Family- Team

Moving to a higher context level, there is still another way to

describe how paradoxical directives work. Here we may ignore indi-
vidual factors, or even recoil factors from within a given set of
relationships,and look at the matter from the point of view of a
family system worked upon by a therapy system. In thinking about
symptoms in families, we have talked about a recursive cycle
whereby the system is opposed, then encouraged, opposed, then
encouraged, endlessly. Thus it remains, a wanted yet unwanted
guest, neither sitting nor standing, staying nor going, but hovering
inan endless and tantalizing stasis.
Imagine this recursive cycle in terms of a balance or scale in which
one side has an "NC" (for "No Change") upon it, and the other side

310
THEORIES ABOUT THERAPEUTIC BINDS

Figure 16.2
Family Plus Therapist or Team: Version One

has a "C" (for "Change") upon and imagine further that the C
it,

and NC are assigned roughly the same weights (see Figure 16.1).
One could invent a rule or constraint for the family that keeps these
two weights in equal balance at all times. If the NC becomes less

important or slacks off, C

the have to compensate by becoming
will
weaker too, or else a new element of NC will have to be added. One
finds, interestingly enough, that if a person who usually supports
a problem does not do so, or is absent, another family member will
fulfill that task. Alternatively, a covert expression of symptom sup-
port will become overt, as when a mother of an anorectic who is

eating well begins to worry that her daughter is eating too fast and
follows her about, keeping watch over her food intake. Further-
more, let us keep in mind that this balancing act is vital to the
family, for some reason that is not spelled out. A rigid family system
faces the most danger when forced to reorganize, and the symptom,
as we have said, may be one answer to this dilemma.
Now imagine what will happen if such a family, with its interior
scale or balance, is joined to a larger scale or balance, represented
by a therapy team or a powerful single therapist. The whole can be
represented by a new double balance, with the family scale on one
side, and the therapist scale on the other. (See Figure 16.2). The fact

311
 Foundations of Family Therapy
that the family has entered therapy, or just their presence in the
room, must also be assigned a weight in favor of change; hence the
"C" at the top of the therapy side of the scale.
If, as in the model portrayed above, the weight of the therapist
is placed on the C side, in addition to the family's presence in
therapy, one can see that the family "rule" that forces for change
must be balanced equally by forces against change has been vi-
olated. Therefore one could predict that both sides of the family
have to shift to NC, interpreted usually as "resistance."
scale will
What must the therapist do, if that is the case? Logically, his move
would be to reverse his C position with a large NC, especially as any
NC of his is still countered by the C of the family being connected
to treatment. Then one might predict that the family will, if it

follows its own rule for balance, change both its NCs to Cs, produc-
ing the following picture (See Figure 16.3).

Figure 16.3
Family Plus Therapist or Team: Version Two

}12
THEORIES ABOUT THERAPEUTIC BINDS

The Counterparadox and Levels of Logical Type

A serious problem with the balance scale analogy for change

processes in therapy is that it does not offer a way to include the
contradictions between overt and covert messages. The Bateson
group originally based their ideas about double binds and therapeu-
tic double binds on the confusion of levels of classification they

found in families of schizophrenics and used when making inter-

ventions with such families. Here, then, is a double-level way to
schematize our model, illustrated by another of the Milan Associ-
ates' cases.
A paper read by Selvini during a workshop in New York in 1977,
7
entitled "First Session of a Systemic Family Therapy/ describes the
team's work with the family of a seventeen-year-old anorectic girl,

Antonella. Her family consisted of mother, father, and a fourteen-

year-old brother, Fabrizio. During the initial interview the thera-
pists were struck by the impassive way the mother described her
daughter's condition, which was extremely serious. By contrast,
when she spoke of how much pain Antonella was causing her
maternal grandmother, Grandmother Theresa, the mother wept.
Father acted uninvolved, and the brother seemed interested only in
his outside pursuits. The family had no hypothesis about the onset
of the illness; all they could say was that the girl had been working
in a factory when suddenly she stopped eating. The therapists asked
Antonella herself what was going on just before she became ill, and
she answered that she had a boy friend but had been afraid to tell
her family because she felt they would not approve of him. Sure
enough, when the parents found out about him they put a stop to
the alliance. The fiance subsequently got drunk in a local bar and
spent a few days in jail for disorderly conduct. When asked, An-
would never stop loving this young man, and
tonella said that she
if she were to become well she would start to see him again. The

mother looked upset and said that Grandmother Theresa would be

horrified tohave her granddaughter going out with a "jail bird."
Some questioning about the early married life of Antonella's par-
ents brought out useful information. The parental couple had first

313
 Foundations of Family Therapy
lived with Grandmother Theresa, and Antonella had always been
a marked favorite of the older woman. This living arrangement
came to an end when the father's mother, who lived not far away,
said that she could no longer visit her son while he lived in the
house of "strange people/' So the husband insisted that they get
their own apartment, much to the sorrow of his wife, who was most
unhappy about leaving her own mother.
Noting that the feelings and well-being of Grandmother Theresa
were treated as if they were as serious as Antonella's illness, the
team decided to attack the triangle of the three women first. Accord-
ingly, this prescription was read and given to the family:

The specialists' team has decided that family therapy is very suitable,
considering that Antonella's life is in danger and there are, in her case, high

probabilities of recovery.
But we are faced with an extremely serious risk and worried about it.

The risk is connected with Grandma Theresa. If Antonella should recover,

she would probably start meeting Franco again. Grandmother would feel
ashamed and suffer mortal pains because of this; in fact, she would suffer
even more than she would if Antonella should die of her illness. During
the next session . . . we shall discuss this danger threatening Grandmother.

This is a classic version of a therapeutic double bind. Look how

neatly it reverses the contradictory messages of the family. Within
a covert framing message of "Do not eat, because it will be too
threatening if you get well" the family says, overtly, "Eat and get
well." The team, by stating that they feel family therapy is in-
dicated, imply covertly that the girl should get well; however, their
overt message is that she should not get well, because Grandmother
Theresa will suffer too much. It is a reverse of the contradictory
messages the family gives the girl, and places the family in a meta-
bind, a state of confusion and upset that hopefully will produce
some change. (See Figure 16.4).
At the next two weeks later, Antonella looks much bet-
session,
ter; she has started to eat and has gained weight. But the mother

complains that the girl eats at night and the mother has to get up
to monitor her eating. It is a less covert version of the message
"Don't eat." Interestingly enough, the girl had run to see her grand-
mother as soon as the family got home after the first session, to

3H
THEORIES ABOUT THERAPEUTIC BINDS
1. Family Paradox

Covert message: No Change

Overt message: Change

2. Team Counterparadox

Covert message: Change

Overt message: No Change

3. Family Countermove

No Change

Result: Antonella eats.

Figure 16.4
Overt and Covert Elements of a Paradoxical Prescription

show her grandmother the letter. The old woman had said, "Non-
sense, of course I want you to eat. Besides, there are many Francos
in this world." As one would expect, the mother looked very de-
pressed during the session, while the father and Fabrizio seemed the
same as usual.

315
 Foundations of Family Therapy
According to Selvini, who was asked about the progress of the
case,Grandmother Theresa was invited to the next session. The
team wished to block off the grandmother, but they did nothing
more than treat her respectfully and thank her for coming; they did
not communicate with her again. This time the message went by
mail, to Antonella's household only, excluding the grandmother by
design. The letter contained three different messages, which each
family member in turn had to read to Antonella before supper each
night. Father's message read something like: Thank you, Antonella,
for refusing to eat because this will keep Mother at home and close
to me. The mother was to say: Thank you, Antonella, for refusing
to eat, because when Grandma Theresa scolds me for neglecting her,
I can tell her I am taking care of you. And the brother was to
say: Thank you, Antonella, for refusing to eat because in that way
all the attention goes to you, and I can go out and play with my
friends.
Only after the girl began to gain weight, to become more inde-
pendent and less part of the family, did the underground quarrels
between her parents begin to surface. The original fight around
whose mother should have primacy had never been resolved, nor
had other issues, and the struggle between husband and wife came
to the fore.
It is, of course, an immense simplification to try to diagram such
a complex process on the moves people in
as placing a reverse fix
a family make that simultaneously invite and discourage change.
And yet, like a beckoning finger, the puzzle draws us on to assay
more and more elaborate explanations. One clear idea emerges,
however: Whatever configuration the family puts forward to the
therapist must be seen as a punctuation peculiar to that context; the
therapist then imposes a counterpunctuation that switches the first

one about. Neither punctuation is more or less true, better or worse,

than the other; both are equally drawn from the universe of linear
causality. What seems to happen is that the confluence of the two
opposing punctuations enforces a new punctuation, not heretofore
suspected or present.
It is possible that the new punctuation comes from a vocabulary
which was not available to people in the system as it used to be. The
fascination of the phenomenon of binocular vision or the bicameral

316
THEORIES ABOUT THERAPEUTIC BINDS
brain lies, as Bateson suggests, in the possibility of creating a tran-
—
scendency different perspective as with two eyes one moves from
an experience of two-dimensional to three-dimensional space. Per-
haps the juxtaposition of contraries is for living systems what meta-
phor is to poetry: a force that pulls, not pushes, for a leap.

Holding One Part Still

A growing discomfort made me start to rethink these models I

had been playing with. They still separated therapist and family
into two units, and they did not sufficiently attend to the character-
istics of the entire field. Looking about me, as one might gaze idly

at the ground, I began to see that what I had thought were pebbles
—
were in some cases precious stones they had been lying about my
feet all this time. I am referring to a group of ideas that keeps
reappearing in different forms, ideas that have been spelled out by
different people, but which should be given the status of a concept.
This concept would not only explain why therapeutic binds work,
but include other kinds of therapeutic moves as well.
Let me start with Bateson's haunting phrase: "the infinite dance
of shifting coalitions." There is a way in which a system stays
structurally the same by virtue of a constant flutter of smaller move-
ments that counteract any serious efforts to reorganize it. This ob-
servation, as we have said, accounts for the chaotic surface look of
families with very disturbed members and the contradictory finding
that on an underlying level these families are extremely rigid. This
phenomenon is not unlike the experience of the canoeist who, see-
ing a particular configuration of wavelets on the surface of a stream,
knows that there is a rock or log stuck firmly beneath.
In Mind and Nature, Bateson clarifies this general idea: Talking
about stability, he argues that one must always keep in mind the
entity whose stability is in question, for there are many types of
mechanisms for maintaining equilibrium. When one comes to living
systems, one finds that

the whole mass of interlocking processes called life may be involved in

keeping our object in a state of change that can maintain some necessary
constants, such as body temperature, blood circulation, blood sugar or
even life itself. 3

317
 Foundations of Family Therapy
As a result, one must "follow the example of the entities about
which we are talking/' For the mammal, stability depends in part
on the temperature of its body, with its variables that can change
reactively with the environment. For the acrobat on the highwire,
stability means balance. To describe that, one must take into ac-
count the smaller —the
instabilities of the pole or changes of
shifts
posture —which keep the acrobat balanced. In these instabil- cases,
ity (on one level) means stability (on another level), and Bateson
argues that these different levels must always be held in mind.
In a family with a problem, one important stability is the way
coalitions in the family are organized. In his early book, Strategies of
Psychotherapy, Haley puzzles about how to describe a three-person
system. He shows by an example how a relationship system is kept
the same by compensatory shifts in the coalitions, as evidenced by
family members' behaviors. For instance: son misbehaves, father
reprimands son, mother chastises father, father defends his action,
and mother looks exasperated with him. Change may be difficult to
introduce in that system because minor shifts (small instabilities)
will occur that counteract it. Haley notes a possible sequence:

If one merely persuades the child not to misbehave, the system does not
change, for father then likely to say to the child, "Why are you so
is

quiet?" and mother will respond with, "He can be quiet if he wants, leave
him alone," and father will say, "I was only wondering." That is, a change
in an individual can lead to an adjustment ... so that the system remains
unchanged. 4

There is a more abstract way to describe a system that is so

constructed as to have trouble changing, and here we go back to
Ashby's concept of the too-richly cross-joined system (see Chapter
5). Such a system, as we know, tends to erase inputs, since no
change will persist unless all parts conjointly change. This is a
highly unlikely possibility, given the fact that each time one part
changes it sets off a change in another part, which sets off a change
in another part —
the result being that the first change can never
become a state of equilibrium for the whole. The only help for this
situation, Ashby says, is to introduce "interactive sequence con-
stancies." These will temporarily interrupt the deviation-coun-
teracting motions induced by a new input. Actual severing is nc

318
THEORIES ABOUT THERAPEUTIC BINDS
essential; it is enough to put in a "null function" by making one part
quiescent.
In my 1975 article on this idea, 5 I was interested in the fact that
clinical work with families with symptomatic members seemed to
consist of putting in "interactive sequence constancies" all the time.
Selvini then used my commentary on Ashby and Ashby's own ideas
in an article, "Why a Long Interval Between Sessions." 6 Referring
to Ashby's observations, she says that the therapeutic task is to
avoid being caught in the organizational pattern of a too richly
cross-joined family. From the beginning, the Milan Associates had
regarded the family-therapist suprasystem as the unit to consider
whenever making interventions, rather than thinking of the thera-
pist or team as one independent unit acting upon another. Thus
Selvini warns that the main danger would be for the therapists to
be drawn by the family into a new system as richly cross-joined as
its own. This would make it impossible for the therapists to keep

outside the system enough to retain control, let alone introduce

change.
The Milan team's answer to this dilemma, suggested to them by
Ashby, is first on interventions that would disrupt
to concentrate
the connections between parts. They favor moves that will make
reactive parts nonreactive or that mark off certain behaviors in such
a way as to keep one part of the system still. This approach consti-
tutes a pragmatic application of Ashby's remark, which struck me
at the time I read it as both poetic and true: "Constancies can cut
a system to pieces." 7
Most of the Milan Associates' interventions are, like many other
therapeutic interventions, designed to block, disrupt, or derail cus-
tomary sequences. A ritual that places two persons together one day
a week disrupts heaven knows how many redundancies, and that
is one type of constancy. Reading a message every night is another
constancy. Thus, Selvini notes, the therapeutic intervention itself

disconnects the family members from their usual positions in the

family system.
For tasks and rituals to succeed, however, the family must carry
them out. Some families will comply, others will not. With families
that will not or cannot comply, the best way to introduce a con-
stancy is to seek out the most rigid and immoveable sequence or

3*9
 Foundations of Family Therapy
behavior that a family presents in therapy, and simply prescribe
that. This may be the refusal of the family as a whole to attend
therapy, or it may be a symmetrical struggle between the two
groups of in-laws, or it may be the problem itself, linked to the
position of every person in the family as a hub is linked by spokes
to the rim of the wheel. Whatever that "most immoveable" item
might be, that is the configuration to prescribe. Thus the therapists
join the family in reinforcing a constancy that is already there.
To be sure, the so-called constancy probably is in Bateson's "state
of change" and is not so constant. Whatever is activated by the
therapeutic encounter is likely to be the family balancing pole, at
moment. To rigidify it further by prescribing it is to
least for that
amplify and put the family off balance, just as the acrobat is
it

unbalanced when his pole is held still by an invisible force. If the

pole is unusable, then other possible ways to maintain balance
(body posture, slower movements) may be adopted. This may be
why other "problems" so often follow the maneuver of freezing
(amplifying) the presenting complaint or the configuration in which
it is embedded. In any case, this is one possible explanation for the
force of a systemic injunction targeted at the family's most palpable
rigidity —the one facing the therapist or evoked by the therapeutic
situation.
An additional therapeutic bind is placed on the family by the
other method Selvini describes: the "long interval between ses-
sions." At first accidentally, then by design, she and her associates
scheduled sessions at least a month apart. This is not just because
a richly joined system requires more time than less connected sys-
tems to process an input or intervention, but because it offers one
more way what Ashby calls a null-function or "wall of
to place
constancies" between the family and the therapy team. They are
still connected, but the family cannot reach the team. The usual

feedbacks are not allowed to dissipate the force of the intervention

by being bounced back on the team or otherwise disqualified. The
block to reactivity is placed within the family, and then the thera-
pistsystem withdraws and a block to reactivity is placed between
family and team, allowing the first block to have greater effect. It
is like a therapeutic bind plus a metabind.
The idea that a constancy can disrupt a system was always lying

320
THEORIES ABOUT THERAPEUTIC BINDS
about in the back of my mind, and in my 1975 article I cited
maneuvers from both structural and Bowenian schools that illus-
trated this principle. Now I would add maneuvers from the strategic

and systemic schools, especially the techniques of prescribing the

symptom or "system" (as a systemic prescription has been loosely
called). In very richly joined systems, these moves may be stronger

than direct attempts to introduce change. The Milan stance of pre-

scribing elements from the entire field, including the referring con-
text, is just a widening of the methodology. Perhaps the concept of
"neutrality" is another way of describing the judo posture of that
methodology.
One more pebble that fit my collection was a contribution from
Paul Dell's paper "Beyond Homeostasis." Dell describes living sys-
tems as always evolving, always accommodating new inputs. He
leans on Bateson in calling such systems "self-healing tautologies."
What caught my eye was a sentence that included a reference to
constancies: "An output or feedback that introduces a constancy
into the system is a strong force that disrupts the preexisting organi-
zation of the system. When thusly disrupted or distorted, the sys-
temic tautology will heal itself into a somewhat new tautology and

a somewhat new steady state." 8

Dell moves into line with Selvini's position on counteracting
family suction by saying that the input or feedback introduced by
a therapist will be "swallowed" by the family system. He observes
that this process is a way the family cancels out attempts to change
it. He defines therapeutic input on the basis of whether or not it
really affects the problem or complaint. If it does not, then it can no
longer be defined as input. It is just a piece of family life. Dell
further notes, and many therapists will agree with him, that when
this happens, the therapist himself has become a piece of family life.

And if the family stays in therapy for any length of time, this is

always true. "The only relevant question," Dell wryly remarks, "is
whether the therapist is now part of a still dysfunctional family
system or whether it has changed, and the therapist is now part of
a non-symptomatic family." 9
The genius of the Milan Associates' approach is that it addresses
point in the following ways: (1) by their
itself so precisely to Dell's

concept of "neutrality" (always remaining nonreactive); (2) by the

321
 Foundations of Family Therapy
tactic of introducing blocks of time between sessions (reducing the
chance that they or their input will be swallowed up by the family
before it can be effective, and lengthening the amount of time
before they become totally engulfed and useless); (3) by the use of
tasks or rituals that introduce elements that remain invariant; (4) by
their attention to those elements in the widerwhich might
field
nullify their input; (5) by which not only
their circular questioning,
introduces new information at the system level but makes them less
vulnerable to the family's attempts to "swallow" them; and (6) by
their systemic prescriptions which hold one part still. Using these
techniques, they have achieved a remarkable methodology for "cut-
ting the system to pieces" and forcing it to find alternatives. Like
all therapists, they are always attempting to introduce what they

call new punctuations, but more importantly they attend to the

family's maneuvers to deamplify these punctuations or disqualify

them, and have invented a brilliant set of devices for preventing the
family from doing this.
This, then, would be another way to see the reason for the useful-
ness of so-called paradoxical prescriptions. I say "so-called" be-
cause the term is beginning to make many serious thinkers in the
field uneasy. The Milan Associates are beginning to turn their backs
on it. The term paradox in its conventional sense suggests that the
maneuvers designated as paradoxical are illogical. They are only
illogical, it seems to me, from the point of view of traditional norms

of treatment. Seen from the perspective of intervening in too richly

cross-joined systems, they seem to be extremely logical.
Obviously, this does not close the subject, which should continue
to generate heated discussion for some time to come. I will now
move on to a consideration of a few other issues that working in a
systemic way brings to the fore.

322
Chapter 17

Issues on
the Cutting Edge

The Turning Facets of Resistance

One of the first signals to early researchers that a psychiatric

symptom might be a family phenomenon was the often noted fact
that if the problem of the identified patient improved, other prob-
lems might arise. Clinicians called this tendency "symptom substi-
tution" (a term used in individual psychotherapy), as if there were
something about certain families that required a symptom, much
the way a dragon needs a virgin for breakfast. This puzzing idea
why would a family "need" a symptom? —baffled therapists in the
days before it was realized what a beautiful, serviceable, well-con-
structed artifact a symptom is, thoughtfully provided by nature to
help families that were terrified (probably for good reasons) by the
threat of change.
We have become more respectful of families since we have gained
mechanisms that evolve to
a better understanding of the intricate
monitor such threats. Living systems usually have a number of
ways to damp down fluctuations that might lead to change. The
therapist may find himself feeling like those princes who keep going
off and completing tasks against great odds to win the hand of the

323
 Foundations of Family Therapy
princess, only to be told that there is one more task, and yet one
more. The therapist feels like the unhappy prince: Here I have
solved one problem, only to be confronted by ten others. (Sotto
voce: Is this princess really worth it?)

But for many families this is the normal course of therapy. They
have a repertory of arrangements, usually seen as problems (at least
by the therapist), that fall out of the defeat of the problem that
brought the family into treatment. To foresee at least the probabili-
ties of what these other
stages might be is part of the art of therapy,
even when seems impossible to make specific predictions. A new
it

problem may be a less life-threatening or serious problem than the

very big or) it may not. Will the "good" sibling become
first, or (a

a "bad" sibling if the "bad" sibling reforms? Or will the marriage,

which seemed to depend for its unity on the worry caused by sick
child, go on the rocks? Or will the mother become seriously de-
pressed if moves toward
the psychotic son regains his sanity and
independence? The negative consequences of not having a symp-
tom can be (although they don't have to be) at least as serious as
the seriousness of the symptom, and any therapist who works with
an entire family begins literally to ask: Is it better or worse for this
family not to have this problem? If I tamper with it, will I be able
to handle the consequences?
The Milan model, with team or a screen, is useful
or without a
because it so clearly seeks to define and coopt whatever counter-
move the family presents in any given session. The experience of
working with this model almost mandates the expectation that any
intervention powerful enough to interrupt a symptomatic redun-
dancy will elicit a countermove. The therapist must, accordingly,
always be watchful to meet the challenge of the new weapon and,
by coopting it, prescribing it, or reframing it, set the stage for the
next round. The Milan team's case histories show how adroitly they
swerve to hold fast to the protean shape that presents itself in such
different forms from session to session.
This does not mean that every family offers a variety of blocking
maneuvers. Very often the simple shift from child symptom to
marital problem is the version encountered in child-focused set-
tings. And sometimes the family makes a jump to a new integration
without any of these intermediate steps, or goes through its own

324
ISSUES ON THE CUTTING EDGE
series of stages on its own. But if these shifts, tossing up new
obstacles, appear, this does not mean that the therapy is not going
well, or that the therapist is not a good therapist, or that the family
is too "sick" to be worked with. It may actually be a sign that the
therapy is having an effect.

This brings us to the next facet of this way of working. If therapy

is succeeding, one is often faced with an extremely frightening
escalation —a sudden worsening of the problem—or the eruption of
a new and more destructive one. This can be difficult for the thera-

pist, but it may be that without a crisis of this sort, some families
will never achieve a major structural change. There is no threat so
likely to make a therapist draw back from the decision to try to
crack the symptom as the sudden explosion of forces which may be
set loose by a powerful systemic intervention. Therapists working
way
this for the first time often feel that all is lost when such
mayhem breaks loose. The more chronic and severe the primary
system is, the more disastrous the crisis is likely to seem. And yet
without this crisis, the family may never escape its deadly dance.

The Evolution of a Problem

The systemic approach, as we have seen, is deeply concerned with

the evolution of a problem, which frequently means going back to
previous generations and to collateral groups in the present genera-
tion as well. But the reason for focusing on the past is not to cast
as wide a net as possible in the belief that understanding the whole
process will help people to escape from it or change it. Rather, the
idea is to be economical, and to trace out only as much as one needs
for a hypothesis. If interventions based on this hypothesis do not
work, or work incompletely, then explorations into a whole new
branch of the family or its history are in order. This information is
used to disrupt the symptomatic configurations in the present.
There no idea that "working through" an event or relationship or
is

pattern in the past will relieve a problem or symptom, as the more

historically oriented family therapist believes.

3*5
 Foundations of Family Therapy
I treated one family with a boy who had always had trouble at
school. He was a likable, spirited seventeen-year-old, but since his
mother had engaged in endless battles with him over
early years his
He was not doing well during his senior year at a
his school work.
demanding prep school, and had been caught cheating in writing a
paper. The boy's anxieties were centered mainly on academic
achievement, and psychological tests revealed that he was very
nervous about doing anything without exact instructions. The par-
ents were in the process of finding a less demanding and more
relaxed school, but the school psychologist had recommended fam-
ily therapy because the boy's problems were "emotional."
A look at the family showed many strengths. Boy and parents
were quite capable of coping with the problem of finding another
school. It was hard to understand why, in this family where non-
conforming brightness and originality were prized as well as con-
ventional abilities, the son was having such a problem. He seemed
like a person who did not have an academic bent but would un-
doubtedly find his own level of achievement if not pressured. The
one who seemed to feel that pressure was necessary was, of course,
the mother. There was some hint that the boy, who seemed to be
repeating his artistic father's own early school difficulties, reflected
the mother's disappointment with her husband, especially his
choice of an artistic career, with all the financial problems that
implied. But the couple's fondness for each other seemed much too
genuine to require such a serious problem to keep their relationship
going.
When we went back to the mother disclosed that her
past, the
own mother was a hard-driving perfectionist who had always made
this oldest daughter the living proof of her own success. As a little
girl, the mother had had to be more brilliant, more responsible, more
handsome, more successful, than any other mother's little girl. She
ended up with many resentful feelings about her mother, but did
not outwardly rebel except in one respect: She chose for a husband
a man who was an artist, not the banker or businessman her mother
would have preferred. She continued to hold a high standard for
herself and for everybody else in the family, especially her son.
It seemed that the only possible move would be to try to break

the hold this perfectionist grandmother still had over her striving

326
ISSUES ON THE CUTTING EDGE
daughter, which had made her try so hard to produce the perfect
son. Of course the harder she tried, the more the boy resisted (possi-
bly also out of an identification with his father), even though con-
sciously he was doing what he could to live up to her expectations
and feeling miserable when he failed.

The therapist told the mother that, in a curious way, she was still
showing her mother how hard she was trying, and would always
try, to be the best in every way: the best career woman, the best

wife, and the best mother. This was the greatest sign of devotion
she could possibly pay her mother: that she was still trying to please
her at any cost. And this was why she had to continue to pressure
her son to succeed: out of loyalty for this mother whose opinion she
had long since thought she did not care about.
At the same time the boy was told that by presenting himself as
an academic failure, which had been his father's problem, he was
not only being loyal to his father but was in some way representing
his mother's resistance to her own mother's perfectionism. In this
family where there was a rule that failure must never be allowed,
he was a kind of pioneer.
The aim of this intervention was from the
to activate a recoil
strictures the mother did not realize she was laboring under; and to
redefine the son's apparent disgrace as a way of helping the family,
especially his father. The statement to the son would not have been
sufficient had not been coupled with the mother-grandmother
if it

issue, making it hard for the mother to continue her effort to "per-

fect" her family without feeling that she was playing into the hands
of her own deeply resented mother.
A number of changes followed this session. The father an-
nounced that he was quitting a poorly paid but stable job he had
held for some years and going back into free-lance work. His wife,
although it obviously cost her something, backed him and said she
was proud that he had the courage to leave a job that was unworthy
of him. The boy was placed in a school that emphasized athletics,
which he was good at, rather than scholarship.
This, then, is an example of a three-generation systemic interven-
tion. We can see how forces coiled within the past can be used as
a source of momentum for change in the present. History, used in
this way, is no longer dead but alive, carrying within it the means

327
 Foundations of Family Therapy
for the destruction of the very patterns it predicted and sustained.
This is the systemic use of the past, very different from the older
view that to understand or "work through" emotions or attitudes
or patterns that are no longer appropriate to the present will elimi-
nate them. They would not be still in use if they were not, in some
way, still appropriate.

Establishing a Benevolent Contest

One aspect that differentiates the strategic/systemic approaches

to therapy from many others is the seemingly less compassionate
role of the therapist. Milton Erickson has spoken of the "contest"
between therapist and client, and most of his therapy consists of a
brilliant balancing act between bringing out positives in a client's
situation and bringing out — —
almost eliciting his resistance.
It is possible that a contest of some sort is a natural prerequisite
for any therapeutic change. This idea goes against the helper/healer
notion, whereby the therapist is a benevolent doctor of the soul,
offering advice and support or acting as his client's advocate or
guide. But implicit in most therapies is an adversary position, usu-
ally covert, sometimes overt. From the momentum generated by this
position comes the energy for change. An ineffective therapist is like
a fisherman whose line hangs limp in the water; at times there is a
—
tug the fish nibbles, then gets away. The therapist who is caught
in the family's or client's "game" is painfully aware of this fact; the
interview drags on tediously and the therapist feels confused, in-
competent, and humiliated. By contrast, a systemic prescription that
hits every relationship unit in the group produces an indescribable
effect. It is as if the fish, thinking itself quite safe with this booby

of a fisherman, suddenly finds itself with a hook in its mouth. There

is a general motion of shock, surprise, rejection, as the family tugs

hard on the line. A quiet period often follows, as family members

try to figure out their position, much as the fish swims around with
the slack line, not realizing that it has been caught. Then comes the
relentless tugging as family members begin to react; one will chal-

328
ISSUES ON THE CUTTING EDGE
lenge the therapist; another will laugh and insist that the therapist
is joking; another may take the message very seriously and, in some
cases, even weep. Bafflement, bewilderment, anger, indifference are
common. But the major fact is that the therapist, perhaps for the
first time with this family, feels that he has a fighting chance to

catch this elusive fish, which is not any particular family member
but the relentless pattern that enslaves them all.

Of course a therapeutic contest is not indicated for families whose

resistance to change is low, or whose range of behaviors can readily
be reset. As we have previously said, neither is it appropriate for a
family that is so badly fragmented that there is no potential for
recoil. The contest stance is for the family at the "consensus-sensi-
tive" or "too richly cross-joined" end of the continuum. Here is

where one finds the entrenched psychotic or neurotic behaviors that

mark the family stuck in a "game without end," and where the
therapist knows it is vital to harness the forces that reside within
the family's resistance to change. And if in addition to the contest
with himself the therapist can add elements of intense ongoing
contests within the relationship system itself, there is even more
momentum to draw on.
A very real problem for the therapist who works systemically,
however, is that he or she may attract all kinds of hostilities and
negative reactions from members of the family. Even if he escapes
blame, he will not receive credit. An intervention, especially a mes-
sage encouraging a problem, may by itself create a seemingly magic
turn for the better, so that the symptom the family complained of
will disappear. However, they will not always associate this change
with the therapist or even remember it. A child with daily head-
aches may suddenly stop having them, but when the therapist asks
how the headaches are, the mother may have complete amnesia
about the complaint that brought the family into treatment, and
proceed to complain about some other problem.
Sometimes it even seems as though the therapeutic relationship
may have to be sacrificed in favor of accomplishing at least some
minimal shift, which takes place even though the family is irritated
and leaves treatment. To deliberately engineer things so that the
family drops out of therapy and leaves feeling angry with the thera-
pist is extremely uncomfortable, but it is sometimes the price of a

329
 Foundations of Family Therapy
change. If there must be a scapegoat, better that it be the profes-
sional.
More often the family, after a major shift has occurred, merely
becomes cool toward the therapist or team. I have wondered some-
times whether the systemic approach may not draw on the mysteri-
—
ous extrusion process the need to draw bodies in or push them out
— so often noticed in families in crisis. At least one person becomes
symptomatic, and there seems to be a decree that someone has to
go: through hospitalization, divorce, running away, or worse. A
therapist, by appearing to take the side of all the things the family
is complaining about, offers himself as a substitute for extrusion.
The family, in pushing the therapist away, also pushes away the
behaviors they have so extravagantly attached themselves to. This
would be analogous in pharmacology to giving a person with gastric
acidity an aluminum gel compound that coats the offending sub-
stances and is expelled along with them. How else to account for the
amazing indifference families display toward a therapist or team
after the problem has dropped away, no matter how intensely all
have been involved before?
In any case, a systemic approach is not for therapists who must
be liked at all costs. The therapist will not get cards at Christmas
or invitations to weddings. Therapists who work this way must
have a cotherapist, or a team, or some kind of sympathetic profes-
sional group. This is not just for support while in braving out some
hostility-arousing intervention, or to get suggestions for making a
stronger impact on a family that is eluding efforts at change, but,
first, to get the rewards we all need during difficult and unap-
preciated work and, second, to have a support group to counteract
therapist extrusion.

A Graceful Exit

If this approach to therapy offers difficulties for the therapist, it

also presents acute problems for the family. Termination of therapy,

in particular, is an issue for the family, not just the therapist. The

330
ISSUES ON THE CUTTING EDGE
situation created by the therapeutic contest poses a problem in
ending therapy because the family should not leave with the feeling
that the therapist has defeated them and is gloating. Thus Milton
Erickson speaks of supplying a graceful exit for the patient: Con-
ventional therapy provides solemn rituals to end therapy —working
through the loss, as if termination were like a death, or as if the
therapist were so important a person that to lose him or her would
be to lose a familymember or dear friend. But this does not apply
to the methods we have been describing here.
Credit must be given to the strategic-systemic schools of therapy
for having organized therapy in such a manner that the sought-for
change takes place out of the room, away from the therapist, and
usually with a disclaimer by the therapist that anything he or she
did, or anything in the therapeutic relationship, brought about the
change. If anything, the therapist challenges the change or declares
that it won't last: a sure sign to the patient that the therapist hasn't
much stake in it. The patient can then own it, rather than having
to give the therapist the credit.
A particularly nice example of a way to give a family a "graceful
exit" and some confirmation of change is to ask each mem-
still get
ber the question: What would you personally have to do to bring
things back to the way they were when you first walked into the
room? This question can be asked of a family whose members have
absolutely no understanding of their own contribution to maintain-
ing the problem, or any idea of the processes that made it disappear,
and yet they answer as if they had each studied in detail the
will
part they played in the whole event. A mother of a "behavior
problem" boy will say, "I would have to think I had to solve every-
thing by myself"; the father will say, "I would have to be away from
home all the time, working late"; and a younger "good" sibling may
say, "I would have to yell as if he were killing me every time he
came near me."
In this therapy, termination can often be presented in the form
of a recess. A therapeutic model which does not think of emotional
problems as illnesses does not think in terms of cure. If a psychiatric
problem is defined as evidence that a family is having difficulty
negotiating a transition to a new phase, the therapist's job is to help
with the transition, then get out, leaving the family to settle itself

33*
 Foundations of Family Therapy
around its new integration. There is no assumption that the family
will then be trouble-free, because new transitions or crises will
come along and new symptoms will arise. As with the old-fash-
ioned G.P., this new-fashioned therapist sees his or her function as
being on hand when needed, any time the family is having problems
as they face a change. An episodic therapy is the result: with ses-
sions placed close together or else far apart; with recesses common
when it looks as if the family can go it alone; with invitations to
come back at any time for a "checkup," or for more serious things,
but with a hope and expectation that this will not be necessary.

The God in the Machine

One objection to the team approach is that it seems like an enor-

mous outlay of therapist time on comparatively few cases. How can
this become a workable model when it takes four people to deal
with one problem, even though the number of sessions may be
relatively few? One can justify it for training purposes; there it is
even rather economical. One can justify it for research, of a soft
clinical nature but nevertheless breaking new ground. But how can
one justify it as a major therapeutic method?
The obvious answer is that a team is forcibly called for in severe,

difficult cases that do not yield to years of intensive individual

treatment, or to family treatment of a more conventional type. We
are talking about serious problems, the psychosomatic "killer" dis-
orders, the class of conditions known as the psychoses, and many
other stubborn, resistant symptoms which, though not always life-
threatening, cripple the activities of the person and the family in
which they are found. Surely, in every clinic, there should be a team
following the systemic model for problems that have proved resist-
ant to other approaches.
What is of immense team approach is that it calls
interest in this
upon so many sources of power. When using one of these teams for
consultation, the therapist who has been stumbling along with a
difficult family finds his or her own position immensely enhanced.

332
ISSUES ON THE CUTTING EDGE
Call it a deus ex machina effect, born of the expectation that
an effect

goes with the fact of unseen "experts," a cheap magic trick, or what
you will, the presence of the group exerts tremendous leverage. As
a matter of fact, at least one therapist I know, having no group and
no one-way screen, used a phone in his office and pretended to be in
touch with a group of consultants, giving his family a message from
this imaginary team which had the same effect as if the team were
real.

However, it is more than an atmosphere of mystery that produces

the sometimes amazing changes induced by this method; it is the
synergistic effect of several minds applied to a given problem. I
agree entirely with the position of the Milan Associates that the use
of the therapeutic paradox, and the logic it is based on, is indepen-
dent of the charismatic influence of a particular therapist. It would
be incorrect to say "independent of the therapeutic relationship,"
but with that as a proviso, the method is indeed, if correctly applied,
amazingly effective. This is true even when the family is working
with a comparatively inexperienced therapist, if there are ex-
perienced practitioners behind the screen.
Peggy Papp points out an extra resource the team provides for the
therapist in "The Family That Had All the Answers." She describes
the team as being able to differ from or criticize the therapist in the
room, thus acting as "a dissenting voice forming a triangle and
forcing the family to take sides." 1 S. J. "The Social Base
Miller, in
2
of Sales Behavior," describes a similar maneuver. The car salesman
impresses the client with how hard it has been to convince his boss
to get him to accept the supposedly low price he has worked out for
his client. As he is writing up the
manager comes in, and
sale the
the following prepackaged scene takes place. The manager "discov-
ers" that his salesman has forgotten to charge for a number of quite
expensive items, which were supposedly included, and he
all of
angrily criticizes his employee. He then marches out, after com-
miserating with the client on how shabbily he has been treated, and
the salesman looks helplessly and miserably at the client. The client
— better yet if it is a couple with a kindly wife infuriated by the —
insolent manner of the manager, and quite sympathetic to the sales-
man, may agree to pay the difference or at least part of it. The object
of the maneuver is to ally the client and the salesman against the

333
 Foundations of Family Therapy
manager; the client then protects the salesman against the (sup-
—
posed) anger of the manager and incidentally purchases a car.
Using same strategy, although we hope for more
a version of the
constructive ends, Papp prearranged for her team to call her out
toward the end of the fourth session with the family, and told them
when she came back in: 'They feel that I am wrong again in trying
to get you, Mother, to be strong and firm with John [a twelve-year-
old problem child who had been in individual therapy for six futile
years], because that means that father would have to be soft, and
since you seem to have such a nice balance going now and your roles
are clear, they're afraid it would upset the balance in your marriage
to change the relationship with the children. Their feeling is, your
marriage is more important than anything else, and if the children

have a few problems, that's a small price to pay for maintaining the
stability in your marriage/' 3
The effect was came
dramatic. In the following session the mother
in saying that she had completely reversed her usual role and had
been telling everybody off, especially her husband. The boy had
already begun to improve and, once the husband-wife issues were
out in the open, began to behave well at home and do well at school
too. To the parents' intense surprise, the boy was also able to tell
them that he had been worried for years that they would get
divorced.
In discussing the deliberate splitting of the team, Gillian Walker
has written that most efficacious if the therapist in
it is the room is
in favor of change, and the team takes the part of the "loyal opposi-
tion." The team can be skeptical, suggest outrageous intensifications
of the behaviors the family is already engaged in, and declare out-
right that the family can in no way succeed in making the changes
"their" therapist has been working for. To go back to the story of
Bateson's porpoise, the therapist is like the handler, giving the por-
poise unearned fish, while the team can continue to be the cold-
hearted experimenter.
Quite apart from the direct therapeutic uses of the team, there is
an unlooked-for special effect that people using the Milan Associ-
ates'format discover to be of inestimable value: the combined ener-
gies and talents of a small cell of people, behind a one-way screen,
seeing families in a peer or semipeer situation. Once started, these

334
ISSUES ON THE CUTTING EDGE
little foursomes, or sixsomes, can become a magnetic center in any
agency or institution where they take root. A form of practitioner-
research becomes possible without grants, immense outlays of
money (except somehow to get a one-way screen), or the deploy-

ment of many people or much on four

time. Four people can take
families, each seen by one or two therapists once every two weeks
or on a monthly basis. The effort to focus on a hypothesis and then
to plan interventions for each session, whether or not "paradoxical"
techniques are used, seems to stir up much enthusiasm and a lively
sense of participation, an effect which is bound to make waves
throughout any institution. It may be wise for these sporelike cells
to keep as quiet as possible about their activities for as long as
possible, for fear that too great a show of excitement will breed envy
and and produce efforts to extrude the too-
hostility in the host,
successful spore, now rapidly growing into a large, productive
mushroom.

Questions: Resolved and Unresolved

There are still, however, many questions for which there are no
easy answers. We can, for example, ask whether this systemic ap-
proach is so team-bound as to be of little value to the lone clinician
working with families. My own view is that it is of enormous value.
Many clinicians who have worked with teams report that a subtle
shift takes place that carries over to their individual work, as if the
colleagues behind the screen had taken up residence in the back of
the clinician's head like the tiny people children sometimes think
live inside the radio. I often feel while working alone that I am
consulting with these little people.
If I need to seek real advice, I will unhesitatingly use my col-
leagues as an ad hoc team. No therapist, no matter how indepen-
dent, should ever be without access to back-up professionals. Here
is where the team idea works as a blessed antidote to the notion that
the therapist should always be effective alone. I tell trainees who

335
 Foundations of Family Therapy
sometimes complain about their lack of success that without more
troops, without stronger ammunition, nothing will alter the situa-
tion and that they should not be ashamed about these "failures."
In assessing a trainee's performance with a family, one now must
add to the usual letter grades SWMH
(Struggle Without Much
Hope) and LLTG (Learning to Let Them Go).
Another question is whether the systemic approach, with or
without a team, is applicable to individuals or couples who are less
caught up in the family web, that is, to those who present them-
selves in smaller units and seem to have smaller problems. Of
course, one must determine whether these units are not, in fact,
deeply entangled in the family web, and hence whether one should
either join them to other familymembers or else redesign the treat-
ment to address this issue. The Milan Associates focus on problems
with large, messy relationship contexts, thus automatically elimi-
nating small units, but they will occasionally treat a couple relation-
ship in vitro. It is only when they speak of their private practice that
one hears of them dealing with individuals. Even then they claim
that their research has changed the way they work, although they
find it hard to explain the nature of this change.
There is a real question here as to what the practitioner must
address: the tip of the iceberg (the problem confined to a small
relationship unit) or the entire iceberg (the problem plus all rela-
tionships attached to it). The systemic approach seems to argue for
addressing the entire iceberg. The strategic therapists have shown,
however, that one can do very decent work even if limited to the
tip. Clinicians experimenting with systemic therapy are beginning
to feel that one can be limited to the tip and still treat the entire
— —
iceberg that is, still be "systemic" as long as one deals with the
iceberg in hypotheses and interventions.
A current argument proposes that "family therapy" should be
redefined as "systems therapy," and that the word "family" should
be eliminated on the grounds that it limits and confuses therapeutic
issues. Proponents of this argument believe that one works with
problem and context as an indivisible whole, and that this indivisi-
ble whole does not have to appear all at once in one's office. (Prag-
matically speaking, it never does.) It is possible that one can remain
faithful to the systemic vision even if one sees only an individual

336
ISSUES ON THE CUTTING EDGE
or a fragment of a family. It is up to the therapist to decide whether
working with this fragment provides enough leverage to bring
about change.
Another objection to the systemic approach is that since it is
based on work with "resistant" families, to use it on less severe
problems is like using a cannon to shoot a canary. Those who harbor
this reservation believe that one should first try a straightforward
approach to see how change-worthy the family or person is, and
should go on to "paradoxical" moves only after more conventional
interventions have failed. However, the reverse could be argued.
Some clinicians believe that these methods should be used first, to
break up the log jam, so to speak, before even attempting straight-
forward moves. They feel that this will shorten the therapy process
and make it more effective.

A question that will not (and perhaps should not) be resolved is

the relationship of the major schools of systems change to one

another. Are the systemic and strategic approaches basically the
same, or do they come from different conceptual universes, despite
their seeming likeness and common roots? If the latter is true, does
this split derive from their different ancestry, one school being
by Bateson and the other by Erickson? Again,
primarily influenced
does the common interest in family history and kinship which links
the systemic approach to the historical approaches, especially
Bowen's, imply that there should be a joining? If not, what are the
different premises that would prevent this?
By might the structural approach be less different from
contrast,
the systemic approach than it appears to be? The work of the Milan
Associates seems to be moving in a more structural direction, as
practitioners experiment with interventions that primarily entail a
revision of the family. One could say, of course, that all systemic
interventions touch the organization of the family, the "paradoxi-
cal" ones indirectly, and many of the rituals directly.
Finally, one cannot help but wonder what lies ahead. Will there
be a mingling of these tributaries into a mighty river? Will one
approach subsume the others? Or will each accentuate its borders
and harden into a sect? These issues will increasingly occupy clini-
cians and researchers in this swiftly expanding field.
In pursuit of answers to these difficult questions, I am working

337
 Foundations of Family Therapy
with a number of colleagues at the Ackerman Institute for Family
Therapy in New York: Olga Silverstein, Peggy Papp, Gillian
Walker, Joel Bergman, Peggy Penn, John Patten, and Jeffrey Ross.
In different groupings, and at different times, we have been strug-
gling with issues such as those outlined above. In particular, we
have been engaged in clinical research with both brief, strategic
techniques and the Milan model.
And here we face yet one more important problem the
issue. A
systemic approach raises for a teaching institute is whether new-
comers to family therapy should be exposed to this model from the
start. Fears are voiced about hordes of inexperienced paradoxical

therapists being loosed upon the world. My own sense is that some
of these fears are unjustified. The approach cannot be reduced to a
set of techniques, paradoxical or otherwise, but rather embodies a
huge epistemological shift. Being exposed to systemic work, even
without fully understanding it, can often help a beginner get to the
heart of a systems view of problems without sacrificing the com-
plexity of the matter or reducing to a blueprint what ought to be
experienced as a cathedral.
But despite cautionary advice there is bound to be an increasing
interest in experimenting with the systemic approach. More than
any other therapeutic design, it translates Bateson's abstract and
visionary formulations into an elegant pragmatics of clinical action.
An influential new paradigm, deeply concerned with clinical episte-

mology, is beginning to justify and explain the work of the Milan

Associates,and it is this paradigm and its implications for psycho-
therapy that I wish to talk about in the Epilogue to this book.

3)8
Epilogue

Toward a
New Epistemology

The Evolutionary Paradigm

This final chapter points toward the future but builds upon the
past. We have spoken of a circular or recursive epistemology. Bate-
son addresses this concern by stressing the self-recursive nature of
7
living forms, choosing for his metaphor that "odd worm/ Ourobor-
ous, the snake that eats its own tail. Expanding on his meaning, he
says:

We live in a universe in which causal trains endure, survive through time,

only if they are recursive. They "survive" — i.e. literally live upon themselves —
and some survive longer than others. 1

These recursive loops, however, are never totally closed, since there
is always a space for new information. Each cycle comes round to

a new position, sometimes so minutely different from the previous

one as to be imperceptible, but sometimes representing a major
shift.

For instance, take the evolution of family therapy from 1950 until
now. During these three decades the "systems" metaphor for family
groups, with its emphasis on homeostasis and equilibrium, became
a major model for the field. The analogy of the cybernetic machine,

339
 Foundations of Family Therapy
always returning to a presumed steady-state, was a convincing
analogy for the redundancies in interaction observed not only in
families with symptomatic members but in all families. It was also
useful in jogging the field out of earlier analogies that applied pri-
marily to the individual in isolation.
Recently, it has become clear that a new template is forming. A
powerful group of new ideas is challenging the equilibrium model,
not just for families, but for many other entities as well. During the
1970s a group of scientists in physics, chemistry, mathematics, and
other fields began to question the almost sacred Second Law of
Thermodynamics. This law held that all entities in the universe tend
toward a state of entropy: a gray, random sameness without move-
ment or change. Notable among researchers who began to challenge
this law was the physicist Ilya Prigogine. A number of physical and
chemical processes seemed exempt from it, he observed, and living
forms seemed almost to flout it. Living forms moved in a negen-
tropic direction much of the time, toward greater complexity and
new and different states. In addition, in the eternal, static realm of
classical physics, time, with its property of irreversibility, was ig-
nored. Tomorrow was the same as yesterday and only that which
never varied was a proper subject of study. The interesting case of
instabilities was ignored by physics, as was much else in the natural

world that changes and evolves. Prigogine says:

I often use the example of the cathedral and the brick. As long as you think
"brick," you see that the brick stays the same for millions of years. But
if you think "cathedral," there is the moment when it is built and the
moment when it falls into ruins. And with these same bricks, you can
construct cathedrals and palaces. 2

Living forms, and how they come into being, could not be addressed
by classical physics because it could only "think brick."
A central concept of Prigogine's, described in Chapter 9, is what
he calls "evolutionary feedback." This means that a movement that
is only a fluctuation in a system at one time can suddenly become
the basis for an entirely new arrangement of the system at another
time: "The evolution of the system goes spontaneously to situations
" Prigogine says, and this can be true for physical
that are less probable,

processes as well as life processes. 3

340
TOWARD A NEW EPISTEMOLOGY
Of particular relevance to family theory is the idea that living
systems have the capability of mutating to more complexly orga-
nized regimes through small fluctuations, usually random ones.
Two writers who have been linking family theory with this model
for nonequilibrium dynamics are the psychologist Paul Dell, whose
writing on this theme has been described earlier in this book, and
the psychiatrist Mony Elkaim of the Institute for Family and
Human Systems Study in Brussels. Both criticize the cybernetic
model for family systems and systems change accepted by many
family theorists. Current reasoning, according to Elkaim, states that,

The family is caught between two forces: one force that leads toward

change and a force that tends to preserve internal equilibrium. The

. . .

attempt was thus made to understand the symptom as having the function
of protecting the homeostasis, in that it protected the family from the
4
dangers represented by change. . . .

As an alternative, both Elkaim and Dell turn to the ideas of

Prigogine, who depicts all living forms as vulnerable to change,
especially when driven farfrom equilibrium. At this point, any
small instability can amplify, causing the system to pass beyond its
limits and in an almost magical way reappear in quite a different
state. The form that results from this process of evolutionary feed-
back is unpredictable, notes Prigogine. It is impossible to say which
of the system's fluctuations will drive it from its former state.

Applying model to therapy, Elkaim states that the task of the

this
therapist is to try to push the system away from equilibrium, forc-
ing it to search for a different solution; and above all, recognizing
the importance of chance in determining which instability will be
the determining one. The structure will be transformed according to
its own special laws, laws that the therapist cannot possibly guess
at, since they are part of what Elkaim respectfully calls the family's
"singularity." 5
It is interesting to consider that the systemic model pioneered by
the Milan Associates provides a living illustration of this theoretical
model, derived from such a different field. It is as if their therapeutic
approach had been founded expressly on the idea of evolutionary
feedback, rather than emerging separately and concurrently. The
attention to small perturbations that might lead a system away from

341
 Foundations of Family Therapy
equilibrium; the expectation of discontinuous change; the concern
for time and its irreversibility; the respect for the self-organizing
capabilities of the system — all this suggests an unusual congruence.
An added fascination for me (as for Prigogine) is the possibility
of developing theories that begin to unite living and nonliving
forms. 6 To go back to Bateson's words, quoted earlier, ''We might
expect to find the same sort of laws at work in the structure of a
crystal as in the structure of society. . .
." The development of an

evolutionary model for social change that so closely parallels laws

forchange from the world of chemistry and physics seems to offer
a hope that this prophecy might one day be fulfilled.

The Importance of Epistemology

For Bateson, the subject of epistemology was an intensely moral

concern. Epistemology for him meant the rules one uses for making
sense out of the world. These rules —not always conscious—deter-
mine much of our behavior and our interpretations of the behavior
of others. Bateson isolates two particular villains, epistemologically
speaking. One is "linear thinking/' which appears to assign a cause
and often ends up assigning blame. The other is any form of dual-
ism. In this passage, Bateson is going after one of the most outstand-
ing products of dualism, the idea of the "self":

To draw a boundary between a part which does most of the computa-

line
tion for a larger systemand the larger system of which it is a part, is to
create a mythological component, commonly called a "self." In my episte-
mology, the concept of self, along with all arbitrary boundaries which
delimit systems or parts of systems, be regarded as a trait of the local
is to
—
culture not indeed to be disregarded, since such little epistemological
monsters are always liable to become foci of pathology. The arbitrary
boundaries which were useful in the process of analyzing data become all
too easily battlefronts, across which we try to kill an enemy or exploit an
environment. 7

Bateson is always at pains to make it clear that he is talking about

total circuits. To "chop up the ecology" is what one does when one

342
TOWARD A NEW EPISTEMOLOGY
takes the parts and pieces of what one is describing and decides that
one part "controls" another or one part "causes" another.
Dell, applying Bateson's ideas to family therapy, points out the
tendency of many family therapists to negate the epistemological
revolution that the family therapy movement represents. 8 He par-
ticularly objects to the common use of the idea of homeostasis.
Family therapists have taken too literally the notion that a family
is like a homeostatic machine with a governor. Thus it is said that

a "family needs a symptom," or "a symptom serves a homeostatic

function in the family." To use this kind of language is to assume
a dualism between one part of the system and another part. It is
more correct to say that all parts are engaged in whatever ordering
of constancy or change is in question, in an equal and coordinate
fashion. To speak otherwise is to engage in what Dell describes as
a kind of "fuzzy systems animism."
What must be kept in mind is the continuous recursiveness of all
circuits in complex systems. It is not valid to say that the parents

are "using" the child's problems to keep them together. One could
just as well say that the child is using the parents' overprotective-
ness to keep him from the perils of leaving home; or that without
his problem, there would be no link between the mother and the
father's mother; or that a valued older child keeps being drawn back
home because of it; or that the problem child is the primary com-
forter of mother. Dell uses analogies to biology and other sciences:
"DNA is not a governor of biological systems; biological functions
are regulated by the total system of DNA
and cytoplasm." 9
Another Batesonian thinker, psychologist Bradford Keeney, pro-
poses an ecosystemic epistemology. 10 This framework prohibits
blaming the patient for his behavior or blaming his symptoms on
etiological factors. Instead, Keeney suggests that symptoms must be
seen as metaphoric communications about the ecology of the pa-
tient's relationship systems. Above all, he repeats the characteristic

Batesonian warning that the therapist must never consider himself

as an outside agent but rather as part of the therapeutic system, or
"part of an ecosystem." This position offers an alternative to the
formulations that place therapist versus patient in a power struggle
or game that the patient is always trying to "win."
Dell raises one problem that epistemological purity poses for

343
 Foundations of Family Therapy
many of us: When we reject dualism, we reject most of the concepts
we have been raised by, concepts that order our thinking —causal-
ity, purpose, objectivity, and the like. In fact, we cannot claim any
accuracy at all for our attempts to make statements about the world,
because we cannot set ourselves apart from that which we are
describing. The reality "out there" is unknowable because it

changes as we watch, and because our watching changes it.

11

The end point of this thinking is, of course, rather radical. State-
ments cannot be made about one's own experience because we are
spectators and participants at the same time, and our very grammar
we become totally mute, Bateson and those
violates this unity. Lest
following him state that we may go on committing epistemological
errors as long as we know we are committing them. Dell takes the
comforting position that as long as we are aware that we are always
operating in the context of a self-recursive network, it does not
matter what epistemologies we use or what theories we adopt:
"What matters is our awareness that both we and our chosen theory
participate in a self-recursive way in the emerging, evolving flow of
" 12
events. Bateson has aptly labeled this awareness 'wisdom/

The Second Generation

The new theoretical positions, the systemic approaches linked

with these positions, and the epistemological consciousness-raising
that provides a context for both are beginning to have an impact on
the family field, not just here but abroad. Some of the work in

Europe was originally influenced by American models; Satir has

heavily colonized Sweden, for instance, and Minuchin's structural
model has taken hold in several countries. Italian psychiatrist Mau-
rizio Andolfi's Institute of Family Studies in Rome offers an ap-
proach that blends the stress techniques of Minuchin with the quix-
in-the-room binds of Whitaker. The pressure-box model that
otic,

has resulted may become an alternative to the Milan approach,

which otherwise seems to be sweeping Europe.

344
TOWARD A NEW EPISTEMOLOGY
Other pioneers who have contributed to the international flower-
ing of family therapy are figures such as John Byng-Hall, Rosemary
Whiffen, and Robin Skynner London; Philippe Caille and Rolv
in
Blakar in Oslo; Helm Stierlin in Heidelberg; Luc Kaufman in Lau-
sanne; Yrjo Alanen in Helsinki; Geoffrey Goding in Melbourne;
Siegi Hirsch in Paris; and many others who will have to forgive me
for not mentioning them.
Even more important for Europeans, who are often more concept-
oriented and less pragmatic than Americans, is the development of
a body of theory for family therapy. For instance, my own observa-
tions suggest that more European clinicians interested in family
therapy have read Pragmatics of Human Communication in the last three
years than their counterparts in the United States have in the last
ten. The proliferation of international conferences, in which Euro-
peans are a very real presence, is also an indication of the increas-
ingly important position they are taking.
In America, there is an even more interesting picture. Here and
there, at universities younger clinicians and researchers
and clinics,

are asking about and struggling with a new kind of therapy and a
new way of thinking about therapy. This new way isn't Bowenian;
it it isn't strategic; and it isn't attached to any single
isn't structural;

therapist. some sense systemic but it is not necessarily mod-

It is in
eled on the work of the Milan Associates, even though most of the
energy stream of the seems to be emanating from that source.
field

It is profoundly Batesonian, and yet Bateson does not explicitly

address it.

What I believe we are witnessing is the emergence of a second

generation of family therapists clearly distinguishable from the first.

This second generation is not content with just a change in etiology.

To say that "The family, not the individual, is the 'cause' of the
problem" is not a real change for them. They are grasping the real
meaning of Bateson's thought, and they are understanding what the
strategic theorists in Palo Alto, the systemic practitioners in Milan,
and other voices in the United States and Europe have been trying
to convey: the need for a new epistemology. This new epistemology
would influence profoundly not only the way one thought about
therapy but also how one practiced it.

345
 Foundations of Family Therapy
Let me
address the issues that are being raised by this second
generation of family and systems thinkers and therapists. I will
highlight in particular the shift from a homeostatic to an evolution-
ary paradigm that Dell in particular has addressed, and the even
more extraordinary implications of a circular epistemology. The two
are in many ways entwined. Here are some of the ideas I believe will
be the benchmarks of the new generation.
1. An emphasis on circular rather than linear thinking is, of

course, basic. This applies very much to what used to be called

diagnosis, or a guess as to what "caused" the particular illness or
distress. A circular model takes one completely out of that frame.
Bateson says that wisdom is awareness of how all the circuits in the
system and are connected. This means not deciding that the
fit

family has "caused" the individual's problem, any more than the
other way round. No one element takes precedence over or controls
another. Brodey once described a case in exactly those terms, and
with the economy of a poet:

'They never let me 7

leave the house/ the child states. "We tried but he
7
always got lost/ the parent states. And so the circle now spirals through
time. 13

2. We must change from the idea of causation to a concept that

is nearer that of "fit." This concept is beautifully developed by Dell
in "Beyond Homeostasis." Describing the difference, Dell says,

Without making reference to etiology or causation, fit simply posits that

the behaviors occurring in the family system have a general complemen-
tarity; they fit together. Causation, on the other hand, is a specified type of
interpretation of fit that considers the observed complementarity to have
the form: A causes B. For instance, bad parents make their children sick.
14

3. We
must add positive to the usual negative interpretations of
symptomatic behavior, not just as a strategy for change, but because
doing so adds a layer of complexity that guards against linear think-
ing. Instead of assuming that a symptom is a kind of minus sign
indicating a dysfunctional family, we may regard it instead as the
one factor that keeps pushing the family toward a new and different
state.

346
TOWARD A NEW EPISTEMOLOGY
I first began to think of symptoms as harbingers of change when
I was learning how to put together a "positive connotation/' The
work involved in this process amounts to finding an organizing
principle for the presence of a problem at the family system level.
In addition, one can often construe the symptom as a solution,
however uncomfortable or destructive it may seem, to a dilemma
faced by the family on its evolutionary path. This is not to say that
a positive connotation is thereby "true," or that it is not just as
two "linears" can make a
linear as a negative one, but in this case,
"circular," much as Yin plus Yang can make a round.

4. We would have to once more legitimize Time, banned for so

long as a piece of useless psychotherapeutic baggage. Current think-
ing about living systems emphasizes that the processes of life are
always irreversible. Nothing can ever go back or step in the same
river twice. A shift in the entire gestalt of a system, in particular,

can never be reversed, witness again the kaleidoscope.

5. We would have to accept the concept of unpredictability. It

seems that major systems changes can never be foreseen, although

in some cases probabilities can usefully be assessed. This means
abandoning our emphasis on goals and giving more attention to
chance. We will have to substitute a respect for what Elkaim calls
the "singularity" of each family, with its universe of possible solu-
tions, rather than imposing our own ideas of what a family ought
to look like. Families can think up far more amazing solutions than
we can.
6. We
would have to abandon the idea of the therapist as a
bullfighter, pushing and pulling the family to where he wanted
it. If we are going to give up the notion of a Newtonian universe,

with forces acting upon things, we are going to have to give up the
notion of the therapist as a force, acting upon a client or a fam-
ily. The careful way the Milan Associates position themselves in

the therapeutic field and their emphasis on what they call "neu-
trality" express this idea, as does Bateson in his epistemological
thinking.
7. We would have to give up the traditional idea of resistance,
usually thought of as an oppositional trait residing in the client or
family. The idea that the client is "resistant," or that the family

347
 Foundations of Family Therapy
"homeostasis" causes it to resist, is totally linear. As Dell says, "the
system does not 'resist/ it only behaves in accordance with its own
15
coherence."
Homeostasis is the counterpart in family therapy of the concept
of resistance in individual therapy. more accurate to describe It is

resistance as the place where the therapist and client or family

intersect. Resistance is merely an artifact of that time and place. In
addition, we can think positively about resistance, since it often
generates the momentum needed to accomplish change. It is a fact
that all strategic/systemic therapists use this judo technique in their
work.
8. We must learn to favor instability over equilibrium. If Prigo-
gine is right, living instabilities. A town, for
systems are permanent
instance, is always changing as flows of people and goods come in
and out. Evolving systems might even be seen as going from a state
of instability, to states characterized by relative rigidity, to new
instabilities. This is the reverse of classical physics and the reverse
of common sense. It also has some trying implications for therapists,
alone their clients, who may not take kindly to the message:
let

Come in and I will get you out of this state of anxiety, nervousness,
depression you are in and make you unstable again.
9. In "Beyond Homeostasis," Dell replaces homeostasis with a
new concept, that of coherence. Coherence has to do with how
pieces of a system fit together in a balance internal to itself and
external to its environment. Homeostasis has a more rubbery,
punchy feel to it than coherence, which is why I hate to give it up
(and won't) but coherence is purer in the epistemological sense. I

also like coherence because it has to do with the family in a field.

The family has to fit with its environment, just as the individual has

to fit within the family, or the separate organs have to fit together
in a system that is the biological self. And all have to fit together
in the ecology of the whole.
I would like to end here with an excerpt from the Notebooks of Paul
Klee, because it so well expresses the position I myself have come
to:

Formerly, artists depicted things that were to be seen on the earth, things
people liked to see or would like to have seen. Now the relativity of visible

34 S
TOWARD A NEW EPISTEMOLOGY
things is made clear, the belief expressed that the visible is only an isolated
case taken from the universe, and that there are often more truths unseen
than seen. . . .

A few examples: A man of antiquity, sailing a boat, quite content and

enjoying the ingenious comfort of the contrivance. The ancients represent
the scene accordingly. And now: What a modern man experiences as he
walks across the deck of a steamer: 1. his own movement, 2. the movement
of the ship which may be in the opposite direction, 3. the direction and
velocity of the current, 4. the rotation of the earth, 5. its orbit, 6. the orbits
of the moons and planets around it.

Result: an interplay of movements in the universe, at the centre "I" on

the ship.

An apple tree in blossom, the roots, the rising sap, the trunk, a cross
section with annual rings, the blossom, its structure, its sexual functions,
the fruit, the core and seeds. An interplay of states of growth.

A sleeping man, the circulation of his blood, the measured breathing of

the lungs, the delicate function of the kidney, in his head a world of dreams
related to the powers of fate. An interplay of functions, united in rest. 16

The epistemology of this passage is circular not linear. It is mul-

tifaceted and therefore systemic. It sets no part over any other part,
so it is not dualistic. It does not chop the ecology to pieces, so it is

holistic. It links the viewer of one period to the viewer of another,

so it is recursive. And it is evolutionary because it highlights a shift
toward greater complexity between these different times.
The above passage also represents a metaphor for the many-
layered perspectives of this book. From the early 1950s, when the
family movement began, to the year 1980, when its two presiding
geniuses, Gregory Bateson and Milton Erickson, died, there has
been the span of a generation. This Epilogue shall fittingly be dedi-
cated to the new generation. After all, we are talking about an
evolutionary paradigm.

349
Notes

Prologue: Behind the Looking Glass

1. Bateson, G., Mind and Nature. New York: E. P. Dutton, 1979.

2. Jackson, D. D., "The Question of Family Homeostasis," The Psychiatric Quarterly Supple-
ment 31 (1957), 79-90-
3. Dell, P.,and H. Goolishian, "Order Through Fluctuation," Speech at the Annual
Scientific Meeting of the A. K. Rice Institute, Houston, Texas, 1979.
4. Bateson, Mind and Nature, p. 41.
5. Bateson, M. C, "Daddy, Can a Scientist Be Wise?" in Brockman, J. (ed.), About Bateson.
New York: E. P. Dutton, 1977, p. 65.
6. Bateson, G., "The Birth of a Double Bind," in Berger, M. (ed.), Beyond the Double Bind.
New York: Brunner/Mazel, 1978, p. 53.
7. Bateson, ibid., p. 45.
8. Hoffman, L., "Deviation-Amplifying Processes in Natural Groups," in Haley, J. (ed.),
Changing Families. New York, Grune and Stratton, 1971.
9. Speer, A., "Family Systems: Morphostasis and Morphogenesis," Family Process 9 (1970),
259-278.
10. Dell, P., "Researching the Family Theories of Schizophrenia: An Exercise in Epistemo-
logical Confusion," Family Process 19 (1980), 321-335.

Chapter 1 Early Research on Family Groups

1. Guerin offers developments and figures in the field; see Guerin, P., Family
a history of
Therapy: Theory and Practice, New
York: Gardner Press, 1976. For a clear presentation of the
major shift in ideas represented by this group, see "A Review of the Family Therapy Field,"

351
 NOTES
in Haley,J., Changing Families, New
York: Grune and Stratton, 1971. And for an excellent
critique of eaarly studies on the family and schizophrenia, see Paul Dell, "Researching the
Family Theories of Schizophrenia: An Exercise in Epistemological Confusion," Family Process
19 (1980), 3*1-335-
2. Watzlawick, P., D. Jackson, and J. Beavin, Pragmatics of Human Communication. New York:
W. W. Norton, 1967.
3. Jackson, D. D., "The Question of Family Homeostasis," Psychiatric Quarterly Supplement

31 (1957), 79-90.
4. Haley, J., Strategies of Psychotherapy. New York: Grune and Stratton, 1964, p. 189.
5. Haley, J., "Research on Family Patterns: An Instrument Measurement," Family Process

3 (1964), 41-65.
6. Haley, J., and L. Hoffman, Techniques of Family Therapy. New York: Basic Books, 1968,
p. 227.
7. Bateson, G., D. Jackson, J.Haley, and J. Weakland, "Toward a Theory of Schizophre-
nia," Behavioral Science 1 (1956), 251-254.
8. Haley, J., "The Family of the Schizophrenic: A Model System," Journal of Nervous and
Mental Disease 129 (1959), 357-374-
9. Jackson, D. D., Therapy, Communication and Change and Communication, Family and Marriage.
Palo Alto, Calif.: Science and Behavior Books, 1967.
10. Weakland, J., and D. D. Jackson, "Patient and Therapist Observations on the Circum-
stances of a Schizophrenic Episode," A.M. A. Archives of Neurology and Psychiatry 79 (1958),
554-574 Jackson, D. D., and I. Yalom, "Conjoint Family Therapy as an Aid to Intensive
Psychotherapy," in Burton, A. (ed.), Modern Psychotherapeutic Practice. Palo Alto, Calif.: Science
and Behavior Books, 1965, pp. 80-98. Weakland, J., and W. Fry, "Letters of Mothers of
Schizophrenics," American Journal of Psychiatry 32 (1962), 604-623.
11. Reprinted in Sluzki, C, and D. Ransom (eds.), Double Bind: The Foundation of the Communi-
cational Approach to the Family. New York: Grune and Stratton, 1976.
12. Sluzki, C, and D. Ransom (eds.), Double Bind: The Foundation of the Communicational Approach
to the Family. New York: Grune and Stratton, 1976.

13. Ibid., pp. 23-37.

14. Bateson, G., D. Jackson, J. Haley, and J. Weakland, "A Note on the Double Bind
1962," Family Process 2 (1963), 154-161.
15. Watzlawick, P., "A Review of the Double Bind Theory," Family Process 2 (1963),
132-153.
16. Weakland, J., "The Double Bind Theory by Self-Reflexive Hindsight," Family Process
13 (1974), 269-277. Bateson, G., "The Birth of a Matrix or Double Bind and Epistemology,"
in Berger, M. (ed.), Beyond the Double Bind. New York: Brunner/Mazel, 1977.
17. Bateson, G., Steps to an Ecology of Mind. New York: Ballantine Books, 1971, p. 241.
18. Neumann, J. von, and O. Morgenstern, Theory of Games and Economic Behavior. Princeton,
N.J.: Princeton University Press, 1947.
19. Bateson, Steps to an Ecology of Mind, p. 240.
20. Haley, J., "Development of a Theory," in Sluzki, C, and D. Ransom (eds), Double Bind.
21. Bateson, Steps to an Ecology of Mind, p. 236.
22. Paul Dell. Personal communication.
23. Haley, J., "Development of a Theory," in Sluzki, C, and D. Ransom (eds.), Double Bind,

p. 78.
24. Haley, J., Strategies of Psychotherapy. New York: Giune and Stratton, 1963.
25. Lennard, H., and A. Bernstein, Patterns in Human Interaction. San Francisco, Calif.: Jossey-
Bass, 1970, p. 134.
26. Whitehead, A. N., and B. Russell, Principia Mathematica. Cambridge: Cambridge Univer-
sity Press, 1910.
27. Riskin, J., and E. Faunce, "An Evaluative Review of Family Interaction and Research,"
Family Process 11 (1972), 365-455.
28. Bowen, M., Family Therapy in Clinical Practice. New York: Jason Aronson, 1978.
29. Bowen, M., "The Use of Family Theory in Clinical Practice," Clinical Psychiatry 7
(1966), 345-374-
30. Wynne, L. C, "The Study of Intrafamilial Splits and Alignments in Exploratory

352
NOTES
Family Therapy," in Ackerman, N. (ed.), Exploring the Base for Family Therapy. New York: Family
Service Association of America, 1961, pp. 95-115.
31. Ibid.
32. Wynne, L. C, et al., "Pseudo-Mutuality in the Family Relations of Schizophrenics,"
Archives of General Psychiatry 9 (1963), 161-206.
33. Ibid., p. 206.
34. Laing, R. D., and A. Esterson, Sanity, Madness, and the Family. New York: Basic Books,
1971.

Chapter 2 The Dynamics of Social Fields

1. Bateson, G., Steps to an Ecology of Mind. New York: Ballantine Books, 1972, p. 7 A-
2. Bateson, G., Naven. Stanford, Calif.: Stanford University Press, 1958 (rev. ed.).
3. Bateson, Steps to an Ecology of Mind, p. 77.
4. Bateson, Naven, p. 175.
5. Boulding, K., Conflict and Defense. New York: Harper and Row, 1963.
6. Bateson, Steps to an Ecology of Mind., pp. 107-112.
7. Bateson, Naven, p. 194.
8. Barth, F., "Segmentary Opposition and the Theory of Games: A Study of Pathan
Organization," Journal of the Royal Anthropological Institute 89 (1959)/ 5-21.
9. Bateson, Steps to an Ecology of Mind, p. 110.
10. Bateson, Naven, p. 289.
11. Ashby, W. R., Design for a Brain. New York: Wiley, 1952.
12. Nett, R., "Conformity-Deviation and the Social Control Concept," in Buckley, W.
(ed.), Modern Systems Research for the Behavioral Scientist. Chicago: Aldine, 1968.

Chapter 3 The Second Cybernetics

1. Maruyama, M., "The Second Cybernetics: Deviation- Amplifying Mutual Causal Pro-
cesses," in Buckley, W. (ed.), Modern Systems Research for the Behavioral Scientist. Chicago: Aldine,
1968, p. 304.
2.Wiener, N., The Human Use of Human Beings. New York: Anchor Books, 1954, p. 25.
3.Hardin, G., "The Cybernetics of Competition: A Biologist's View of Society/' in
Shepard, P., and D. McKinley (eds.), The Subversive Science: Essays Toward an Ecology of Man.
Boston: Houghton Mifflin, 1969, pp. 275-295.
4. Schroedinger, E. What Is Life? Cambridge: Cambridge University Press, 1945.
5. Buckley, W., "Society as a Complex Adaptive System," in Buckley, W. (ed), Modern
Systems Research for the Behavioral Scientist. Chicago: Aldine, 1968, p. 491.
6. Reiss, D., "The Working Family: A Researcher's View of Health in the Household."
Distinguished Psychiatrist Lecture, Annual Meeting, American Psychiatric Association, San
Francisco, Calif., 1980.
7. Buckley, "Society as a Complex Adaptive System," p. 500.
8. Maruyama, "The Second Cybernetics," p. 312.
9. Minuchin, S., and A. Barcai, "Therapeutically Induced Family Crisis," in Masserman,
J. H. and Psychoanalysis. New York: Grune and Stratton, 1969.
(ed.), Science
10. Bateson, G., Naven. Stanford, Calif.: Stanford University Press, 1958 (rev. ed.), p. 197.
11. Simon, H., "Comments on the Theory of Organization," American Political Science Review
46 (1952), 1130-1139.

353
 NOTES
12. Ruesch, J., and G. Bateson, Communication: The Social Matrix of Society. New York: W. W.
Norton, 1951, p. 287.
13. Ibid., p. 289.
14. Merton, R., On Glencoe, 111.: Free Press, 1967, p. 115.
Theoretical Sociology.

15. Haley, J., Strategies of Psychotherapy. New

York: Grune and Stratton, 1963, Ch. 1.
16. Wilkins, L. T., "A Behavioral Theory of Drug Taking," in Buckley, W. (ed.), Modern
Systems Research for the Behavioral Scientist. Chicago: Aldine, 1968, pp. 421-427.
17. Vogel, E. F., and N. W. Bell, "The Emotionally Disturbed Child as the Family Scape-
goat," in Bell, N. W., and Ezra F. Vogel (eds.), The Family. Glencoe, 111.: Free Press, i960, pp.
382-397.
18. Dentler, R. A., and K. T. Erikson, "The Functions of Deviance in Groups," Social
Problems 7 (1959), g8-l07.
19. Daniels, A., "Interaction Through Social Typing: The Development of the Scapegoat
in Sensitivity Training Sessions." Mimeographed manuscript.
20. Haley, J., and L. Hoffman, Techniques of Family Therapy. New York: Basic Books, 1967,
p. 205.
21. Haley, J., "Toward a Theory of Pathological Systems," in Zuk, G. H., and I. Boszor-
menyi-Nagy (eds.), Family Therapy and Disturbed Families. Palo Alto, Calif.: Science and Behavior
Books, 1969, pp. 11-27.
22. Lederer, W. J., and D. D. Jackson, The Mirages of Marriage. New York: W. W. Norton,
1968, pp. 161-173.
23. Jackson, D. D., "The Role of the Individual." Address to Conference on Mental Health
and the Idea of Mankind, Annual Meeting, Council for the Study of Mankind, Chicago, 1964.
24. Taylor, W., "Research on Family Interaction I: A Methodological Note," Family Process
9 (1970), 221-232.
25. Bales, R., "In Conference," in Etzioni, A. (ed.), Readings on Modern Organizations. Engle-
wood Cliffs, N.J.: Prentice-Hall, 1969, p. 150.
26. Reiss, "The Working Family."

Chapter 4 Typologies of Family Structure

1. Singer, M. T., and L. C. Wynne, "Differentiating Characteristics of Parents of Child-

hood Schizophrenics, Childhood Neurotics, and Young Adult Schizophrenics," American Jour-
nal of Psychiatry 120 (1963), 234-243.
2. Jackson, D. D., "Family Interaction, Family Homeostasis and Some Implications for
Conjoint Family Therapy," in Masserman, J. (ed.), Individual and Familial Dynamics. New York:
Grune and Stratton, 1959.
Jackson, D. D., The Mirages of Marriage. New York: W. W. Norton, 1968.
3.
4. Watzlawick, W., D. D. Jackson, and J. Beavin, Pragmatics of Human Communication. New
York: W. W. Norton, 1967, p. 110.
5. 110-117.
Ibid., pp.
6. 107-108.
Ibid., pp.

7. Minuchin, S. et al., Families of the Slums. New York: Basic Books, 1969.
8. Ibid., p. 352.
9. Ibid., p. 358.
10. Minuchin, S., Families and Family Therapy. Cambridge, Mass.: Harvard University Press,
1974-
11 248-249.
Ibid., pp.
12. Ashby, W. R., Design for a Brain. London: Chapman and Hall, Science and Behavior
Books, 1969, p. 79.
13. Ibid., p. 154.
14. Ibid., p. 155.
15. Ibid., p. 207.

354
NOTES
16. Ibid., p. 208.
17. Ibid., p. 210.
18. Bowen, M., "The Use of Family Theory in Clinical Practice," Comprehensive Psychiatry

7 (1966), 345-374-
19. Stierlin, Helm, Separating Parents and Adolescents.New York: Quadrangle/New York
Times Book Co., 1972.
20. Thomas, Lewis, Lives of a Cell. New York: Bantam Books, 1974/ PP- 62-63.
21. Minuchin et al., Families of the Slums, p. 355.
22. See Aponte, H., "Underorganization in the Poor Family," in Guerin, P., Family Therapy:
Theory and Practice, New York: Gardner Press, 1976, for a more general discussion of this type
of family.

Chapter j The Concept of Family Paradigms

1. Dell, P., "Researching the Family Theories of Schizophrenia," Family Process 19 (1980),
321-335-
2. Reiss, D., "Varieties of Consensual Experience," Family Process 10 (1971), 1-35.
3. Ibid., p. 6.
4. Ibid., p. 4.
5. Hess, R., and G. Handel, Family Worlds. Chicago: University of Chicago Press, 1959.
6. Wertheim, E., "Family Unit Therapy and the Science and Typology of Family Sys-
tems," Family Process 12 (1973), 361-376.
7. Beavers, W. R., "A Systems Model of Family for Family Therapists." Unpublished
manuscript. See also Beavers, W. R., Psychotherapy and Growth, New York: Brunner/Mazel,
1977-
8. Haley, J., Leaving Home. New York: McGraw-Hill, 1980.
9. Kantor, D., and W. Lehr, Inside the Family. San Francisco, Calif.: Jossey-Bass, 1975.
10. Reiss, D., "The Working Family: A Researcher's View of Health in the Household."
Distinguished Psychiatrist Lecture, Annual Meeting, American Psychiatric Association, San
Francisco, Calif., 1980.
11. Ibid., p. 32.
12. Fivaz, R., "Une Evolution Vers l'lmpasse?" Polyrama (Ecole Poly technique Federale de
Lausanne), January 1980, no. 45, pp. 9-11.

Chapter 6 The Pathological Triad

i. Weakland, J., "The Double Bind Hypothesis of Schizophrenia and Three-Party In-
teraction," in Sluzki, C, and D. Ransom (eds.), Double Bind: The Foundation of the Communicational
Approach to the Family. New York: Grune and Stratton, 1976.
2. Ibid., p. 29.
3. Wynne, L. C, and Singer, M. T., "Thought Disorder and Family Relations of Schizo-
phrenics: I. A Research Strategy. II. A Classification of Forms of Thinking," Archives of General
Psychiatry 9 (1963), 191-206.
4. Lidz, T.,A. R. Cornelison, S. Fleck, and D. Terry, "Schism and Skew in the Families
of Schizophrenics," in Bell, N. W., and E. F. Vogel (eds.), A Modern Introduction to the Family.
Glencoe, 111.: Free Press, i960, pp. ^g^-607.
5. Stanton, A., and M. Schwartz, The Mental Hospital. New York: Basic Books, 1954.

355
 NOTES
6. Haley, "The Family of the Schizophrenic: A Model System," Journal of Nervous and
J.,

Mental Disease, 129 (1959), 357-374.

7. Haley, J., "Development of a Theory," in Sluzki, C, and D. Ransom (eds.), Double Bind.
8. Haley, J., "Toward a Theory of Pathological Systems," in Watzlawick, P., and J.
Weakland (eds.), The Interactional View. New York: W. W. Norton, 1977.
9. Ibid., p. 37.
10. Caplow, Two Against One. Englewood Cliffs, N.J.: Prentice-Hall, 1968.
T.,
11. Haley, "Pathological Systems," p. 38.
12. Haley, "Pathological Systems," p. 41.
13. Davis, J., "Structural Balance, Mechanical Solidarity, and Interpersonal Relations,"
American Journal of Sociology 6S (1963), 444-462.
14. Boulding, K., Conflict and Defense: A General Theory. New York: Harper and Row, 1963,
p. 83.
15. Caplow, Two Against One, p. 78.
16. Haley, "Development of a Theory," p. 81.
17. Ruesch, J., and G. Bateson, Communication: The Social Matrix of Psychiatry. New York: W.
W. Norton, 1951, p. 193.
18. Ibid., p. 196.
19. Haley, J., Strategies of Psychotherapy. New York: Grune and Stratton, 1963, Ch. 5.
20. Haley, "Pathological Systems," pp. 42-44.
21. Ibid., p. 44.
22. Haley, J., The Power Tactics of Jesus Christ. New York: Grossman, 1969.
23. Wynne, L., "Intrafamilial Splits and Alignments in Exploratory Family Therapy," in
Ackerman, N., et al. (eds.), Exploring the Base for Family Therapy. New York: Family Service
Association of America, 1961.
24. Haley, "Pathological Systems," p. 40.
25. Caplow, Two Against One, p. 106.
26. Haley, "Pathological Systems," p. 39.
27. Hsu, F., "Kinship and Ways of Life: An
Explanation," in Psychological Anthropology:
Approaches to Culture and Personality. Homewood,
Richard D. Irwin, 1961.
111.:

28. Ross, A., "The Substructure of Power and Authority," in Barash, M., and A. Scourby
(eds.), Marriage and the Family. New York: Random House, 1970, p. 86.
29. LeVine, R., "Intergenerational Tensions and Extended Family Structures in Africa,"
in Barash and Scourby, Marriage and the Family, pp. 144-164.

Chapter 7 The Rules of Congruence for Triads

i. Cartwright, C. and F. Harary, "Structural Balance: A Generalization of Heider's The-

ory," Psychological Review 63 (1956), 277-293.
2. Taylor, W., "Research on Family Interaction: Static and Dynamic Models," Family
Process g, ig70, 221-232.
3. Caplow, T., Two Against One: Coalitions in Triads. Englewood Cliffs, N.J.: Prentice-Hall,

1968, p. 59.
4. Davis, J., "Clustering and Structural Balance in Graphs," Human Relations 20 (1967),
181-187.
5. Wolff, Kurt H., The Sociology of Georg Simmel. New York: Free Press, Collier-Macmillan,
1950, 141.
p.
6. Coser, L., The Functions of Social Conflict. New York: Free Press-Collier, 1969, Ch. 4.

7. Davis, "Clustering," p. 187.

8. Apple, D., "The Social Structure of Grandparenthood," American Anthropologist 53
(1956), 656-663.
9. Bott, E., Family and Social Network. London: Tavistock Publications, 1957-

356
NOTES
10. Flomenhaft, K., and D.M. Kaplan, "Clinical Significance of Current Kinship Relation
ships," Social Work, Jan. 1968, 68-74.
11. Taylor, H. F., Balance in Small Groups. New
York: Van Nostrand-Reinhold, 1970.

Chapter 8 Triads and the Management of Conflict

1. Freilich, M., "The Natural Triad in Kinship and Complex Systems," American Sociological

Review 29 (1964), 529-540.

2. Stanton, A., and M. Schwartz, The Mental Hospital. New York: Basic Books, 1964.
3. Ibid., p. 345-
4. Ibid., p. 363.
Minuchin, S., B.L. Rosman, and
5. L. Baker, Psychosomatic Families. Cambridge, Mass.:
Harvard University Press, 1978.

Chapter 9 The Simple Bind and Discontinuous Change

1. Mind and Nature. New York: E. P. Dutton, 1978, p. 194.

Bateson, G.,
2. and H. Goolishian, "Order Through Fluctuation: An Evolutionary Epistemol-
Dell, P.,
ogy for Human Systems." Presented at the Annual Scientific Meeting of the A. K. Rice
Institute, Houston, Texas, 1979.
3. Bateson, G., Mind and Nature, pp. 47-48.
4. Dell and Goolishian, "Order Through Fluctuation," p. 10.
5. Piatt, J., "Hierarchical Growth," Bulletin of Atomic Scientists (November 1970), 2-4,
14-16.
6. Ashby, W. R., Design for a Brain. London: Chapman and Hall, Science Paperbacks, i960.
7. Bateson, Mind and Nature, p. 98.
8. Erikson, E., Childhood and Society. New York: W. W. Norton, 1963.
9. Lindemann, Eric, "Symptomatology and Management of Acute Grief," in Parad, H.
J., and G. Caplan (eds.), Crisis Intervention: Selected Readings. New York: Family Service Associa-
tion of America, 1969, p. 18.
10. Rapoport, L., "The State of Crisis: Some Theoretical Considerations," in Parad and
Caplan, Crisis Intervention, p. 23.
Reuben, Families Under Stress. New York: Harper and Bros., 1949.
11. Hill,
Solomon, M., "A Developmental Premise for Family Therapy," Family Process 12
12.
(1973), 179-1S8-
13. Eliot, Thomas D., "Handling Family Strains and Shocks," in Becker, Howard, and
Reuben Hill (eds.), Family, Marriage and Parenthood. Boston: Heath and Co., 1955.
14. LeMasters, E. E., "Parenthood as Crisis," in Parad and Caplan, Crisis Intervention, pp.
111-117.
15. Holmes, T. H., and R. H. Rahe, "The Social Readjustive Rating Scale," Journal of
Psychosomatic Research 11 (April 1967), 213-218.
16. Ibid., p. 215.
17. Haley, "The Family Life Cycle," in Uncommon
J., Therapy: The Psychiatric Techniques of
Milton Erickson, M.D. New York: W. W. Norton, 1973.
18. Ashby, Design fa a Brain.

357
 NOTES
19. Ibid., pp. 87-89.
20. Gramsci, Antonio, Prison Notebooks:
Selections, trans. Quintin Hoare and Geoffrey N.
Smith. New
York: International Publishing Co., 1971, p. 71.
21. Reiss, D., "The Working Family: A Researcher's View of Health in the Household,"
Distinguished Psychiatrist Lecture, Annual Meeting of the American Psychiatric Association,
San Francsco, 1980.
22. Rabkin, R., "A Critique of the Clinical Use of the Double Bind," in Sluzki, C, and
D. Ransom (eds.), Double Bind: The Communicational Approach to the Family. New York: Grune and

Stratton, 1976, pp. 287-306.

23. Sluzki, C, et al., "Transactional Disqualification: Research on the Double Bind," in
Watzlawick, P., and J. Weakland (eds.), The Interactional View. New York: W. W. Norton, 1977,
p. 217.
24. Rabkin, "Critique of the Double Bind," and Ransom, Double Bind, p. 297.
in Sluzki
25. Bateson, G. Steps to an Ecology of Mind. New
York: Ballantine Books, 1972, p. 277.
26. Wynne, L., "On the Anguish and Creative Passions of Not Escaping the Double Bind,"
in Sluzki and Ransom, Double Bind, pp. 243-250.

Chapter 10 The Thing in the Bushes

Bateson, G., Mind and Nature. New York: Holt, Rinehart, and Winston, 1979, p. 103.
1.

Ravich, R., Predictable Pairing. New York: Peter H. Wyden, 1974, p. 269.
2.

3. Wild, C, L. Shapiro, L. Goldenberg, "Transactional Disturbances in Families of Male

Schizophrenics," Family Process 14 (1975), 131-160.
4. Lidz, T., et al., "The Intrafamilial Environment of Schizophrenic Patients: II. Marital
Schism and Marital Skew," American Journal of Psychiatry 114 (1957): 241-248.
5. Lederer, W., and D. D. Jackson, Mirages of Marriage. New York: W. W. Norton, 1968.
6. Ravich, Predictable Pairing, Ch. 7.

7. Steinglass, P., I. D. Davis, and D. Berenson, "Observations of Conjointly Hospitalized

'Alcoholic Couples' During Sobriety and Intoxication," Family Process 16 (1977), 1-16.
8. Berman, E., C. Pittman, and V. Ratliffe, "A Relational Approach to Spouse Abuse."
Unpublished manuscript.
9. Sampson, H., S. L. Messinger, and R. D. Towne, "Family Processes and Becoming a
Mental Patient," American Journal of Sociology 68 (1962), 88-96.
10. Sluzki, C. "Marital Therapy from a Systems Therapy Perspective," in Paolino, T. J.,
and B. S. McCrady (eds.), Marriage and Marital Therapy. New York: Brunner/Mazel, 1978;
Papp, P., "The Use of Fantasy in a couples Group," in Andolfi, M., and I. Zwerling (eds),
Dimensions of Family Therapy. New York: Guilford Press, 1980; Paul, N. and B. Paul, A Marital
Puzzle.New York: W. W. Norton, 1975; Sager, C, Marriage Contracts and Couple Therapy. New
York: Brunner/Mazel, g76.
11. Erikson, Kai T., Everything in Its Path. New York: Harper & Row, 1978.
12. Chappie, E. D., Culture and Biological Man. New York: Holt, Rinehart, and Winston,
1970.
13. Ibid., p. 48.
14. Raush, H., et al., Communication, Conflict and Marriage San Francisco, Calif.: Jossey-Bass,
1974-

35S
NOTES

Chapter 1 1 Breaking the Symptomatic Cycle

Watzlawick, P., J. Weakland, Jr., and R. Fisch, Change: Problem Formation and Problem
1.

Resolution.New York: W. W. Norton, 1974/ P- 9-

2. Dell, P., and H. Goolishian, "Order Through Fluctuation: An Evolutionary Epistemol-
ogy for Human Systems." Presented at the Annual Scientific Meeting of the A. K. Rice
Institute, Houston, Texas, 1979.
3. Minuchin, S., Psychosomatic Families. Cambridge, Mass.: Harvard University Press, 1978.
4. Ibid., pp. 165-66.
5. Haley, J., and L. Hoffman, Techniques of Family Therapy. New York: Basic Books, 1967,
p. 6.
6. Ibid., p. 63.
Haley, J., "Strategic Therapy when a Child is Presented as the Problem," Journal of the
7.
American Academy of Child Psychiatry 12 (1973), 64-74.

Chapter 12 Family Therapy and the Great Originals

1. Madanes, C, and J. Haley, "Dimensions of Family Therapy," Journal of Nervous and

Mental Disease 165 (1977), 88-98.

2. Satir, V., Conjoint Family Therapy. Palo Alto, Calif.: Science and Behavior Books, 1964.

3. Haley, J., and L. Hoffman, Techniques of Family Therapy. New York: Basic Books, 1967.
Ch. 2.

4. Ackerman, N. "The Family

as a Social and Emotional Unit," Bulletin of the Kansas Mental
Hygiene October, 1937.
Society,

5. Ackerman, N. and P. Franklin, "Family Dynamics and the Reversibility of Delusional

Formation: A Case Study in Family Therapy," in Boszormenyi-Nagy, I. and J. Framo (eds.),
Intensive Family Therapy. New York: Harper and Row, 1965, Ch. 6.
6. Ackerman, N., Treating the Troubled Family. New York: Basic Books, 1966.

7. Whitaker, C, "Power Politics of Family Psychotherapy." Presented at the American

Group Psychotherapy Association Conference, February 1973.
8. Ibid.
Whitaker, C, "Psychotherapy of the Absurd," Family Process 14 (1975), 1-16.
9.
10. Napier, A. Y., with Carl Whitaker, The Family Crucible. New York: Harper and Row,
1978.
11. Whitaker, C, "Psychotherapy of the Absurd," p. 11.
12. Haley, J., Uncommon Therapy: The Psychiatric Techniques of Milton H. Erickson, M.D. New
York: W. W. Norton, 1973.
13. Erickson, M., in Haley, J. (ed.), Advanced Techniques of Hypnosis and Therapy. New York:
Grune and Stratton, 1967, pp. 395-397.
14. Ibid., p. 396.
15. Ibid., pp. 393-395-
16. Grinder, J., et al., Patterns of the Hypnotic Techniques of Milton H. Erickson, M.D. Cupertino,
Calif.: Meta Publications, 1977.
17. Haley and Hoffman, Techniques of Family Therapy, Ch. 3.
18. Weakland, J., and D. D. Jackson, "Patient and Therapist Observations on the Circum-
stances of a Schizophrenic Episode," A.M. A. Archives of Neurology and Psychiatry 79 (1958),
554-574-
19. Bateson, G., D. D. Jackson, J. Haley, and J. Weakland, "Toward a Theory of Schizo-
phrenia," Behavioral Science 1 (1956), 251-264.

359
 NOTES
20. Watzlawick, P., J. Beavin, and D. D. Jackson, Pragmatics of Human Communication. New
York: W. W. Norton, 1967, pp. 243-244.
21. Ibid., pp. 74-75.
and I. Yalom, "Conjoint Family Therapy as an Aid to Intensive
22. Jackson, D. D.,
Psychotherapy," in Jackson, D. D. (ed.), Therapy, Communication and Change. Palo Alto, Calif.:
Science and Behavior Books, 1968.
23. Haley and Hoffman, Techniques of Family Therapy, p. 180.
24. Ibid., p. 174.
25. Ibid., p. 177.

Chapter 13 Historically Oriented Approaches to Family Therapy

1. Bowen, M., Family Therapy in Clinical Practice. New York: Jason Aronson, 1978.
2. Anonymous, "Towards the Differentiation of a Self in One's Own Family," in Framo,
J. New York: Springer Publishing Co., 1972.
(ed.), Family Interaction.
Guerin, P., and K. Guerin, "Theoretical Aspects and Clinical Relevance or the Multi-
3.
Generational Model of Family Therapy," in Guerin, P. (ed.), Family Therapy: Theory and Practice.
New York: Gardner Press, 1976.
4. Carter, E., and M. Orfanides, "Family Therapy With One Person and the Therapist's
Own Family," in Guerin (ed.), Family Therapy, p. 207.
5. Guerin (ed.), Family Therapy, p. 104.
6. Paul, N., "The Role of Mourning and Empathy in Conjoint Marital Therapy," in Zuk,
G., and I. Boszormenyi-Nagy (eds.), Family Therapy and Disturbed Families. Palo Alto, Calif.:
Science and Behavior Books, 1969.
7. Framo, J., "Family of Origin as Therapeutic Resource for Adults in Marital and Family
Therapy," Family Process 15 (1976), 193-210.
8. Papp, P., "Family Sculpting in Preventive Work with 'Well' Families," Family Process
12 (1973), 197-212.
9. Boszormenyi-Nagy, I., and G. Sparks, Invisible Loyalties. New York: Harper and Row,
1973-
10. Ibid., p. 6.
11. Ibid., pp. 47-48.

Chapter 14 Ecological, Structural and Strategic Approaches

Minuchin, S., et al., Families of the Slums. New York: Basic Books, 1968.
1.

Rabkin, R., Inner and Outer Space. New York: W. W. Norton, 1970.
2.

3. Auerswald, E. H., "Interdisciplinary versus Ecological Approach," Family Process 7

(1968), 205-215.
4. Hoffman, L., and L. Long, "A Systems Dilemma," Family Process 8 (1969), 211-234;
Hetrick, E., and L. Hoffman, "The Broome Street Network," in Sanders, D. S., J. Fischer, and
O. Kurken (eds.), Fundamentals of Social Work Practice. North Scituate, Mass.: Duxbury Press,
1981.

36O
 NOTES
5. Langsley, D., and D. Kaplan, Treating Families in Crisis. New York: Grune and Stratton,
1968.
6. Haley, J., and L. Hoffman, Techniques of Family Therapy. New York: Basic Books, 1967,
Ch. 5.
Speck, R., and C. Attneave, Family Networks. New York: Vintage Books, 1974.
7.
Aponte, H. "The Family-School Interview: An Eco-Structural Approach," Family Process
8.

15 (1976), 303-311. Aponte, H. "Under-organization in the Poor Family," in Guerin, P. (ed.),

Family Therapy: Theory and Practice. New York: Gardner Press, 1976.
9. Minuchin, S., Families and Family Therapy. Cambridge, Mass.: Harvard University Press,
1974-
Minuchin, S., Psychosomatic Families. Cambridge, Mass.: Harvard University Press, 1978.
10.
Ch. 5.
11. Ibid.,
12. Aponte, H., and L. Hoffman, "The Open Door: A Structural Approach to a Family
with an Anorectic Child," Family Process 12 (1973), 1-44.
13. Malcolm, Janet, "A Reporter at Large: The One- Way Mirror," The New Yorker (May
1978,) 40.
14. Weakland, J., R. Risch, P. Watzlawick, and A. Bodin, "Brief Therapy: Focused Prob-
lem Resolution," Family Process 13 (1974), 141-168. Watzlawick, P., J. Weakland, and R. Fisch,
Change: The Principles of Problem Formation and Problem Resolution. New York: W. W. Norton, 1974.
15. Haley, J., Uncommon Therapy: The Psychiatric Techniques of Milton H. Erickson, M.D. New
York: W. W. Norton, 1973.
16. Haley, J., Problem-Solving Therapy. San Francisco, Calif.: Jossey-Bass, 1977.
17. Haley, J., Leaving Home. New York: McGraw-Hill, 1980.

Chapter is The Systemic Model

1. Selvini Palazzoli, M., Self-Starvation. New York: Jason Aronson, 1978, p. 19.
2. Selvini Palazzoli, M., et al., Paradox and Counterparadox. New York: Jason Aronson, 1978,
p. 8.
3. Selvini Palazzoli, M., "Why a Long Interval Between Sessions," in Andolfi, M., and
I. Zwerling (eds.), Dimensions of Family Therapy. New York: Guilford Press, 1980.
4. Selvini Palazzoli et
al., Paradox and Counterparadox, p. 55.

5. Selvini Palazzoli, Self-Starvation, p. 208.

6. Selvini Palazzoli et al., Paradox and Counterparadox, p. 56.

7. Madanes, C, "Protection, Paradox and Pretending," Family Process 19 (1980), 73-85.

8. Selvini Palazzoli, M., et al., "Hypothesizing Circularity — —
Neutrality," Family Process
19 (1980), 3-12.
9. Selvini Palazzoli, Self-Starvation, p. 231.
10. Selvini Palazzoli, M., et al., "The Problem of the Referring Person," Journal of Marital
and Family Therapy, 6 (1980), 3-9.
11. Ibid., p. 4.
12. Selvini Palazzoli et al., "Hypothesizing —Circularity— Neutrality," pp. 3-12.
13. Ibid.

361
 NOTES

Chapter 16 Theories About Therapeutic Binds

1. Watzlawick, P., D. Jackson, and J. Beavin, Pragmatics of Human Communication. New York:
W. W. Norton, 1967, Ch. 7.
2. Stanton, M. D., "Strategic Approaches
Family Therapy," in Gurman, A., and D.
to
Kniskern (eds.), Handbook New
York: Brunner/Mazel, 1981.
of Family Therapy.
3 Bateson, G., Mind and Nature. New York: E. P. Dutton, 1979, p. 62.
4. Haley, J., Strategies of Psychotherapy. New York: Grune and Stratton, 1963, p. 159.
5. Hoffman, L., " 'Enmeshment' and the Too Richly Cross-Joined System," Family Process

14 (1975), 457-468.
6. Selvini Palazzoli, M. "Why a Long Interval Between Sessions," in Andolfi, M., and I.

Zwerling (eds.), Dimensions of Family Therapy. New York: Brunner/Mazel, 1980.

7. Ashby, Design for a Brain, p. 207.
8. Dell, P., "Beyond Homeostasis," Family Process (forthcoming).

9. Ibid.

Chapter 17 Issues on the Cutting Edge

1. Papp, P., "The Family That Had

All the Answers," in Papp, P., (ed.), Family Therapy:
Full Length Case Studies. New
York: Gardner Press, 1977, Ch. 9.
2. Miller, S. J., "The Social Base of Sales Behavior," Social Problems 12 (1964), 15-24.
3. Papp, "The Family That Had All the Answers," p. 152.

Epilogue Toward a New Epistemology

1. Bateson, G., "Afterword," in Brockman, J., (ed.) 7 About Bateson. New York: E. P. Dut-
ton, 1977, p. 242.
2. Salomon, M., "Entretien avec Prigogine," Prospective et Sante 13 (June 1980), 41-58.

3. Prigogine,"L'ordre a partir du chaos," Prospective et Sante 13 (June 1980), 29-39.

I.,

4. Elkaim, M., "Debat entre Ilya Prigogine, ses collaborateurs, Felix Guattari et Mony
Elkaim," Cahiers Critiques de Therapie Familiale et de Pratiques de Reseaux 3 (Paris: Editions Gamma,
1980) 6-17.
Elkaim, M., "Non-Equilibrium, Chance and Change in Family Therapy," to be pub-
5.
lished in "Models of Therapeutic Intervention with Families: A Representative World View,"
special issue of the Journal of Marital and Family Therapy (summer 1981).
6. Prigogine, I., "Structure, Dissipation and Life," in Theoretical Physics and Biology. Amster-
dam, Holland: North-Holland Publishing Co., 1969, pp. 23-32.
7. Bateson, G., "The Birth of a Matrix or Double Bind and Epistemology," in Berger, M.,
(ed.), Beyond theDouble Bind. New York: Brunner Mazel, 1977, p. 53.
8. Dell, P., and H. Goolishian, "Order Through Fluctuation: An Evolutionary Paradigm
for Human Systems." Presented at the Annual Scientific Meeting of the A. K. Rice Institute,
Houston, Texas, 1979.
9. Dell and Goolishian, "Order Through Fluctuation."

362
NOTES
10. Keeney, B., "Ecosystemic Epistemology: An Alternative Paradigm for Diagnosis,"
Family Process 18 (1979)/ H7~
12 9-
11. Dell, P., "Beyond Homeostasis," Family Process (forthcoming).
12. Dell and Goolishian, "Order Through Fluctuation."
13 Brodey, W., "Some Family Operations and Schizophrenia," Archives of General Psychiatry
1 (1959)' 379-402.
14. Dell, P., "Beyond Homeostasis."
15. Ibid.
16. Paul Klee, Notebooks, vol. 2, trans. Ralph Manheim, ed. Jiirg Spiller. New York: George
Wittenborn, 1973, PP 7^-79-

363
Index

Abdication by single parents, 179 Anarchistic family, 98, 99

Abreaction of repressed material, 242, 243, Andolfi, Maurizio, 344
249/ 250 Anorexia nervosa: conflict management and,
Absurd, therapy of, 228-31 150, 152-54; parental support for, 311;
Accession,crisis of, 161 structural approach for, 264-70; sympto-
Achievement, Rabkin on, 168 matic cycle in, 203-9; systemic approach
Ackerman, Nathan, xiii, xiv, 17, 220, 223-28, to, 284, 313-17
261 Anxiety: in couple cycle, 190; in multiprob-
Ackerman Institute for Family Therapy, lem family, 84
298^, 307, 338 Aponte, Harry, 83, 85, 262, 272
Addict families, 57; paradoxical techniques Apple, D., 133
with, 306 "Applied Behavioral Sciences Program,"
Adolescence: academic problem during, Gouverneur Health Services, 257
326-27; Ackerman's approach to treat- "Art of Being Schizophrenic, The" (Haley),
ment in, 225; and anorexia nervosa, 266; 118
interpretive approach in, 210-12; para- "Ascribing noble intention" technique,
doxical injunctions in, 168, 171; parental 306
fear of homosexuality in, 235-36; psycho- Ashby, W. Ross, 136, 166, 318; on bi-modal
sis in, 282; scapegoating in, 101; schizo- feedback mechanisms, 159, 197; on null-
phrenia in, 215-17; second order change function, 319, 320; step-mechanisms of,
in, 198; sexual acting out in, 272; simple 162-63, 165; on too richly cross-joined
bind becoming symptom in, 172-75 systems, 74-78, 100, 286; on two-level
Afghanistan, Pathan clans of, 44 model for change, 47-48; on "variety"
African societies, emphasized relationship and "constraint," 51, 142
in, 122-24 Asian societies, emphasized relationship in,
Alanen, Yrjo, 345 122
Albee, Edward, 70 Asthma: breaking symptomatic cycle for,
Alcoholic couples, 188, 189; therapeutic 202-3; conflict management and, 150,
double bind in treatment of, 237 152, 153
"Alcoholic" family type, 66 Attneave, Carolyn, 261
Alliances: coalitions distinguished from, Auerswald, E. H., xiii, 17; ecological model
108; in family of schizophrenic,H3 of, 256-58
Ambivalence, 114; versus conflict of levels, Authoritarian families, 98-99
115-17 Avoidance-avoidance conflict, 107, 113-14

365
 INDEX
"Bad" label, 138 Boszormenyi-Nagy, Ivan, see Nagy, Ivan
Backlash of therapeutic encounter, 230 Bott, Elizabeth, 134, 142
Balancein Small Groups (Taylor), 137 Boulding, Kenneth, 41-43, 107, 113-14
Balance theory, see Structural balance theory Bowen, Murray, 17, 28-32, 72, 89, 146, 176,
Balancing responses, 309-12 251; on degree of connectedness, 135;
Bales, Robert, 61, 141 "detriangling" technique of, 78-79; his-
Balinese culture, 43 torical approach of, 243-49, 337
Bander, Richard, 234 Brain, cybernetic model for, 75
Barragan, Mariano, 281 Brief family crisis therapy, 258-61
Barth, Fredrik, 44 "Brief Therapy: Focused Problem Resolu-
Bateson, Gregory, xii, 60, 70, 75, 104, 124, tion" (Weakland, Fisch, Watzlawick, and
154, 168, 280, 306, 320, 334, 342; on am- Bodin), 271
bivalence, 115-16; Ashby and, 48; bicam- Brodey, W., 346-47
eral format of, 4; Boulding and, 42; circu- Bronx State Hospital, 260-61
lar epistemology of, 7-$, 342-44; "Broome Street Network" (Hetrick and
communication research of, 5, 11, 17-19, Hoffman), 258
26, 33, 67, 107, 256; couple observations Brown, Howard, 257
by, 184-85; cybernetic paradigm of, 36, Buckley, Walter, xiii, 51, 52
45-47, 202; death of, 350; Dell influenced Buffalo Creek flood, Erikson's study of,
by, S7, 321; double bind hypothesis and, 191-92
xiii, 105, 235, 239, 287, 313; on epigenesis Byng-Hall, John, 345
and evolution, 157; game theory used by,
24-25; "infinite dance of shifting coali-
tions" of, 25, 74; on information, 300; on
inverse progressive changes, 53; on
"learning to learn," 169; on levels of sys-
tems, 55; Logical Paradox analogy of, ill, Caille, Philippe,
345
112; model for mapping phenomena at in- Caplow, Theodore, 134, 136, 148; on ambiv-
creasing levels of complexity of, 156; on alence, 115; balance theory and, 124-25,
nature of stability, 177, 317-18; on pat- 131; coalitions research of, 107, 109-12,
terning, 37-40; on phenomenon of binoc- 120-21, 137, 140-43; on emphasized re-
ular vision, 316; prescribing the symptom lationship, 122;on Freud, 121
by, 305; on processes of change, 159; on Carter, Elizabeth, 246, 247, 253
recursive epistemology, 339, 348; schis- Cartwright, Dorwin, 126, 127, 129
mogenesis concept of, 12, 40-41, 43-45, Catharsis, psychoanalytic concept of, 243
69, 73, 138, 197; Selvini and Milan group Cathode-ray oscilloscope, 77
influenced by, 14, 285, 287, 337, 338, 345 Cecchin, Gianfranco, 284, 291, 304
Bateson, Mary Catherine, 7-S Center for Family Learning (New Rochelle,
Beavers, R., 97-98, 101 N.Y.), 213
Beavin, J., 17, 23, 236 Centrifugal family, So, 98, 102
Beels, Chris, 261 Centripetal family, So, 98, 102, 104
Behavior disorders in centrifugal families, 98 "Chain of displaced retributions," 251, 253
Behavior therapy, 220 Change: balance scale analogy for, 309-13;
"Behavioral theory of Drug Taking, A" constancies and, 318-22; discontinuous,
(Wilkins), 57 102, 156-75; negative consequences of,
Bell, John, 17 238-41; symptoms as harbingers of, 347;
Bell,Norman, 37, 59 two-level model for, 47-49; vulnerability
Bergman, Joel, 307^ 338 of living forms to, 341
Berman, E., 188 Change: Principles of Problem Formation and Problem
Bertalanffy, Ludwig von, 5 Resolution (Watzlawick, Weakland, and
"Beyond Homeostasis" (Dell), 321, 347, 348 Fisch), 198, 199, 271, 277
Bicameral model, 4-5 Chaotic families, 97-98
Bimodal feedback mechanisms, 48, 159, 197 Chappie, Eliot D., 192
Bipolar families, 13 Character disorders, research on, S7
Birdwhistle, Ray, 261 "Choreography" technique, 251
Blakar, Rolv, 345 Circadian rhythms, interactional rhythms
Boas, Franz, 38 and, 192-93
Book of Changes (I ChingJ, 295 Circular epistemology, 346
Boscolo, Luigi, 284, 297-98, 304 Circular questioning, 300-302, 322

366
INDEX
Circularity, concept of, 5-9, 287 Coping patterns, 160-61
Clarkin, John, 307^ Coser, L., 132
Classification, confusion of levels of, 313 Cotherapy, Whitaker's use of, 230, 231
Closed family, 98-99 "Counterautonomous superego," 252
Closed pseudointegrated family, 93 Counterparadox, 287-89; levels of logical
Clustering theory, 132-34 type and, 313-17; see also Therapeutic
Coalitions, 13, 106-7; in balance theory, double bind
128, 129, 136-37; Caplow on, 100-11; in Countervailing coalitions, 120
family sculpting, 250; fixed sequences of, Couple cycles, 184-90
138; Haley on, 111-13; shifting, 24-26, Couples, see Complementary couples; Dyads;
317; stability and, 318; study of, in rela- Symmetrical couples
tion to symptomatic behavior, 293; see also Couples therapy: systemic approach in, 336;
Cross-generation coalitions therapeutic double binds in, 237
Cocoanut Grove fire, Lindemann's study of, "Couples Therapy: Prescription for a Sys-
161 tems Experience" (Sluzki), 190
Cognitive dissonance, 126 Creative leaps, 168-70
Coherence, 349 Crises, life stage, 160-62
Cohesion, promoted by deviance, 59 Crisis intervention, 258-61
"Collaborative Geniuses" marriage type, 185 "Critical state" in step-function changes,
Colorado Psychiatric Hospital, 258-60 163, 164
Communication: ambivalence in, 115; "Critique of the Clinical Use of the Double
avoidance of definition in, 117-19; Bate- Bind, A" (Rabkin), 167
son's research on, 17-19; in Bowen's ap- Cross-generation coalition, 108, 139, 177;
proach, 246; control theory of, 288; dis- and avoidance of defining relationships,
crepancies in, 220; excessive, 77-78; 118; balance theory and, 136; conflict
hidden patterns of, 261; Satir on, 221-23; in levels in, 116; in extended family,
shift of emphasis from family structure to, 119-22
105; symptom type and, 67-69; systemic Cross-joining: excessive, see Too richly cross-
view of, 288 joined systems; inadequate, 70-81
Community psychotherapy, 257-58 Cross-pressure hypothesis, 113, 115
Communication, Family and Marriage (Jackson), Cross-sectional process model, 97-98
Cults, children from permissive homes in,
"Competitive" marriage pattern, 187 101
Complementary couples, 60, 60-70; change Culture and Biological Man (Chappie), 192
to symmetrical couples from, 308-9; Cybernetic mechanisms, 45-46; in brain,
"homeostatic seesaw" of, 180-84 163; evolutionary, 157, 158; Maruyama
Complementary differentiation, 41-45 and on, 40-54; naven ceremony as, 46-47; in
dominance-submission cycles, 43 symptomatic cycle, 197, 202
Compression, theory of, 306 Cybernetic model, 36, 330-40; for adap-
Conflict: of levels, 115-17; among newly- tive system, 75; criticism of, 341
weds, 194-95; triads and management of,
140-55
Conflict and Defense (Boulding), 41, 113
Conjoint Family Therapy (Satir), 220
"Conjoint Family Therapy as an Aid to In- Daniels, Arlene, 59
tensive Therapy" (Jackson and Yalom), 23 Davis, James, 113, 115; balance theory and,
Consensual Homeostasis, 94, 96 125; on clustering, 131, 132, 134
Consensual Morphostasis (Mc), 92 Death phobia, 302
Consensus-sensitive family, 88, 89, 93-94, "Degradation ceremonies," 65
104, 220; contest stance with, 329; ho- Delinquents, families of, 71-72
meostasis in, 95 on co-
Dell, Paul, 5, 12, 26n, 100, 101, 346;
Conservative coaltion pattern, 110 herence, 349; on concept of fit, 347; on
Constancies: interactive, 318-22; wall of, 77 epistemology, 343-44; on evolutionary
Constraint, 51, 142 feedback, 156, 202, 341; on resistance,
"Context replication," 262 348; on self-healing tautologies, 321; on
Contextual issues of treatment, 290-300 transactional versus etiological model,
Control theory of communication, 288; 86-87
schizophrenic transaction and, 26-28 Delusion, allegorical nature of, 224
"Cooperative" marriage pattern, 187 Democratic family, 98, 99

367
 INDEX
Dentler, R. A., 59 Drug abuse, couples with problems related
Depression: in mother of multiproblem fam- to, 188
ily, 84; of single parents, 179; of submis- Dualism, 342-44
sive spouse, 187 Duhl, Bunny, 250
Design for a Brain (Ashby), 74, 75, 163 Duhl, Fred, 250
"Detouring-Attacking" triad, 150-51 Durkheim, Emile, 55, 59
"Detouring-Supportive" triad, 150-51 Dyads, 24; in Bowenian therapy, 246; critical
Developmental models, 13 state in, 164; cycles of, 184-90; in en-
Deviance: angle of, from normative model, meshed families, 73; governing, in differ-
263; evolutionist theory of, 51; meaning ent cultures, 122; interactional rhythms
of, for social systems, 59-60; role of, 49; of, 193; typology for, 69
typecasting of, 56-59 Dysfunctional family system, 220
Deviation-amplifying process, 9, 53-54, 56;
levels and, 63; role of society in, 65; in
"special case," 148; safe, substitution of,
62
"Devil's Pact," 279 Ecological approach, 219, 242, 255-62
de Young, Carol, 259 Economy of technique, 213-15
Diabetic children, conflict management and, Eco-structural approach, 262
150, 152, 153 Ecosystemic epistemology, 343
Dialectic of polarities, 295 Ego, 244
37-
Differentiation: in biological structures, Ego mass, undifferentiated, 28-32
39; Bowen's concept of, 29, 31-32; lack "Ego strengths," 160
of, in enmeshed family, 72; through schis- Eliot, Thomas, 161
mogenesis, 40; of self, 76, 243-49 Elkaim, Mony, xiii, 341, 348
Differentness, acceptance of, 220 Emotional cutoff, differentiation distin-
Disadvantaged families: organization of, 71, guished from, 32
80-85; see also Poverty "Emotionally Disturbed Child as a Family
Discontinuous change, 102, 156-75; evolu- Scapegoat, The" (Vogel and Bell), 57
tionary feedback and, 156-58; expectable Emotionally disturbed children: in cross-
life stage crises and, 160-63; hierarchical generation coalitions, 108; mediation
growth and, 158-60; paradoxical injunc- function of deviance and, 59; scapegoat-
tions and, 166-75; step-mechanisms in, ing of, 57; systemic approach for, 284; un-
163-66 even marital arrangement and, 180, 181
Disengaged families, 71, 72, 79-86, 90, 102, "Emphasized relationship" as cultural char-
136; redundancies in, 81-85 acteristic, 122
Disintegrated family, 93, 95 "Encouraging the resistance," 232
Dismemberment, 161
crisis of, Enmeshed families, 31, 71-75, 79, So, 89,
Disqualified message, 26-27 104, 176; communication patterns in, 220;
Dissociation, 149 cross-connections in, 77 78; degree of
,

"Dr. Homeostat," 299 adaptiveness of, 102; "resistance" of, 270

Dominance-submission cycle, 42-43 Entropy, 104, 340
"Dominant-Submissive" marriage pattern, Enuresis, hypnosis in treatment of, 233-34
187, 188 Environment-sensitive families, 88, 90-91,
Donne, John, 95 93-94, 104; homeostasis in, 96
Double bind, 15, 176; ambivalence distin- Epigenesis, 157
guished from, 115; dyadic formulation of, Episodic therapy, 332
24; Haley's interpretation of, 26; origins Epistemology: importance of, 342-44; new,
of hypothesis of, 10-23; and shifting coa- need for, 345-50
litions, 24-26; therapeutic, see Therapeutic Equifinality, principle of, 72
double bind; in three-party interaction, Equilibrium model, 339-40
105-7 Equilibrium point, Boulding's concept of, 42
"Double Bind Hypothesis of Schizophrenia Erickson, Milton, xiv, 220, 232-35, 279, 337,
and Three-Party Interaction" ( Weakland), 350; on "contest" between therapist and
24, 105-6 client, 328; on termination of therapy, 331
Double Bind: The Foundation of the Communicational Erikson, Erik, 160, 161
Approach to the Family (Sluzki and Ransom), Erikson, Kai T., 59, 191-92
23 Error-activated feedback cycles, 45
Drama, Milan Associates' use of, 303 Esterson, A., 36

368
INDEX
Ethnicity, 261; normativemodel and, 264 escalation of, 19; in symptomatic cycle,
Etiological model,86-67 203
"Evaluative Review of Family Interaction Feedback loops: interplay of, 52; and levels
and Research, An" (Riskin and Faunce), of system, 54-56; timing in, 53-54
28 Female-headed households, coalitions in,
Evan, Richard, 307« 122
Evolution of problem, 325-28 Ferber, Andrew, 261
Evolutionary feedback, 156-5 8, 202, First order change, 47-49; symptomatic

340-41 cycle and, 197-202

Evolutionary paradigm, 5, 9, 101-2, 339-42; "First Session of a Systemic Family Therapy"
shift to, 346 (Selvini), 313
Expectable life stage crises, 160-62 Fisch, Richard, 61, 198, 203, 213-14, 241;
Expressive leaders, 141, 142 strategic approach of, 271, 280, 289
Extended family, cross-generation coalitions Fish, Vince, 346
in, 110-22 Fit, concept of, 347

Externally open-integrated family, 93 Fivaz, Roland, 103-4

Extrusion, therapist as substitute for, 330 Five-person game, 25
Fleck, Stephen, 17
Flipflop Phenomenon, 187-88
Flomenhaft, Kalman, 135, 259
Flow hierarchy, 158
Fairbairn, W. R. D., 250 Fogarty, Thomas, 261
Fairly integrated family, 93 Forced Homeostasis, 94, 96
Families and Family Therapy (Minuchin), 72, Forced Morphostasis (Mf), 92
262, 264 Formal logic, paradox of, 115
Families of the Slums (Minuchin), 71, 81-82, "Formality systems," 143
256 Framo, James, 17, 250
"Family of Angels, A" (Satir), 221 Franklin, Paul, 224
Family balance scale, 310-12 Freilich, Morris, 140-43, 148
Family Crisis Unit (Colorado), 250-60 Freud, Sigmund, 121, 123, 219, 249
Family Crucible, The (Napier), 231 Fromm-Reichmann, Frieda, xi, 236
"Family Dynamics and the Reversibility of Frye, W., 23
Delusional Formation" (Ackerman and Fulweiler, Charles, 210, 212, 214
Franklin), 224 "Functions of Deviance in Groups, The"
Family Health Unit (New York), 258 (Dentler and Erikson), 59
Family homeostasis: coining of term, 18; see Fusion, 176, 244; oscillating, 306
also Homeostasis
Family ledger, 251-55
Family life cycle, 14; expectable crises in,
160-62
Family paradigms, 86-104; grid model for, Game theory, 24-25; therapeutic double
91-96; and metaphysics of pattern, 86- binds and, 306
91; process models for, 96-100; spiral "Game without end," 172, 198, 201, 207,
platters as, 100-104 288; contest stance and, 329
Family Study Section (Bronx State Hospital), Genogram, 244, 246
261 Gestalt therapy, 220
"Family Systems: Morphostasis and Mor- Goding, Geoffrey, 345
phogenesis" (Speer), 9 Goffman, Erving, 65
"Family That Had All the Answers, The" Goolishian, Harold, 156
(Papp), 333 Gouverneur Health Services (New York),
Family Therapy in Clinical Practice (Bowen), 29 257
Family Therapy Institute (Washington), 281 Graham, Billy, 212
"Family Unit Therapy and the Science and Gramsci, Antonio, 16^-66
Typology of Family Systems" (Wer- Greeks, ancient, 56
theim), 91 Grid models, 13, 91-96
Family Worlds (Hess and Handel), 90 Grief reaction, intensity of, 160-61
Faunce, E., 28 Grinder, John, 234
Feedback: error-activated, 45; evolutionary, Growth, hierarchical, 158
156-5^, 202, 340-41; levels of, 76; rapid "Growth therapies," 243

369
 INDEX
"Gruesome Twosome" marriage type, 185, Homeostatic model, 5, 9, 72; Ashby on, 75;
186 change in, 56; evolutionary feedback and,
Guerin, K., 247, 248 1 57~58; Hardin on, 51; irresistible runs in,
Guerin, Philip, 246-49, 261 60-62; pathological balance and, 62-63;
plateaus in, 61; Wynne's, 33
Horney, Karen, xi
Hsu, Francis, 122, 123
Hubris, 56
Haley, Jay, xiii, xiv, 17, 18, 29, 148, 162, 219, Human Education Movement, 220
235, 260; on ambivalence, 115; on avoid- Human potential movement, 223
ance of defining relationships, 117-18; Hypnotism: Erickson's use of, 232-34;
balance theory and, 125; on coalitions, Haley's interest in, 279, 280
106-7, 120-21, 123, 134, 138-39; on Hypothesis, systemic, 293-95, 324
cross-generation joinings, 79; double bind "Hypothesis-Circularity-Neutrality" (Milan
research of, 21, 28-29; on double conse- Associates), 293-95, 300, 302
quence of symptoms, 56; Erickson and,
232, 234; on homeostatic monitoring of
behaviors, 72; interpretive approach of,
210; Milan Associates and, 287, 288;
"perverse triangle" of, 108-13, 140, 143; Iatmul culture, 38-40, 44-49, 124
problem-solving approach of, 279-83; I-Ching, 295
redefinition of schizophrenia by, 60, 149; Id, 244
therapeutic approach of, 98; on therapeu- "Improper coalition," 107, 144
tic double binds, 306; on three-person In and Out of Psychosis (film), 224
system, 318; on treatment of hospitalized Inadmissible triangle, 129, 150
adolescent schizophrenics, 215-16; on Incest tabu, 121-22
types of parental roles, 212, 214; on un- Incongruous hierarchy, prescription of, 292
stable dyads, 33 Individual therapy, systemic approach in,
Handel, Gerald, 00 336
Harary, Frank, 126, 127, 129 Individuated self, 242
Hardin, Garrett, 51, 54 Individuation: achievement of, 243; fear of,
Headaches, therapeutic double bind in treat- 36; in normative model, 263
ment of, 237 Induced Morphostasis (IM), 93
"Heavenly Twins" marriage type, 185 "Infinite dance of shifting coalitions," 25,
Heider, Fritz, 126-27, 129, 142 74, 317
Hess, Robert, 90 Injunctions, paradoxical, see Paradoxical
Hetrick, Emery, 258 techniques
Hierarchical growth, 158-60 Inner and Outer Space (Rabkin), 257
High Status Authority (HSA), 141-43 Insight, psychoanalytic concept of, 242, 243
High Status Friend (HSF), 141-43 Instabilities, 340; in evolving systems, 348
Hill, Reuben, 161 Institute for Family and Human Systems
Hindu culture, emphasized relationship in, Study (Brussels), 341
122, 123 Institute ofFamily Studies (Rome), 344
Hirsch, Siegi, 345 Instrumental leaders, 141, 142
Historical approach, 219, 242-55; differen- Integrated family, 93
tiated self in, 243-49; and theory of re- Interaction typologies, 69-71
pression, 249-5; Interactive sequence constancies, 318-19
Hoffman, Lynn, 60, 210, 235, 257-5$, 260 "Interdisciplinary versus Ecological Ap-
Holmes, T. H., 161-62 proach" (Auerswald), 257
Home vists: in crisis intervention, 259; in Interfering Grandmother, 179
problem-solving approach, 279 Interpersonal Behavior Game-Test (Train
Homeostasis, 339; coherence and, 349; Dell Game), 186, 195
on, 343; in family of anorectic, 208; in Interpersonal-distance sensitive families, 88,
family sculpting, 250; in grid model, 94- go, 93-94; homeostasis in, 95

96; in hospitalization of schizophrenic ad- Interpretation in Bowenian therapy, 247

olescents, 215; in process models, 98; resis- Interpretive approach in family therapy,
tance and, 348; symptomatic behavior 210-13
and, 180-84; see also Homeostatic model Intimacy: access to, 191-96, denial of,

Homeostat, 76 267-68

370
INDEX
Intrapsychic theories of development, 160 Loops, see Feedback loops
"Introjects," 250 Low Status Subordinate (LSS), 141-43
Inverse progressive changes, 53 Lunar cycles, 193
(Nagy), 251
Invisible Loyalties Lundquist, Betty, 3Q7«
Irresistible runs, 60-62

Madanes, Cloe, 219, 281, 292

Jackson, Don, xiii, xiv, 53, 220, 284;
5, 17, 23, Malcolm, Janet, 270
on couple interaction, 66, 185-86; predic- Mapping, Minuchin's method of, 264-65
tive methodology of, 68-69; "rescue oper- Marital maladjustment, self-reinforcing re-
ation" observed by, 5g, 62; "runaway" lationship sequence in, 41
observed by, 19; on "symmetrical" cou- Marital Puzzle, A (Paul and Paul), 190
ples, 60; therapeutic double bind of, "Marital schism," 185
235-41 "Marital Schism and Marital Skew" (Lidz),
Joking behavior, 121 182
Josiah Macy, Jr., Conferences, 45 Marital skew, 106
Marital subsystem in normative model,
262-63
Marriage Contracts and Couple Therapy (Sager),
190
Kantor, David, 98-99, 250 Maruyama, Magoroh, 49-54, 57
Kaplan, David, 135, 258 Mathematical Theory of War (Richardson),
Kaufman, Luc, 345 xii

Keeney, Bradford, 343 Maturana, Humberto, 5

Kinesics, 261 Meaning: disqualification of, 26-27; levels
King, Charles, 256 of, 17
Klee, Paul, 349 Mediation function of deviance, 59
Koestler, Arthur, 168 Medical model, 6
Mendelsohn, Marilyn, 261
Menninger Clinic, 29
Menstrual cycle, 193
Mental Hospital, The (Stanton and Schwartz),
Laing, R. D., 36 106, 144
Langsley, Donald, 258 Mental Research Institute (Palo Alto), xiv, 5,

"Leap theory," 167 11, 23, 61, 68, 70, 220, 235, 238, 241; see

Learning: levels of, 17; to learn, 169 alsoPalo Alto group

Leaving Home (Haley), 98, 215, 282 Merton, Robert, 55
Lehr, W., 98-99 Messinger, S. L., 190
LeMasters, E. E., 161 Metameric differentiation, 37; cultural, 39
Lemmings, destructive positive feedback ex- Metaphysics of pattern, 86-91
emplified by, 54 Milan Associates, 14, 285-87, 305, 306, 319,
"Letters of Mothers of Schizophrenics" 321-22, 333, 337, 338; circular question-
(Weakland and Frye), 23 ing format of, 300-302; concept of "neu-
Levels of logical type, 17, 18, 313-17 trality" of, 302-4, 348; counterparadox
LeVine, Robert, 123 of, 287-89, 313-17; evolutionary feed-
Lidz, Theodore, 17, 67-68, 106, 182, 185 back concept and, 341; influence of, 344,
Life stage crises, expectable, 160-62 345; referring context of, 299-300; sys-
Lindemann, Erich, 160-61 temic hypothesis of, 293-95; technique
Linear thinking; 342; circularity versus, 5-9, for dealing with resistance, 324
288, 346; paradoxical techniques and, 306 Milan Institute for Family Studies, 284
Live observation, 16-17 Miller, S. J., 333
(Thomas), 80
Lives of a Cell Mind and Nature (Bateson), 4, 7, 157, 317
Logical Paradox, 111 Minnesota, University of, 161
Logical type: levels of, 17, 18, 313-17; Rus- Minuchin, Salvador, xiii, xiv, 17; Ackerman
sellian Theory of, 27, 28 and, 225, 228; on conflict-detouring
Long, Lorence, 257-58 triads, 150-55, 177, 178; crisis induced
"Long, brief therapy," 287 by, 53; disengagement concept of, 71, 72,

371
 INDEX
Minuchin, Salvador (continued) "Neutrality," Milan Associates concept of,
79, 82, 83, 90, 136; ecological model of, 302-4, 321, 348
1 5°-57> 261-62; enmeshment concept of, New Guinea, Bateson's studies of, 38-40
31, 71-75, 78, 79/ 89, 176; Haley and, New York Department of Social Services,
280; structural approach of, 262-71, 344; 258
structural typology of, 66; symptomatic New Yorker magazine, 168, 270
cycle derailed by, 203-1 1; triangle con- Newcomb, T. M., 127
cept of, 29 Newlyweds, study of conflict among,
Mirages of Marriage (Jackson), 69, 185 194-95
Mirror-image disagreement, 148-50, 202; Nonequilibrium dynamics, 341
emergence of, 215; interpretive approach Normative model, Minuchin's, 262-64
to, 211; between parents, 154; of parents Nuclear family: boundaries between family
of anorectic, 206, 265; about symptomatic of origin and, 73; coalition choices in, 110;
behavior, 182 community networks and, 261; fusion
Mobile Crisis Unit (New York), 258 with, 306; intimacy function of, 191;
"Modern Little Hans, A" (videotape), 281 Minuchin's work with, 262; in normative
Montalvo, Braulio, 71, 82, 83, 256, 257, 272, model, 263; scapegoating in, 134
281 Null function, 163, 319, 320
Morawetz, Anita, 307^ "Null-function" movement, 163
Morgenstern, O., 25
Morphogenesis, 50-52, ^6; in grid model,
91-92, 94; therapy and, 167
Morphostasis, 50-52, 56; in grid model, 91,
92, 94; therapy and, 167 Object relations theory, 250
"Mother-in-law avoidance," 121 Obligation ties, 251-55
Mourning, unresolved, 249 Obsessions in couple cycles, 190
Multigenerational paradoxical interventions, Oedipal conflict, 121
251 "On the Anguish and Creative Passions of
Multigenerational transmission of emotional Not Escaping the Double Bind" (Wynne),
illness, 29; Bowen's theory of, 243 170
Multiple group affiliations, 132 One-way screen, 3; bicameral model and,
Multiproblem families, 66, 81-85; coordina- 4-5; Milan Associates use of, 285, 303,
258
tion of services for, 304; team approach and, 334, 335
Mutation, ^o; through random fluctuation, Open family, 98, 99
341 Open-integrated family, 93
Mutual reaction processes, 41-46 Operant conditioning, 169
Mystery, Milan Associates' use of, 303 Orfanides, Monica, 246, 247
Mystification, 36 Organization theory, 144; structural ap-
proach and, 264
Origin, families of: boundaries between nu-
clear family and, 73; in Bowen's approach,
29, 31, 243-45, 248-49; fusion with, 306
Nagy, Ivan, 17, 251-55, 306 "Outdoor" therapy, 258
Napier, A. Y., 231 "Outer ring" in systemic approach, 299
National Institutes of Mental Health, 17 Overinvolved parent, 212
Natural triads, 140-44 Oxford English Dictionary, 294
Naven (Bateson), xii, 38, 45, 48, 73
Naven ceremony, 38-40, 43, 124; as cyber-
netic mechanism, 46-47; as first order
change, 48, 49
Negative consequences of change, 238-41 Palo Alto group, 15, 28, 285, 289, 345; cy-
Negative feedback: interplay of positive bernetic analogy of, 177; on double bind,
feedback and, 52; in morphostasis, 50; in 168, 287; see also Bateson, Gregory; Mental
stable system, 75 Research Institute
Negative practice, 30577 Palo Alto Veterans Administration Hospital,
Negentropy, 51, 103, 340 19, 21
Neighborhood configurations, 261 Papp, Peggy, 190, 250, 251, 307^ 333/ 334/
Nett, Roger, 49 338
Neurotic behavior in centripetal families, 98 Paranoia: self-reinforcing relationship se-

372
INDEX
quencein, 41; therapeutic double bind in morphogenesis, 50; pathological balance
treatment of, 236-37 and, 63; step-mechanisms and, 165; ther-
"Paranoid Predators" marriage type, 185, apeutic use of, 229
186 Poverty: community psychiatry and, 257-
Paradox and Counterparadox (Milan Associates), 58; crisis intervention and, 258-61; family
285, 287, 304 organization and, 71-72, 80-85; norma-
Paradoxes, Bateson on, 115-16 tive model and, 264
Paradoxical techniques, 166-75, 239; with Power, issues of, 190-96
anorectic child, 206; in Bowenian therapy, Pragmatics of Human Communication (Wat-
248; Erickson's use of, 232; Madanes on, zlawick, Jackson, and Beavin), 17, 69, 70,
292; multigenerational, 251; in problem- 23°-37, 305, 345
solving approach, 280, 283; in strategic Prata, Giuliana, 284, 304
approach, 278; in structural approach, Predator-prey relationship, 42
270; in systemic approach, 287-go, 297- Prescribing the symptom, 19, 236; first use
g8; see also Therapeutic double binds of term, 305; in reframing, 277; resistance
"Parent-Child Coalition," 150 and, 324; in strategic approach, 289; in
Parental subsystem in normative model, 263 structural approach, 270
"Parenthood as Crisis" (LeMasters), 161 Pressure-box model, 344
Parsons, Talcott, 141, 142 Prigogine, Ilya, xiii, 5, 157, 340-42, 348

"Part-function" movement, 163 Princeton University, 45

Pathan clans, 44 Prison Notebooks (Gramsci), 165
"Pathological" balance, 62-64 Privacy, children's need for, 266-67
Pathological triads, 105-25; Caplow on, "Problem of the Referring Person, The"
109-12; communication and structure in, (Milan Associates), 299-300
105-8; conflict of levels in, 115-17; in Problem-solving approach, 27Q-8^
different cultures, 122-25; Haley on, 108, Problem-Solving Therapy (Haley), 280, 281
111-13; interconnection of, 113-14; "Process koans," 231
schizophrenic communication in, 117-20 Process models, 13, g6-ioo
"Patient and Therapist Observations on the Process taxonomy, 91, 156
Circumstances of a Schizophrenic Epi- Professional held, interference from, 299-
sode" (Weakland and Jackson), 23 300
Patten, John, 338 "Providing a worse alternative," 279
Pattern, metaphysics of, 86-91 Provocative approaches: Ackerman's, 223-
Paul, Betty, 190 28; Whitaker's, 228-31
Paul, Norman, xiii, 190; historical approach Pseudohostility, 33, 35, 182
of, 249 Psuedointegrated family, 93
"Pawnbrokers" marriage type, 185 Pseudomutuality, 31, 33-36, 89, 93, 176,
Penn, Peggy, 281, 338 252
Peripheral parent, 212 Psychoanalysis, 219; historical approach
"Perverse triangles," 108-13, 1 4°, 1 43> *44; and, 242, 243, 251; theory of repression
in different cultures, 122-25 in,250; training analysis in, 245; see also
Philadelphia Child Guidance Clinic, 215, Freud, Sigmund
230, 257, 262, 281 Psychodynamic model, 6; Ackerman and,
Phobias: in couple cycle, 190; of death, 302; 223, 225; in historically oriented ap-
problem-solving approach to, 281; school, proaches, 242-43, 249, 253
174/ 279 Psychodrama, sculpting as form of, 250
Pittman, C, 188 Psychosis: Ackerman's approach to, 224; in
Pittman, Frank, 259, 260 adolescence, 282; in centripetal families,
Plato, 104 98; episodes of, and couple bond, 190;
Piatt, J., 158-60, 166 point of intervention for, 201; Satir's tech-
Polarities, dialectic of, 295 nique for, 221-23; structural approach to,
Polarization, 132, and scapegoating,
131, 270; systemic approach to, 286; team ap-
137; in "special case" problem, 146-50 proach to, 332; see also Schizophrenia
Porpoises, learning behavior of, i6g-yo "Psychosomatic Avoiders" marriage type,
"Positive connotation," therapeutic device 185
of, 289-93, 306, 347 Psychosomatic disorders: conflict manage-
Positive feedback, 48; abortive, 60, 61; de- ment and, 73, 150-51, 178; point of inter-
structive effects of, 50-51; hubris and, 56; vention for, 201; structural approach for,
interplay of negative feedback and, 52; in 262; team approach for, 332

373
 INDEX
(Minuchin), 150, 203
Psychosomatic Families Ross, Jeffrey, 338
Psychotic children: parental marriage type Rotating scapegoat, 101
and, 185-86; see also Psychosis; Schizo- "Rubber fence," 35, 89, 94
phrenia Runaways, 43, 46, 60; in closed family, 98-
99; discontinuous change and, 160; pro-
voked by therapist, 203-4; second order
change and, 47-48; see also Positive feed-
back
"Quandary," 114 Runs, irresistible, 60-62
Questioning, circular, 300-302, 322 Russellian Logical Paradox, 112 m,
Russellian Theory of Logical Types, 27, 28

Rabkin, Richard, xiii, 167, 168, 256, 257,

3°5« Sado-masochistic couples, 70
Rahe, R. H., 161 Sager, Clifford, 190
Random families, 98, 99 "Saltology," 167
Ransom, D., 23 Sampson, H., 190
Rapaport, Rhona, 161 Sanity, Madness, and the Family (Laing and Es-
Rappaport, Anatol, xii terson), 36
Ratliffe, V., 188 Satir, Virginia, xiii, xiv, 17, 220-23, 2 5°,
Rausch, Harold, 194, 195 344
Ravich, Robert, 180, 195; couple typology Scapegoating, 36, 64; of adolescents, 101;
of, 186-88 balance theory and, 134, 137, 138; in
Recesses from therapy, 332 closed society, 132; in "Detouring-
Reciprocal marriage, 69 Attacking" triad, 151; in disengaged fam-
"Recurrent states" in family interaction, ily, 82; hospitalization and, 65; morale-
60-61 building function of, 59; as mutual causal
Recursive language, 7, 8 process, 57-58; Nagy on, 255; pressure for
Referring context, 299-300, 321 change and, 167; strategic approach to,
Reframing, technique of, 272, 275-77; re- 275; systemic view of, 288; of therapist,
sistance and, 324; in systemic approach, 330
289 Scheflen, Albert, 261
Reiss, David, 51, 66; family paradigm of, 87- Schismogenesis, 12, 40-41, 73, 138, 155; cy-
91, 93, 96, 99-100, 166 bernetic processes and, 47, 48; mathemat-
Repetition compulsion, multigenerational, ical expressions of, 43; societal checks on,

244 43-45; in spouse abuse, 189; symptomatic

Repression: abreaction and, 242; theory of, sequences and, 197; typology and, 69, 70
240-51 Schizophrenia: in adolescence, 215-17; as
"Rescue operation," 59, 62 appropriate response to real split, 149;
Resistance, 323-25; as balancing response, avoidance of defining relationships
312; in Bowenian therapy, 248; Haley's through, 117-19; and Bateson's commu-
techniques for management of, 279; ho- nication research, 18; bicameral model
meostasis and, 348; structural approach and, 4; clinical studies of family in, 28-29;
and, 270; systemic approach and, 303, control theory of, 26-28, 288; double bind
304; therapeutic contest and, 328-29; hypothesis of, 20-26, 105-6, 168, 313;
therapeutic double bind and, 305-6 early research on, 12; family alliances and,
"Restructuring moves," 225 112-13; family structure and, 81;
Reversals, 247 "homeostatic seesaw" and, 180-82; and
Reverse psychology, 305^ mediation function of deviance, 5Q-60;
"Revolutionary" coalition pattern, 110, 142 and negative consequences of change,
Richardson, Lewis F., xii, 41-43, 45, 70 238-41; Reiss's card-sorting experiment
"Rigid triads," 150, 151 on, 87-89; Rorschach study of, 67-68;
Riskin, J., 28 systemic approach for, 286, 287, 290, 301;
Role theory, 264 therapeutic double bind in treatment of,
Romeo and Juliet (Shakespeare), 178-79 236; three-generation hypothesis of, 29;
Rorschach test, 67-68 transactional model of research on, 86-87
Ross, Aileen, 123 "Schizophrenic" family, 66

374
INDEX
School phobia, 174; problem-solving ap- Speck, Ross, 261
proach for, 279 "Special case," problem of, 144-47; mirror-
Schroedinger, Erwin, 51 image disagreement in, 148-50
Schwartz, Morris, 106, 143-45/ 147-5©/ Speer, Albert, 9, 51
154, 202 Spiral Platters of Family Organization, 100-
Scott, Sam, 215 104
Sculpting technique, 250-51 Spouse abuse, cyclical nature of, 188-89
Second Law of Thermodynamics, 340 Stability, nature of, 317-18
Second order change, 47-49; in grid model, Stable psychological conflict, 114
92; symptomatic cycle and, 197-202 Stable-Satisfactory marriage, 185
Self:Bateson on, 342; differentiation of, 78, Stable system, attributes of, 75, 76
243-49; individuated, 242 Stable-Unsatisfactory marriage, 185, 186
Self-healing tautologies, 321 Stanton, Alfred, 106, 143-45, 147-5©/ 154/
Self-reinforcing relationship sequences, 40- 201
41; model of, 271-72, 276, 278 Stanton, Duncan, 306-7
Self-Starvation (Selvini), 284, 288, 294 State-determined system, 75
Selvini Palazzoli, Mara, 14, 284-86, 288, "Statistics of Deadly Quarrels, The" (Rich-
291, 294, 2Q7-Q&, 304, 313, 319, 321; see ardson), 42
also Milan Associates Status schisms, 143
Sensitivity training groups, scapegoating in, Steady-state systems, change in, 48
59 Steinglass, P.,188
Severe solitary delinquencies, 87, 89-90 Step-mechanisms, 76, 162-63
Sexual acting out, 272 Stierlin, Helm, 80, 254, 345
Shakespeare, William, 303-4 Strategic approach, 219, 235, 243, 248, 253,
Shannon, Claude, 5, 45 255, 271-78, 289; benevolent contest in,
Shtetl culture,
91 328; constancies in, 321; "neutrality" in,
Sibling subsystem in normative model, 263 303; paradoxical directive in, 232; prob-
"Sick" label, 138 lem-solving approach in, 279; reversals in,
Silverstein, Olga, 307*7, 338 247; for small relationship units, 336; sys-
Simmel, Georg, 132 temic approach and, 337; termination in,
Simon, Herbert, ^5 331
Simple bind, 168-69, 17 2 ' x 74' 2 3° (Rabkin), 305*7
Strategic Psychotherapy
Singer, Margaret, 17, 68, 87 106 , (Haley), 26, 318
Strategies of Psychotherapy

Single deviant arrangement, 131, 138, 146, Street gangs, authoritarian nature of, 101
155 Stress: individual responses to, 160; life
Single parents, variables of,179 stage, 161, 162; somatization of, 178;
"Singularity" of family, 341, 348 transfer of, 152-55
Skynner, Robin, 345 Structural approach, 219, 242-43, 248, 255;
Sluzki, Carlos, 23, 167, 168, 190 constancies in, 321; differences between
"Social Base of Sales Behavior, The" strategic approach and, 276-78; of Minu-
(Miller), 333 chin, 262-71; problem-solving approach
"Social envelope," family as, 191-92 and, 279
Social fields, 37-49; balance theory for, 127; Structural balance theory, 13, 113, 124-39;
contiguous systems in, $^; cybernetic clustering and, 132-34; "natural triad"
mechanisms in, 46-47; decoding of in- and, 142; family pathology and, 134-37;
teraction patterns in, 70; first and second family theory and, 126-32; trade-offs
order change in, 47-49; mutual reaction and, 137-39
processes in, 41-46; patterning of, 36-40; Structural typologies, 71-74
schismogenesis in, 40-41 Suicide attempt, systemic approach for, 302
Social psychiatry, 257 Sullivan, Harry Stack, xi
"Social Readjustment Rating Scale," 161,-62 Superego, counterautonomous, 252
Society for General Systems Research, xii Symbiosis, 243-44; unresolved, 29
Sociopathic behavior in centrifugal families, Symmetrical couples, 60, 6g, 70; change of
98 complementary couples to, 308-9; conflict
Solomon, Michael, 161 in, 182
Somatic illness, couples held together by, Symmetrical differentiation, 37, 41, 42, 45
189 Symmetrical relationship between groups of
"Spare-Time Battlers" marriage type, 185 inlaws, 320
Sparks, Geraldine, 251 Symptom substitution, 323-24

375
 INDEX
Symptomatic behavior: in adolescence, 172- Therapeutic double bind, 235-41, 305-22;
7y, as balancing response, 310-12; break- balancing responses to, 309-12; interac-
ing cycle of, 197-218; conflict and, 178- tive sequence constancies in, 318-22; of
79; in couple cycle,
190; double long interval between sessions, 320; par-
consequence of, emotionality and,
56; adoxical injunction and, 167-69, 171; in
247; family ledger and, 251; homeostasis strategic approach, 277; team approach
and, 180; linear models of, 6; Nagy on, and, 333-35
254; positive connotation of, 289-93; Therapy, Communication and Change (Jackson),
positive interpretations of, 347; prescrip-
tion of, see Prescribing the symptom; in Thermodynamics, Second Law of, 340
problem-solving approach, 280; repres- "Thing" language, 7
sion and, 240-50; step-mechanisms and, Third Reich, 101
164-66; strategic approach and, 271-72, Thomas, Lewis, 80
278; structural approach to, 263; systemic Thompson, Clara, xi
hypothesis and, 293, 294; therapeutic Thought disorders: and denial of align-
double bind and, 238; time aspect of, 296; ments, 118, 119; and double bind hy-
trade-off of, 138-39; typology according pothesis, 21, 22; Wynne's view of, 35-36
to, 58, 66-6g; variables associated with, Three-generation hypothesis for schizo-
178-70 phrenia, 29
Systemic approach, 219, 243, 253, 255, 284- Time: ignored in classical physics, 340; ir-
304; benevolent contest in, 328-30; circu- reversability and, 347; paradoxes and,
lar questioning in, 300-302; constancies 116; uses of, in systemic approach,
in, 321; counterparadox in, 287-8Q, 313- 295-99
17; for couples and individuals, 336; and "Time emergence," 159
evolution of problem, 325-28; hypotheses Timing in feedback loops, 53-54
in, 293-95; in individual work, 335; "neu- Todd, Thomas, 306
trality" in, 302-4; positive connotation in, Too richly cross-joined systems, 76-77,
289-93; referring context in, 299-300; 318, 322; contest-stance and, 329; Sel-
strategic approach and, 337; in teaching vini on, 286, 319-21
institute, 338; termination in, 331; uses of "Toward aTheory of Pathological Systems"
time in, 295-99 (Haley), 107, 108
"Systems Dilemma, A" (Hoffman and Long), "Toward a Theory of Schizophrenia" (Bate-
258 son, Jackson, Haley, and Weakland), 20,
Systems theory, structural approach and, 236
264 Towne, R. D., 190
"Sweat box," 170, 201 Train Game, 186, 195
Trade-offs for balanced forms, 137-39
Transactional disqualification, 167
"Transactional Disqualification: Research
on the Double Bind" (Sluzki, Beavin,
Tantrums, meaning of, for different people Tarnopolsky, and Veron), 23
in family, 183-84 Transference, 21; of members of family of
Tarnopolsky (author), 23 schizophrenic, 68
Tautologies, self-healing, 321 "Transformations," 159, 160; in family,
Taxonomy, process, 91, 156 166, 168
Taylor, Howard, 137 Treating the Troubled Family (Ackerman), 225
Taylor, William, 60, 131, 146 Triads, 24; Bowen's concept of, 30, 32; in

Team approach, 332-35; individual work coalition theory, 176; kinship net-
in
and, 335; Milan Associates
see also works, 107; and management of conflict,
Techniques of Family Therapy (Haley and Hoff- 140-55; natural, 140-44; pathological,
man), 6o, 210, 221, 235 105-25; in problem-solving approach,
"Ten-foot pole," 61 280-82; rules for congruence of, 126-39;
Tension, arousal followed by reduction of, Triangles: application of balance theory to,
43 129-31; benign, 140-43; in Bowen's the-
Termination of therapy, 330-32 ory, 29-32; "detriangling" of, 7Q; in fam-
Termite behavior, 80-81 ily of anorectic, 205, 206, 208, 209; in
Texas, University of, 5 historical approach, 243, 244, 246; inter-
Theory of Games (Von Neumann and Mor- pretive approach with, 210, 212; mother-
genstern), 25 daughter-grandmother, 314; perverse,

376
 INDEX
108-13, 122-25; reductive nature of Watzlawick, Paul, xiii, 17, 24, 70, 198, 203,

model based on, 218; in social groups, re- 213, 236, 241; Milan Associates a'nd,
search on, 107; warring, 113-14 287; strategic approach of, 271, 280, 289
"Triangulation," 150, 246 Weakland, John, ziii, 17, 23, 107, 198, 203,
Tribal networks, 261 213, 235, 241; Milan Associates and,
Trough equilibrium, 114 287; strategic approach of, 271, 280, 289;
"Truth" of systemic hypothesis, 293-94 three-party interaction work of, 24,
Two Against One (Caplow), 109, 125, 137, 105-6
140 "Weary Wranglers" marriage type, 185
Typecasting of deviants, 56-59, 64 Weimar Republic, 101
Typologies, 12-13, 67-&5; of couples, 185; Wertheim, E., ziii; grid model devised by,
grid model for, 91-96; interaction, 69- 91-96
71; for kinship and culture, 122; in pro- Western societies, emphasized relationship
cess models, 96-100; structural, 71-74; in, 122

symptom, ^S, 66-6g Whiffen, Rosemary, 345

Whitaker, Carl, ziii, 17, 101, 220, 228-31,
344
Who s Afraid of Virginia Woolf? ( Albee), 70
"Why a Long Interval Between Sessions"
Ulcerative colitis, 69 (Selvini), 286, 319
Ultrastable system, 75 Wiener, Norbert, 5, 45; on destructive
Uncommon Therapy (Haley), 232, 279, 280 effects of positive feedback, 50, 51
Underorganized family, $5, 262 Wild, Cynthia, 180-81
Undifferentiated family ego mass, 28-32, Wilkins, Leslie T., 57
72, 176 Wiltwyck School, 17, 256, 257
Unintegrated family, 93 "Working Family, The: A Researcher's
Unpredictability, concept of, 347-48 View of Health in the Household"
"Unresolved mourning process," 249 (Reiss), 99
Unstable-Unsatisfactory marriage, 185, 186 "Working through," 242, 325, 328
"Use of Fantasy in a Couples Group, The" Wynne, Lyman, ziii, 17, 68>, 87-89, 119,
(Papp), 190 170; pseudomutuality concept of, 31, 35,
93, 176, 252; researchon schizophrenia
D Y/ 33-36, 106

Variables, symptoms associated with, 178

Variety, 142; Buckley's concept of, 51
"Variety pools," 49 Yale University, 17, 67
Veron, Eliseo (author), 23 Yalom, J., 23, 238
"Vertical stressors," 253
Vietnam War, 296
Vogel, Ezra, 57, 59
Von Neumann, J., 25-26
Zen Masters, 168-69
"Zigzag ladder of dialectic between form
and process," 156, 280
Zuk, Gerald, 17
Walker, Gillian, 298^ 299, 307^ 334, 338 Zwerling, Israel, 261
"Wall of constancies," 320

377
 So oks.p
f^M,
CO
B °o
ks
3
$03
r ^