Paradoxical Intention

Promoting Change
Through
Paradoxical Therapy
Revised Edition
Edited by
Gerald R. Weeks, P h . D.
Copyright © 2013 International Psychotherapy Institute (1991 Gerald
Weeks)
All Rights Reserved
This e-book contains material protected under International and

Federal Copyright Laws and Treaties. This e-book is intended for
personal use only. Any unauthorized reprint or use of this material is
prohibited. No part of this book may be used in any commercial
manner without express permission of the author. Scholarly use of
quotations must have proper attribution to the published work. This
work may not be deconstructed, reverse engineered or reproduced in
any other format.
Created in the United States of America
For information regarding this book, contact the publisher:
International Psychotherapy Institute E-Books

301-215-7377
6612 Kennedy Drive
Chevy Chase, MD 20815-6504
www.freepsychotherapybooks.org
ebooks@theipi.org
About the Editor
Dr. Weeks is a Professor at the University of Nevada-Las Vegas and

maintains a small private practice. He is a Fellow, Approved-Supervisor,
and Clinical Member of the American Association of Marriage and
Family Therapy, and is a Senior Examiner and Diplomate of the
American Board of Professional Psychology and the American Board of
Sexology and a Certified Sex Therapist of the American Association of
the American Association of Sex Educators, Counselors, and Therapists.
He has published 20 professional textbook in the fields of sex, marital,
and family therapy with at least one book in each field being considered
a “classic” text.
Dr. Weeks is among a handful of individuals to be awarded the

“Outstanding Contribution to Marriage and Family Therapy in 2009
from the American Association of Marriage and Family Therapy and
was named the “2010 Family Psychologist of the Year.” He lectures
nationally and internationally.
In 1982, he published Paradoxical Psychotherapy: Theory and Practice

with Individuals, Couples, and Families (with Luciano L’Abate). The
book was translated into several languages and was one of the top five
best-selling professional books for years. Among his most popular
books are: Couples in Treatment (with Stephen Fife), Focused
Genograms (with Rita DeMaria & Larry Hof), A Clinicians Guide to
Systemic Sex Therapy (with Katherine Hertlein & Shelley Sendak),
Systemic Sex Therapy (with Katherine Hertlein and Nancy Gambescia),
Treating Infidelity (with Nancy Gambescia & Rob Jenkins), If Only I
Had Known: Common Mistakes in Couple Therapy (with Mark Odell &
Suzanne Methven) and The Handbook of Family Therapy (with Tom
Sexton and Michael Robbins).
For additional information about Dr. Weeks please google, “Gerald R.

Weeks” to learn more about his books, lectures, awards, etc.
Contents
C on tribu tors...........................................................................................vii
P refa ce....................................................................................................... ix
Part O ne Introduction
1. A D elp h i Study o f Paradox in

T h e r a p y .................................................................................................2
by Christine Watson, Ph.D.
Part Two A pproaches to Paradoxical Intervention
2. U n p red ictab ility and C hange: A H olographic

M eta p h o r............................................................................................ 28
by Brian W. Cade, CSW
3. Beyond Paradox and C ou n terp arad ox...................................... 60

by Klaus G. Deissler, D iplompsychologe
4. Paradoxical Intention .......................................................................99

by Viktor E. Frankl, M .D., Ph.D.
5. Paradoxical Techniques: O ne Level o f

A bstraction in Fam ily Therapy ............................................ I l l
by Luciano L'Abate, Ph.D.
6. E ricksonian Styles o f Paradoxical T reatm en t..................... 134

by Stephen R. Lankton, MSW, and Carol H. Lankton, M A
7. Paradox in C o n te x t.......................................................................... 187

by H oward Tennen, Ph.D., Joseph B. Eron, Psy.D.,
and M ichael Rohrbaugh, Ph.D.
v
Vi CONTENTS
Part T h ree R esearch on Paradoxical Techniques

8. A n Introduction to R esearch on the C lin ical
E fficacy o f Paradoxical In ten tio n .......................................... 216
by Ray S. Kim , M A , James Poling, M A , and
L. M ichael Ascher, Ph.D.
Part Four Future D irections

9. T h e M ysteriou s A ffair o f Paradoxes and L o o p s.................252
by Steve de Shazer and Elam N unnally
10. C ontradiction and Its R esolu tion am ong

the Psychotherapies: R esu lts o f a
P relim in ary In v estig atio n ........................................................ 271
by M ichael J. Bopp, Ph.D.
11. A M etatheory o f Paradox ............................................................ 302

by Gerald R. Weeks, Ph.D.
Contributors
L. Michael Ascher, Ph.D. Professor and Director of the Behavior

Therapy Unit, Department of Psychiatry, Temple University
Medical School, Philadelphia, Pennsylvania.
Michael J. Bopp, Ph.D. Private practitioner in Philadelphia,
Pennsylvania.
Brian W. Cade, C.S.W. Family Therapist and Course Organizer at
the Family Institute, Cardiff, Wales.
Steve de Shazer Associate Professor at the School of Social Welfare,
University of Wisconsin-Milwaukee, and Director of the Brief
Family Therapy Center. Steve de Shazer is the author of Patterns
of Brief Family Therapy (Guilford Press, 1982).
Klaus G. Deissler, Diplompsychologe. Co-founder and Director of
the Institut fiir Familientherapie Marburg. Dr. Deissler was
founding member of the German Association of Family Therapy,
editor of Kontext, and an editorial board member of
Familiendynamik.
Joseph B. Eron, Psy.D. Director of the Catskill Family Institute in
Kingston, New York, and a family therapist in private practice.
He received his doctorate in psychology at the University of
Illinois at Champaign-Urbana. Dr. Eron is a clinical faculty
member at Albany Medical College, Mt. Sinai Medical Center,
and Beth Israel Medical Center.
Viktor E. Frankl, M.D., Ph.D. Professor of neurology and
psychiatry at the University of Vienna Medical School and
distinguished professor of logotherapy at the United States
International University (San Diego). Dr. Frankl is the author of
26 books that have been translated into numerous languages,
including Japanese and Chinese.
Ray S. Kim, M.A., Ph.D. candidate, Department of Psychology,
Division of Clinical Psychiatry, Temple University, Philadelphia,
Pennsylvania.
Luciano L'Abate, Ph.D. Former Professor and Director of the Family
v ii
V ili CONTRIBUTORS
Psychology Program at Georgia State University. Dr. L'Abate has

authored numerous books and articles, including Paradoxical
Psychotherapy with Gerald Weeks.
Carol H. Lankton, M.A. Private practitioner in Florida. She
conducts workshops on Ericksonian Therapy throughout the
world. She is the co-author of several books and articles that
describe the use of Ericksonian psychotherapy with families,
children, and adults.
Stephen R. Lankton, M.S.W. Private practitioner in Florida. He
conducts workshops on Ericksonian therapy throughout the
world. He is founding editor of The Ericksonian Monographs and
author or co-author of several books and articles that describe
the use of Ericksonian psychotherapy with families, children,
and adults.
Elam Nunnally Staff member at the Brief Family Therapy Center in
Milwaukee, Wisconsin.
James Poling, M.A., Ph.D. candidate, Department of Psychology,
Division of Clinical Psychology, Temple University, Philadelphia,
Pennsylvania.
Michael Rohrbaugh, Ph.D. Associate professor of psychology,
College of William and Mary, and Director of Family Therapy
Training for the Virginia Consortium for Professional
Psychology.
Howard Tennen, Ph.D. Associate Professor of Psychiatry at the
University of Connecticut School of Medicine, where he is
director of psychology training. He received his graduate
training at the University of Massachusetts. His research
interests include responses to uncontrollable stress, coping with
serious illness, and the role of paradox in strategic therapy.
Christine Watson, Ph.D Private practitioner in Houston, Texas.
Gerald R. Weeks, Ph.D. Director of Training, Marriage Council of
Philadelphia; and Clinical Associate Professor of Psychology and
Psychiatry, University of Pennsylvania School of Medicine.
Preface
A few years ago paradox in psychotherapy was used by only a few

practitioners. M any perceived it as strange, noncom monsensical,
and too directive. Yet it was powerfully attractive because of its
som etim es rapid and dram atic effects. W hen paradox is used
adeptly, the results have been so startling that early practitioners—
and some of the more recent ones— have been called "therapeutic
m agician s."
Following the publication of the book Paradoxical Psychotherapy:
Theory and Practice with Individuals, Couples, and Families (W eeks &
L'Abate, 1982, Brunner/Mazel, New York), a keen interest in this
approach developed. In Paradoxical Psychotherapy we attem pted to
dem ystify the approach by explicating the different m ethods of
working paradoxically. Shortly after the book's publication, several
others appeared, w hich focused directly or indirectly on this
approach.
The psychotherapeutic field was and is rich w ith excitem ent,
controversy, and new ideas about the application of paradoxical
m ethods. M any of those writing about this approach did so in the
context of other therapies such as logotherapy and behavior
therapy.
The purpose of this volume is to bring together the w ritings of
a num ber of well-know n therapists w ho are pioneers of this
approach. A lthough this book was originally published in 1985,
and the editor acknow ledges that the subsequent w ritings o f the
authors show further reprim and, elaboration, and development,
the basic theories and conceptual frameworks are timeless in their
capacities to teach.
The book is divided into four parts. Part O ne asks the question,
W hat is paradoxical therapy? It exam ines the issue of w hat we
claim to do— how we define a therapeutic paradox.
Part Two is the theoretical section o f the volum e. C ontributors
describe their ow n theories o f paradoxical therapy, w hich is a
unique focus considering that m ost o f the literature in this field
ix
X PREFACE
goes over technique. In the third section o f the book M ichael

A scher and his colleagues have com pletely revised and updated
their chapter to present the latest research findings.
The last part of the text, Part Four, is designed to help synthesize
and integrate the different perspectives offered in this volume. O f
particular interest is the final chapter, w hich develops a m etathe
ory of paradox to help to u nify thinking about paradox and to tie
together w hat appear to be divergent theories.
Since the publication of this book in 1985, articles on paradox

have proliferated at an astounding rate, especially in behavior ther
apy. This volum e is an im portant m arker for all those interested
in this approach. It contains much o f the sem inal theory that has
lead to the use of paradoxical techniques in different system s of
therapy.
Gerald R. Weeks, Ph.D.

Part One
Introduction
1
A Delphi Study of Paradox
in Therapy*
by Christine Watson, Ph.D.
The purpose of this study was to provide an operational defini

tion for the concept of therapeutic paradox. M uch of w hat has
been written about paradox in psychotherapy has been in books
explaining theory and ways to use paradox and in articles giving
case study examples. In these publications, paradox as a tech
nique for strategic com m unication is described in m any varying
ways. Further, clinicians appear to differ about which persons can
benefit from its therapeutic use. Part of the difficulty in talking
about the concept in sharing creative solutions with each other
has been the several different languages being spoken with re
spect to paradox. Thus, this study attem pted to develop opera
tional definitions of therapeutic paradoxes and to form ulate them
hierarchically.
*A few participants of this study did not wish to be identified. The author
gratefully acknowledges the cooperation of all the participants, including: Phi
lippe Caille, Gina Abies, Y. L'amontagne, David Keith, Jeffery Brandsma, Susan
McDonald, Lynn Hoffman, Rachel Hare-Mustin, Seymour Radin, Joseph Lisiecki,
Bradford Keeney, d in t Phillips, L. Michael Ascher, Phoebe Prosky, Paul Dell,
Richard Rabin, Donald S. Williamson, Steve de Shazer, Arthur M. Bodin, Carlos
E. Sluzki, Brian Ackerman, Luciano L'Abate, and Gerald R. Weeks.
2
A DELPHI STUDY O F PARADOX IN THERAPY 3
Since the concept of paradox is still in the formative stages, a

m ost suitable design was the Delphi process, a m ethod originated
by the Rand Corporation for exploring group opinion and deci
sion formation. In using this process, participants, who are usu
ally experts, are asked to give their opinions on an unresolvable
or currently evolving topic or theory. Statem ents are collected and
a com posite is form ed, without acknowledgm ent of individual
authorship. The com posite is returned to each participant for con
sideration. Participants are asked to rewrite their opinions based
on the new knowledge contributed by other experts. This process
is repeated until som e consensus is reached.
The process of gathering written anonym ous statem ents and
providing controlled feedback derives from psychological princi
ples that reduce or eliminate many distortions that may occur in
open com m unication w here high status group m embers, domi
nant personalities, m ajority opinion, and prior public sentim ent
are likely to affect the form ation and expression of individual
judgm ent and thus influence group consensus. The idea of con
trolled feedback is related to inform ation theory and cybernetics
and is a technological development of World War II (Weiner,
1967).
In this study, the Delphi m ethod was used to explore three
questions. Participants (panel experts) were persons who had
either published on paradox in therapy or who were know n to
use paradox as a m ajor therapeutic approach. The 26 panel m em
bers (who had responeded to an invitation to participate in the
study) were asked to give their opinions on three questions. Their
responses were collected and analyzed, and com posites were
formed for each of the three questions; individual authorship was
kept anonymous. The com posites were returned to each partici
pant for reconsideration. Panel m em bers were asked to reform u
late their opinions based on the new knowledge they had gained
as a result of contributions by other experts and their knowledge
of the "w h o le." This process was repeated for three rounds, with
a final, fourth step consisting of the com pleted research and a re
quest for further com m ents.
The three questions asked dealt with (a) operational defini
tions of paradoxical intervention, (b) options for paradoxical inter
vention involving a hypothetical family, and (c) what criteria, if
any, the therapist used w hen choosing paradoxical language in
therapy.
This chapter contains the responses to and com m ents on the
first question—the experts' operational definitions of paradoxical
intervention.
4 CHAPTER 1
An interesting sidelight on the Delphi is found in the m eanings

associated with the naming of the technique. Originally, the
Delphi experiments were created to aid in predicting the future
via informed consensus of a group of experts. The name Delphi
was used for the process; it is associated with the historical Greek
oracle at Delphi, a place where the divinities gave answers to
those who consulted them about the future. The findings from
the original Delphi experiments did not predict the future; rather,
the process took an unexpected twist. The Delphi was instrum en
tal in creating future directions by stimulating informed consen
sus among anonymous experts. The name Delphi seem s apropos
for the results of the original experiments, as the priestess of the
oracle at Delphi was known for her ambiguous, paradoxical,
multilevel utterances.
Instrumentation
The research instrum ent used in this study was the question
naire, which was sent to a panel of experts considered knowl
edgeable about paradox in therapy.
Data Collection
The data collection followed these procedures:
1. Fifty-five persons who had published on paradox in therapy
or who were known to use paradox as a meaningful therapeutic
approach were sent a letter describing the study and its purpose
and inviting them to participate. A stamped postcard was in
cluded which those interested in being on the panel were asked to
return to the researcher. Thirty-seven persons responded to the
invitation. O f these, 26 said yes to participation, 7 said no, and 4
dropped out after the first round. All 26 respondents who re
turned the card saying yes were sent each round of the question
naire. Participation was not consistent over the length of the
study period. Som e members responded to Round O ne and not
to Round Two; some did not respond until Round Two; some
answered certain portions of the questionnaire and not others;
and some wrote letters sharing their views in depth.
Panel members were given a choice in Rounds Two and Three
of responding either by mail or telephone. Six experts who were
contacted by phone shared their views orally, an alternative that
proved viable for persons with time constraints.
2. The process of sending the first round of questionnaires, re
ceiving the responses, and analyzing and form ulating the com
posites from the first set of answers is referred to as Round O ne.
[Round One]
Each panel m ember was asked to provide a brief definition (no
more than five or six sentences) in his/her own words of what
constitutes a paradoxical intervention. Ten days after the first
questionnaires were sent to the panel m embers, follow-up re
minders were mailed.
3. Representative definitions of paradoxical intervention were
formulated by the researcher based on the definitions found in
the existing literature.
Definitions obtained from this round were read by the re
searcher and com posite definitions were created based on those
received from the panel members and on the representatives from
the literature. The com posites were formulated by conducting a
conceptual analysis of the data.
[Round Two]
4. A second questionnaire, containing the com posites drawn
from the first round, was sent to the panel. M em bers were asked
to respond to the new information, and to make any changes,
additions, and/or deletions that m ight make the com posites more
acceptable to them.
The actual choices presented consisted of several com posite
definitions, derived from analysis of Round O ne questionnaire re
sponses. Care was taken to ask the questions in a way that re
spected the individual characteristics of original answers.
Panel m em bers' responses to Round Two were returned to the
researcher and read. A second round of com posites was then con
structed by conducting a conceptual analysis of these responses.
[Round Three]
5. These com posites were sent to the panel m embers in a
Round Three questionnaire for final additions, changes, or dele
tions. After the second set of responses was received, final defini
tions were form ulated. These were sent to the panel members
who were asked to select their single favorite response to the defi
nitional question.
[Round Four]
The entire paper was sent to panel m embers with requests for
their com m ents. These responses com pose Round Four.
6 CHAPTER 1
Analysis
A stepwise format for conducting conceptual analysis of verbal
qualitative data, as elucidated by Tesch (1980) in Phenomenological
and Transformative Research: What They Are and How to Do Them was
used to analyze the data. We will present a brief sum m ary of the
steps in this type of analysis to give the reader an overview of the
process before we elaborate on the actual findings.
Step 1. First, each respondent's com plete answers to the ques

tionnaires were read, thereby enabling the researcher to develop a
sense of the entire picture before beginning the actual analysis.
The data were sorted according to the organizing them es pre
sented in the answers. Themes, in this study, refer to clusters of
ideas that panel experts may have held in com m on in their defini
tions of paradoxical intervention.
Step 2. Each them e was studied to determ ine what the panel
members were saying about paradoxical intervention. Som e help
ful study questions suggested by Tesch include:
Which comments seem to express the same idea or opinion about
the theme? What is this idea or opinion, in your own words? Is
there unity about the theme (i.e., most people say something simi
lar), or is there controversy? What is the nature of the difference of
opinions, and how many people take the one or the other stance?
Is the language passionate, strong, personal, or matter-of-fact, un
involved, general? The answers to these questions are recorded in
the form of a summary of the theme substance. (Tesch, 1980, p. 32)
Step 3. The them es were com pared with each other to see if any
were sufficiently alike to be grouped together. For exam ple, the
answers to the question about operational definitions of paradoxi
cal intervention produced several different them es, some of which
were com bined due to similarity; a few distinctly different them es
evolved from the experts' responses to this question.
Step 4. The m ajor them es culled from the initial analysis were
compiled in a second round questionnaire and sent to the panel
members, who were asked to read them and select the one that
they thought came closest to their own definition. (Panel m em
bers were also asked to make com m ents, additions, or sugges
tions for possible deletions that would bring the definition closer
to their own ideas.)
The experts analyzed the responses to the them es (presented to
them as feedback), again looking for com m on m eaning units. Here

the researcher exam ined the panel's com m ents, additions, and so
on, as well as their actual choices of m ost acceptable definition.
Step 5. Based on the initial and feedback responses, three litera

ture definitions and six operational definitions were form ulated
for the concept of paradoxical intervention. These were sent to
panel members for a final selection of the definition that came
closest to their own views.
DEFINITION OF PARADOXICAL
INTERVENTION
Round One
Step 1 . The participants' answers to the first question in this

study were read through twice to enable the researcher to get a
sense of the responses before beginning the actual analysis. The
data were sorted according to organizing them es presented in the
answers.
Step 2: Summary of Theme Substance. The request for an

operational definition of paradoxical intervention was: "W h at ele
m ents are necessary for an intervention to be categorized as a
paradoxical intervention?" Eighteen responses were analyzed.
(Four of these responses were received after the first analysis of
Round O ne responses and subsequent sending of Round Two
questionnaires.) Brief sum m aries of the answers to the defini
tional question are presented below to give the reader a sense of
the variety of approaches taken to define paradoxical interven
tion. Five answers were similar in that they centered on the them e
of "co u n terin g" the presenting sym ptom . O ne response was mat
ter-of-fact in defining a paradoxical intervention as that which
produces second-order change; a therapeutic double bind used to
treat individuals trapped in a pathogenic double bind. O ne respon
dent stated that paradoxical intervention is defined from the
client's perspective, in that the client m ust perceive the interven
tion as not addressing his/her therapeutic goals in a com m on-
sense m anner: The intervention may be illogical, opposed, or ir
relevant to the goals of therapy.
A nother m em ber defined paradoxical intervention as an ap
proach with an elem ent of surprise which resolves a paradox by
8 CHAPTER 1
changing the epistemological frame of reference. O ne respondent

em phasized that the specific frame of reference is the therapist's
prescribing of the symptom. The paradoxical statem ent implicit
here is: “ Do what you were doing, and it will be different." (From
a process perspective, doing now what you were doing, because
you are now being told to do it, is different.) A nother respondent
saw paradoxical intervention as the therapist com m unicating on a
more encom passing level of abstraction than the client, in order to
change the client's frame of reference. Yet another respondent de
fined paradoxical intervention as the com m unication of two ap
parently contradictory m essages which, w hen introduced into a
rigid idea framework, disrupt the rigidity and allow for individua
tion and great creativity. O ne expert preferred the term restraint
from change, seeing it as more descriptive than paradoxical inter
vention. And another participant defined a paradoxical interven
tion as a global positive connotation of the behaviors of all partici
pants "in that context and history." Three panel members sub
mitted no definition.
Two distinct, broad themes em erged from the definitional ques
tion in Round O ne, each of which contained subthem es. These
them es are as follows:
Theme 1. Paradoxical intervention is a term for w hat the thera

pist does in response to client/family behaviors (especially resist
ant behaviors). Specifically, s/he:
a. Restrains the family from change.
b. Sends two contradictory m essages: O ne is the context of the
therapeutic relationship, which implies that change is to take
place; the other, given directly by the therapist, says "d o n 't
change."
c. Exaggerates some client/family behavior, or uses surprise to
shock the family out of rigid frame of reference (as in som e of
the Zen koans).
d. "C o u n ters" the family's paradoxical ways of looking at their
problem with a "th erap eu tic" paradox.
e. Locates the possible function of a problem behavior or any
behavior and prescribes it (behavior and function) as a ho
meostatic maintainer (referred to as second-order change).
f. Positively connotes behaviors of all participants in a particular
context and history.
Theme 2. Paradoxical intervention can be viewed in term s of the

client's/family's perception.
a. It is paradoxical w hen a client/family sees the therapist's inter

vention as illogical, lacking in com m on sense, or irrelevant to
the goals of therapy.
b. The intervention isn 't paradoxical per se—the family's chosen
perception of the behavior creates the paradox. Paradox is in
the eye of the beholder.
These broad them es represent two very different perspectives
on paradoxical intervention: O ne sees it in terms of the therapist's
response to client behavior, the other in terms of the client's/fam
ily's perception of the therapist's intervention or of the behaviors
presented.
Round Two
Step 4. Nine com posite responses to the Round O ne question

(operational definition of paradoxical intervention) were compiled
in a Round Two questionnaire and sent to panel m embers. M em
bers were asked to read the selections and choose the one or two
which they thought came closest to their own definition. In addi
tion, members were given nine different literature definitions and
asked to choose the one or two they thought came closest to their
own operational definition.
Summary Results from Round Two Questionnaire

Nineteen panel members responded to the Round Two ques
tionnaire. The favored definitions are given below, in addition to
the num ber of respondents who chose each of them . There are
two parts to this definitional question: Part A contains the favored
responses to the literature definition of paradoxical intervention.
Part B includes responses to the com posites form ulated by the re
searcher, based on Round One responses from the panel of experts.
Part A : Literature Definitions. The literature definitions fa

vored by most panel members stated:
A paradoxical intervention can be defined as follows: "Within a
context where the client comes to therapy in order to change, the
therapist asks him to produce more of the behavior that the client
wants changed, and within a context of acceptance of the involun
tary nature of the client's behavior, the therapist requests that the
patient produce this behavior voluntarily." (Madanes, 1981, pp.
7-8)1
’ References to the literature definitions were not provided in the study.

10 CHAPTER 1
The second m ost favored definition from the literature was 9, with
nine respondents choosing it. This definition reads as follows:
A paradoxical order is one which, if correctly executed, is dis
obeyed; if disobeyed, it is obeyed.
A paradoxical intervention then, is any intervention, command,
prediction, request that, if followed or accepted, will accomplish
the very opposite of what it is seemingly intended to accomplish. It
is dialectics as applied to psychotherapy, consisting of seemingly
self-contradictory and sometimes absurd therapeutic interventions
which are always constructively rationalizable, although sometimes
very challenging, and which join rather than oppose symptomatic
behavior, leading to increased social interest. (The notion of dialec
tical thinking can be summarized in the idea that things are not
what they seem.) The success of the paradoxical intervention de
pends on the family's defying the therapist's instructions or follow
ing them to the point of absurdity and then recoiling. "If a family
continually defies compliance-based interventions, it can be safely
assumed there is some hidden interaction in the system that under
mines their usefulness—some secret alliance, contest, or coalition
that the family is reluctant to reveal or change. The target of the
systemic paradox is this hidden interaction that expresses itself in a
symptom." (Papp, 1980, p. 46)
The third most favored definition, with six panel m embers choos
ing it, was 5, which reads:
Paradoxical intervention refers to any intervention designed to pro
duce what Watzlawick et al. (1974) called second-order change as
opposed to first-order change. (Second-order change refers to
change in the system itself, e.g., the rules of the system, and is
called paradoxical change.) (Weeks, 1979, p. 62)
Part B. Operational Definitions from Panel M embers. The

favored operational definition in Part B, chosen by eight panel
m embers, was no. 3, which states:
A paradoxical intervention is the prescribing of the symptom or pre
senting problem; i.e., it is to commend that the client (individual,
couple, or family) do something they are already doing, without first
making explicit that they are doing it; and so they should do this, as
if, as it were, they were not already doing it. Secondly, they should do
it (as if they were not already) because in the therapist's view, doing
it (i.e., doing what they were doing) will bring change. The para
doxical statement implicit is: "Do what you were doing, and it will
be different." Process-wise, doing now what you were doing, be
cause you are now being told to do it, is different.
The second m ost favored operational definition, chosen by seven
respondents, was no. 2, which reads as follows:
A paradoxical intervention is one that has an element of surprise

(i.e., is contrary to context-bound expectations) and resolves a
paradox by changing the epistemological frame; it is an unexpected
exaggeration of a pattern of behavior that has been previously re
sistant to other forms of intervention.
In addition, five experts presented new ideas for the definition of
paradoxical intervention. Three of these were included in the Round
Three questionnaire, as they represented totally different ap
proaches to the definition of paradoxical intervention. A fourth re
sponse was received after the Round Three questionnaire had been
sent. It represents a different view and is presented here verbatim:
To me, the archetype of the paradox is most effective when it is the
by-product of the striving for consciousness. And that striving
leads to a confrontation with the paradox in question simulta
neously with an awareness of the paradox in all life.
There is a challenge in the statement, "You can't get there from
here." To contrive to precipitate an awareness of a paradox pre
maturely or as a technique is oppositional and destructive to the
patient.
A paradox functions like a Zen koan. It thrusts you back on
deeper resources. The famous example, "What is the sound of one
hand clapping?" confronts you with the futility of the ordinary in
tellectual defensiveness.
Round Three
Step 5. The favored responses from Round Two plus three of the
new definitions were presented to the panel of experts in the
Round Three section containing operational definitions of para
doxical intervention. Panel members were again asked to choose
which definition came closest to their own (choosing only one defi
nition each). In addition, three definitions from the literature,
which were favored in Round Two, were presented and each m em
ber was asked to choose one which best fit his/her own definition.
Summary of Results from Round Three

Questionnaire
Sixteen responses to Round Three questionnaires were re
ceived. In the first section on the definition of paradoxical inter
vention (from the literature), two definitions were favored.
Definition A was m ost favored, being chosen by six panel
m embers. It reads as follows:
12 CHAPTER 1
A. A paradoxical intervention can be defined as follows: Within a

context where the client comes to therapy in order to change, the
therapist asks him to produce more of the behavior that the client
wants changed, and within a context of acceptance of the involun
tary nature of the client's behavior, the therapist requests that the
patient produce this behavior voluntarily. (Madanes, 1981, pp. 7-8)
It refers to any intervention designed to produce what Watzlawick
et al. (1974) called second order change as opposed to first order
change. (Second order change refers to change in the system itself,
e.g., the rules of the system, and is called paradoxical change.)
(Weeks, 1979, p. 62)
Definition C was second favorite, being chosen by five respon
dents. It read as follows:
C. Paradox is inherently an epistemological phenomenon. The
“existence" of paradox requires the premises and beliefs of an ob
server. There are no phenomena which are paradoxical in them
selves—apart from an observer. The major premise that generates
"paradox" is the (false) assumption that we live in an Aristotelian
universe. This premise requires that self-recursive phenomena be
considered to be somehow unreal or illusory. (Self-recursiveness is
cybernetic feedback wherein a statement, organism, or system al
ters its own behavior because its previous output feeds back to it
and modifies its subsequent behavior.) Such epistemological (and
ontological) disqualification of self-recursive phenomena leads to a
variety of consequences, perhaps the least of which is "paradox."
.. .There is no such thing as "a paradox." Rather, all "paradox"
exists only in the mind of the beholder. Accordingly, all attempts to
explain paradox either as an entity or as a therapeutic intervention
are fundamentally flawed and can lead only to further conceptual
confusion. (Dell, 1981, p. 127)
Responses to the operational definitions (Section II) were spread
among three of the six choices. Definition B received the most
agreement, with five respondents choosing it. This definition
read:
B. A paradoxical intervention is one that has an element of sur
prise (i.e., is contrary to context-bound expectations) and resolves a
paradox by changing the epistemological frame; it is an unexpected
exaggeration of a pattern of behavior that has been previously re
sistant to other forms of intervention.
The second favored response was chosen by four panel m em
bers and read as follows:
A. A paradoxical intervention refers to a counter-intuitive instruc
tion or request made by the therapist, bypassing the patient's dem
onstrated tendency to discount, criticize, forget, postpone, or mis

understand simple, straightforward directions or neutralize them
by claiming inability to comply. Thus, if a person says he/she is
helpless to stop a symptom, by definition he/she is not helpless to
continue performing the symptom if that is what is prescribed.
(Counter-intuitive, in this instance, means "against the direction
prescribed by common sense.")
The intervention is not a "thing," is not the "change agent." The
change agent is within the process which encompasses the "fit"
between the patterns of the client/family and the pattern of the inter
vention.
A third favorite was one of the new responses, included in
Round Three because it represented a new approach. This re
sponse received three votes of agreement from panel m embers.
D. There exists an interpretation or instruction to the family that
restrains them from the change they have come to therapy to ask
help for, or from any change at all. Since this is done within a context
that implies that the family's request will be honored, a contradiction
is set up. This contradiction is often called a "paradoxical interven
tion." As the intervention only seems paradoxical or illogical from the
point of view of client expectations, the descriptive term, "restraint
from change," may be used to refer to the situation mentioned.
Telling a person/family to continue with a problem, and giving a
rationale that is validated by at least some of the family's percep
tions, is not "paradoxical." It is contrary to expectation. The ra
tionale may be trivial or incomplete. Using such a trivial rationale
to restrain the family from change may seem absurd. But it is not
paradoxical because at the system level it makes sense from many
angles; it has justifications any one of which one might choose to
use as a rationale for restraining the family from change. Which
one, of course, is the essence of the art.
Perhaps, just as in dreams and animal communications, there is
no way to say "not" to people caught in a symptomatic configura
tion. To say don't change within a context that implies change seems
to be analogous to Bateson's description of an otter communicating
"play" to another otter by pretending to fight. By biting an ear but
not biting it hard, one otter signals to the other "this is play." Peo
ple get the message when it is put this way where they often don't
when told to change directly.
In addition to the above three choices, there was one distinc

tively new approach to the definition question. In this approach,
paradox is seen as "sy n ta x " of a certain type. In other words, you
can speak in paradox language or in ordinary English ju st like you
can speak in prose or poetry. Quoting the panel m ember further,
" I have yet to work out what the rules are but they deal mostly
14 CHAPTER 1
with valuing and giving directions. Instead of saying in prose,

'That is bad. Do this instead/ in paradoxical syntax you first as
sign levels and then say, 'It is good that that is bad.' A similar
form would be used for the com m and."
Quoting further, "a n operational definition of a syntax is a
grammar. The elem ent of surprise and noncoerciveness would be
probably the same if one spoke in rhym e or in elaborate flowery
gestures, etc."
This last definition em phasizes the language form as im portant
in understanding paradox, especially as it com pares with straight
interventions.
Conclusions
Paul Dell (1981) m entioned in a recent article that any attem pt to
define paradox either as an entity or as a therapeutic intervention
leads to further conceptual confusion. Giving credence to this
statem ent, we can proceed to "m ak e sen se" of the wide variety of
responses to the first question in this study—how to define para
doxical intervention. W hat seem s even more perplexing is attem pt
ing to define an intervention in which paradox is either sponta
neously or purposively used in com m unications with clients, or is
at least recognized as being present in such com m unications. It is
helpful to keep in mind that the term paradoxical intervention is very
new in the literature of therapeutic com m unication.
The rationale for seeking a definition was that by first learning
about the operational definitions of panel m em bers, the re
searcher would be able to grasp more of the m eanings in re
sponses to another question: W hat are the criteria in deciding to
" u s e " or not to " u s e " paradoxical interventions?
The final analysis of the various definitions for paradoxical in
tervention showed that experts used several conceptual categories
in thinking about paradoxical interventions. We will discuss these
categories, starting with the more general and abstract concep
tualizations and proceeding to the more specific and concrete
ones. As one m ight well imagine, the categories were not cut-and-
dried. Rather, several of the conceptualizations overlapped. In
addition, som e panel members defined paradoxical intervention
with differing em phases. The three m ajor areas of em phasis con
sisted of m essage com ponents, process com ponents, and more
theoretical aspects.
There were seven m ajor approaches to defining w hat was
m eant by a paradoxical intervention. They are listed below and a

discussion of each follows (P.I. = paradoxical intervention).
1. P.I. is in the mind of the beholder.
2. P.I. is intuitive and experiential.
3. P.I. is an archetype symbolic of paradox in all of life.
4. P.I. is connected closely to concepts of circularity and the sys
temic nature of interventions.
5. P.I. is dependent upon client's perception of a specific inter
vention.
6. P.I. is a term for som ething specific the therapist does.
7. P.I. is specific syntax.
Paradoxical intervention, w hen viewed as “ in the mind of the
beholder," represents the m ost abstract way of understanding this
concept. Its m essage can be sum m ed up in the statem ent: W hat is
paradoxical to you may not be paradoxical to me; no two “ reali
ties" are alike.
The second approach to a definition considers paradox intui
tive, experiential, and a product of the therapist's personality and
lifestyle; thus it cannot be explained or defined without “killing"
it (just as a good joke loses its punch upon elucidation). This ap
proach, though still som ewhat general and elusive, does give one
a flavor of the koan-like nature of paradoxical com m unication in
therapy. It acknowledges that the therapist's primary “lever" for
change is his/her own personality (which is a totally different
view from the one in which the therapist follows a strategy or
format based on specific hypotheses form ulated earlier, as in the
Milan team 's approach). Additionally, this approach implies that
the therapist is in the situation or paradox with the client; it is
clearly not therapist as guru conceptualization.
The third approach is som ewhat similar to the second, al
though several distinctions are noted. This approach acknowl
edges (a) the archetypal nature of paradox, (b) the connectedness
of the paradox archetype with the striving for consciousness (in
therapy), which (c) leads to a confrontation with the paradox in
question and sim ultaneously to an awareness of the paradox in all
of life. This definition also makes a strong statem ent about the
intuitive nature of paradox in therapeutic com m unication. Q uot
ing one respondent, “There is a challenge in the statem ent, 'You
can 't get there from here.' To contrive to precipitate an awareness
of a paradox prematurely or as a technique is oppositional and
destructive to the patient." This approach also acknowledges the
fact that the therapist's personality is crucial to the therapy proc
16 CHAPTER 1
ess. The therapist might handle a therapeutic situation differently

on any day of the week, but it is clear that in this approach to a
definition, "th e therapist and client are in there struggling in the
same soup."
Additionally, the resolution of a paradoxical intervention re
quires a change of stance. The individual m ust search for solu
tions of a totally different nature from those previously tried and
they must come from the person's whole being. The paradoxical
intervention is not to be contrived. It evolves—from the shared
reality which therapist and client create together within the thera
peutic environment. The theme present in both the second and
third response is that paradox and interventions termed paradoxi
cal are not explainable in the realm of intellect. They surpass logic
and encompass one's whole being: they are not to be understood
through "th in king." These responses give credence to and appre
ciation for the deeper levels of existence involved in the "solving
of paradoxes." They also dem onstrate the relative absurdity of
trying to explain or define, with logic, phenom ena that exist in a
different realm from that of everyday conscious thought.
The latter two conceptualizations about paradox lend them
selves to a wide range of useful and potent linguistic form s—
myths, fables, poetry, and other similar mediums—within which
therapeutic com m unications may be framed.
The fourth approach is connected with system s theory, and,
more specifically, to the therapist's understanding of the way sys
tems work. Quoting one panel member, " I f the therapist has un
derstood the family system correctly, then what the therapist does
will work. This is true whether an observer would label what the
therapist uses as 'logical' or 'illogical,' 'straight-forw ard' or 'para
doxical'." Another way of saying the above is that an intervention
may appear paradoxical w hen one does not understand the sys
tem context and the "lo g ic" in which the particular intervention is
made: if one understands these, the particular intervention does
not seem paradoxical at all, but rather, well fitted to the therapist-
family relationship context.
A continuation of the theme which connects the concept of
paradox with systems thinking is a view that suggests a change in
the language used w hen referring to what have been previously
called paradoxical interventions. O ne panel member stated:
There exists an interpretation or instruction to the family that re

strains them from the change they have come to therapy to ask
help for, or from any change at all. Since this is done within a con
text that implies that the family's request will be honored, a contra
diction is set up. This contradiction is often called a "paradoxical

intervention." As the intervention only seems paradoxical or illogi
cal from the point of view of client expectations, the descriptive
term, "restraint from change," may be used to refer (more appro
priately) to the situation mentioned.
Telling a person/family to continue with a problem, and giving a
rationale that is validated by at least some of the family's percep
tions, is not "paradoxical." It is contrary to expectation. The ration
ale may be trivial or incomplete. Using such a trivial rationale to
restrain the family from change may seem absurd. But it is not par
adoxical because at the system level it makes sense from many
angles; it has justifications any one of which one might choose to
use as a rationale for restraining the family from change. To deter
mine which one, of course, is the essence of the art.
This system s-oriented view puts m uch more em phasis on the

quality of the therapist-family relationship.
Som e panel members em phasized the im portance of capturing
the circularity and/or process nature of paradoxical interventions.
Two examples of this em phasis follow.
1. The definition of paradox belongs to the world of logic. But
the term paradoxical intervention usually refers to reverse psychol
ogy, not necessarily to an intervention designed to bring about
second-order change (as previously defined by Watzlawick). O ne
need not succumb to the tem ptation to view paradoxical interven
tion as "w h at you use w hen the client is 'oppositional' to tip the
hom eostatic balance. That just makes the client into an object.
The key is in the process interface betw een therapist and client. It's
in that in-between. You can only create som ething at the inter
face."
2. "A paradoxical intervention is not a 'thing,' is not the
'change agent.' The change agent is im m anent within the process
which encom passes the 'fit' betw een the patterns of the family/
client (as described by the therapist to him self) and the pattern of
the intervention." [szc]
These general com m ents about the term "paradoxical interven
tio n " point out the circularity involved in conceptualizing about
paradox and change w hen one takes into account the existence of
system s—family system s as well as the therapist-family system
interface.
Som e of the more "co n crete" responses to the definitional
question fell into three general areas:
a. Paradoxical intervention as a term which specifies som ething

the therapist does.
18 CHAPTER 1
b. Paradoxical intervention as a term for defining the client's per

ception of an intervention.
c. Paradoxical intervention as a term used to describe the specific
syntax of certain com m unications.
There were several responses which are subsum ed under the

heading "som ething the therapist d oes." These will be listed
briefly to give the reader an idea of the variety of responses. A
paradoxical intervention can be seen as one in w hich the therapist:
Attempts to alter the frame of reference in w hich the problem is
conceptualized by the patient.
W ithin a context of the client com ing to therapy in order to
change, (the therapist) asks him or her to produce more of
the behavior that s/he w ants changed.
Prescribes the sym ptom or presenting problem , that is, the
therapist com m ands that the client do som ething s/he is al
ready doing. The paradoxical statem ent implicit is "D o what
you were doing, and it will be different." Process-w ise, doing
now what you were doing, because you are now being told to
do it by the therapist, is different.
Gives a global positive connotation of the behaviors of all par
ticipants in the particular context and history. S/he prescribes
the symptom, or defines the sym ptom as positive, and de
fines others' behavior as positive.
D oes som ething or says som ething that has an elem ent of sur
prise (i.e., is contrary to context-bound expectation) and re
solves a paradox by changing the epistem ological frame. If
the therapist can change the frame around, s/he can get a bit
of leverage which could not otherw ise be attained.
Com m unicates in a way that allows for two apparently contra
dictory m essages. Introducing a paradoxical intervention into
a rigid idea framework disrupts the rigidity and allows the
possibility of individuation and group creativity.
The second, more concrete approach to a definition em pha

sized the client's perception of the intervention as being the de
ciding factor about w hether it is paradoxical or not. If clients per
ceive the intervention as not fitting with their own com m onsense
views of how their problem should be solved, then the interven
tion is paradoxical. With this definition one may note a connec
tion with the more abstract statem ent that "paradox is in the
m ind of the beholder."
The third definition belongs on a level totally different from the
others. This particular form ulation forced us to think more than
any others (perhaps because it was so new). This definition is

more of a creative searching than a black-and-white here it is. It
states that "paradox is syntax of a certain type. In other words,
you can speak in paradox language or in ordinary English ju st like
you can speak in prose or poetry."
The rules "d eal m ostly with valuing and giving directions. In
stead of saying in prose 'That is bad. Do this instead/ in paradoxi
cal syntax you first assign levels and then say, 'It is good that that
is bad/ A similar form would be used for the com m and." [sic]
This definition places paradox on a plane with poetry, fables,
and m yths. Poetry is a useful analogy. People's responses to it are
as varied as are responses to the use of paradox. Som e view poe
try as nonsense, a waste of time, word magic, and cryptic com
m unication. O thers view it as sublime expression, language of the
gods, mythic, soul food, the language of free expression. W hat
m akes the difference? The language of poetry has been called the
language of the "u n consciou s." Poetry says w hat prose dares
not—what prose cannot, in some instances. So also with paradox
language. "T h e elem ents of surprise and non-coerciveness" allow
for the suspension of conscious thought and the em ergence of
"intuitive perspectives."
A well-known story illustrates the idea.2
"Laws do not make people better," said the Wise Man to the
King; "they must practice certain things in order to become at
tuned to inner truth. This form of truth resembles apparent truth
only slightly."
The King decided that he could, and would, make people ob
serve the truth. He could make them practice truthfulness.
His city was entered by a bridge. On this he built a gallows. The
following day, when the gates were opened at dawn, the Captain
of the Guard was stationed with a squad of troops to examine all
who entered.
An announcement was made: "Everyone will be questioned. If
he tells the truth, he will be allowed to enter. If he lies, he will be
hanged."
The Wise Man stepped forward.
"Where are you going?"
"I am on my way," said the Wise Man slowly, "to be hanged."
"We don't believe you!"
"Very well; if I have told a lie, hang me!"
"But if we hang you for lying, we will have made what you said
come true!"
"That's right: now you know what truth is—YOUR truth!"
(Shah, 1972, p. 23)
2© 1966, Mulla Nasrudin Enterprises Ltd. Used with permission.

20 CHAPTER 1
Round Four
The entire study, including results from three rounds of ques
tioning, was sent to the 26 panel members who had participated.
Each was asked to com m ent on the study and com m ents were
incorporated in Round Four. Responses were received from 23
persons. O f these 23, 17 responded with favorable com m ents
about the study process with remarks such as "w ell done," "c o n
gratulations," and "favorably im pressed." The remaining re
sponses were cards accepting acknowledgm ent for participation
in the study. Eight panel members sent articles or references to
articles they had written on related topics. Five participants con
tributed additional com m ents based on the entire study. O ne
panel m ember suggested adding three relevant experimental
studies (which he would send references for). A nother stated that
he was shocked by the negative argum ents of other participants
regarding ethics and paradox in therapy. In disagreeing with
them , he suggested that with many destructive and repetitive epi
sodes (such as truancy or excessive drinking) one framework
would be that of planning a "p rob ab le" relapse. For example, one
m ight say to the client, "W h en do you think the relapse will hap
pen this time? W ho will say what so that you will relapse again?
W hat will you be doing before you have your next relapse?" This
probable plan approach does not prescribe the destructive behav
iors; it calls attention to the patterns involved and the feelings
connected with such behaviors.
A third participant described her own experience of working
paradoxically. She stated, " I d o n 't know how I do it. Cases do not
becom e clear to me until I have a m etaphor for th em ." This expert
noted that paradox in therapy seemed very similar to drawing a
picture. Citing the book, Drawing on the R ight Side o f the Brain
(1979) by Betty Edwards, this respondent noted that w hat is nec
essary is to so confuse the left brain that it shuts off. For example,
w hen one draws a hand with the left brain one draws the symbol
of a hand. W hen the left brain is shut off and the right brain is
functioning (the right brain being able to flow freely with details),
the picture really looks like a hand and not merely a symbol. Sec
ond, this panel member agreed with com m ents made by another
participant—that the problem of glibness of attitude regarding
paradox does seem to exist. She noted that this seem s to be the
case with therapists in the learning stage, and less so with more
seasoned therapists. Last, this expert posed an im portant ques
tion: Can paradox really be taught? O r is it somehow absorbed
through experiencing of spiraling paradoxes within a supervisory

or therapeutic relationship?
O ne expert elegantly verbalized her experience of the whole
picture as follows:
I have been tilting crazily in and out of your finished product. This
"delphi" is fascinatingly circular in its process and seems to move
collectively along from Position 1 to Position 2, as earlier ideas ac
cumulate and push people on to new contexts from which to oper
ate. I myself am interested in tracing what I think is a genuine
movement from one “state of the art" sense to a newer one. The
process of facing folks with their old thoughts and feedbacks from
anonymous others about these thoughts, is in itself a statement
about how Mind works. It is like the Milan methods for generating
a hypothesis and their technique of circular questioning. I like it as
a refreshingly non-scientific instrument. Brad Keeney would call it
scientific in the sense of a "context for discovery" as opposed to a
"context for research."
Anyway, I would love to know who said what, but I guess that
would blow everyone's mind if you divulged your secrets. People
would find themselves matching up with their least favorite others,
if not their veritable enemies. And it would be even worse it they
didn't. So keep it all to yourself.
.. .My new slogan will be: Stamp Out Paradox, but the upshot
is to move me along in a conviction of a growing division in the
field, and how the two sides look. (Of course, as we shall soon
know, there are never real oppositions, but only apparent dual
isms, and imbrications of one side recursively into another, the Zig
a context for the Zag and vice versa. In each generation the old Zag
becomes the Zig for the next Zag. It's not better or worse, just how
things change.)
A fifth panel member expressed excitement at the diversity of
approaches, being amazed that the problem could be perceived
from so many different angles. O ne of the critics of paradox
roused this respondent's interest: he wondered if the critic was
som eone who h ad n 't worked much with families in which the
more usual situation is one in which one family member is a
"g o o d " patient, while the others are not interested in therapy.
This type of many-faceted psychosocial situation requires multi
level skills and attentiveness, or the family therapist can easily
burn out in two or three years.
In summary, Round Four feedback elicited responses to the en
tire study process and results. Highlights of these responses in
clude disagreement betw een two panel members regarding critics'
com m ents, an example of a probable plan approach, em phasis on
22 CHAPTER 1
intentionally confusing left brain functioning so it shuts off and

perm its the right brain's creative function to operate, a com pari
son of the process to the work of the M ilan team , and viewing the
Delphi process itself as a statem ent about how “ m in d " works.
RECOMMENDATIONS
There are several areas that are relevant to furthering the re
search on paradox in psychotherapy. A broader historical per
spective on paradox is needed. M uch has been written on the an
cient use of koans as ways fostering "enligh ten m en t." Western
therapists know little about the artful use of these koans and how
their process relates to paradox; an in-depth historical elucidation
is needed. Clearer understanding of the relationship betw een
conscious manipulation and hum an caring w ithin the therapeutic
relationship is urgently needed as is more in-depth study of the
short- versus long-term value of using paradoxes for change.
Phenomenological studies of clients who have experienced para
dox in their therapy relationships are needed; these could lend
invaluable insights into w hat is useful and m eaningful about
paradox in the therapy process. The use o f paradox in therapy
needs to be studied and m easured with respect to its potency and
usefulness. Carl W hitaker (1982), in a personal com m unication,
highlighted this issue with, "O n e of the great paradoxes is long,
continued therapy with the assum ption of change that is not tak
ing place. Ten years of therapy with no effective change is by itself
a powerful paradox" (personal communication). Last, the process-
research m ethod itself needs refining; its usefulness is becom ing
more apparent as more and more clinicians/researchers grapple
with the fluctuating elem ents involved in change.
CONCLUDING REMARKS
Several observations from a more general perspective are rele
vant. First, the m eanings inherent in the differing definitions are
im portant. It would be easy to assum e that all clinicians speak the
same language and that each has the same definitions of term s,
concepts, and m ethods. But such an assum ption would be wholly
unrealistic. People bring their own associations, ideas, memories
to a term or concept: so it is with the defining of paradoxical inter
vention. Granted this evolving process (associations, ideas, etc.)

has its base in som e original definitions of Watzlawick, Haley, and
Erickson. But each new " u s e r " of the concepts form s a new,
idiosyncratic, and in m any cases, creative understanding of that
concept.
This process of conceptualizing poses an interesting challenge
for com m unicators. Each has a specific definition or set of associa
tions to the words and each of these definitions is m eaningful and
useful. Nevertheless, in talking about paradoxical intervention, it
is im portant that each knows the other's definitions, grants that
there is no wrong or right one, and acknowledges that, to the
extent possible, agreem ent am ong com m unicators on w hat is
most useful in understanding the concepts is essential.
The variety of definitions of "paradoxical intervention" in this
study reveals the process of creative discovery within one particu
lar group of therapists/researchers. N one of the experts in this
study copied what som eone else had said before. Each seem ed
involved in sharing his/her own creative search.
To conclude, a quote from Erickson, Rossi, and Rossi's book,
Hypnotic Realities (1976, p. 63), seem s fitting:
Ideally, our therapeutic double binds are mild quandaries that pro
vide the patient with an opportunity for growth. These quandaries
are indirect hypnotic forms insofar as they tend to block or disrupt
the patient's habitual attitudes and frames of reference so that choice
is not easily made on a conscious, voluntary level. In this sense a
double bind may be operative whenever one's usual frames of refer
ence cannot cope and one is forced to another level of functioning.
Bateson (1975) has commented that this other level can be "a higher
level of abstraction which may be more wise, more psychotic, more
humorous, more religious, etc." We simply add that this other level
can also be more autonomous or involuntary in its functioning; that
is, outside the person's usual range of self-direction and -control.
Thus we find that the therapeutic double bind can lead one to experi
ence those altered states we characterize as trance so that previously
unrealized potentials may become manifest.
In actual practice, there is an infinite range of situations that may
or may not function as binds or double binds. What is or is not a
double bind will depend very much on how it is received by the
listener. What is a bind or double bind for one person may not be
one for another.. . . Humans are too complex and individual differ
ences are simply too great to expect that the same words or situa
tion will produce the same effect in everyone. Well-trained hypno
therapists have available many possible approaches to hypnotic ex
perience. They offer them one after another to the patient and care
24 CHAPTER 1
fully evaluate which actually lead to the desired result. In clinical

practice we can only determine what was or was not a therapeutic bind or
double bind in retrospect by studying the patient's response. [Emphasis
added.]
In my opinion, usefulness of double binds or paradoxical inter
ventions does depend on how each person and/or each family
perceives the bind. And another elem ent involves the therapist's
attitude toward strategic thinking. That is, what happens for the
therapist who decides to plan a certain strategy to help the client?
O ne important idea is that part of the power in the pradoxical
approach lies in the therapist's "thinking about" the system and
planning strategies based on his/her observations. This does not
imply that the therapist h asn 't "jo in e d " the family: It does imply
the therapist's recognition and use of a som ewhat paradoxical po
sition in the therapist-family system . S/he can join the family and
establish rapport and also recede to "th in k about" the system dy
namics and plan possible strategies. S/he can operate with flexi
bility in two important dim ensions of therapeutic functioning.
In summary, the therapist's ability to move in and partially out
of the system , the client's perception of the particular paradox,
and the quality of the therapist-client relationship operate as es
sential criteria for acceptance and/or use of paradox in therapeutic
communication.
REFERENCES
Bateson, G. Steps to an ecology of mind. New York: Ballantine, 1972.
Dalkey, N. The Delphi methods: An experimental study of group opinion.
Santa Monica, Calif.: Rand Corporation (RM-588-PR), 1969.
Dell, P. Paradox redux. Journal of Marriage and Family Therapy. 1981, 7.
Edwards, B. Drawing on the right side of the brain. Los Angeles: J. P.
Tarcher, 1979.
Erickson, M., Rossi, E., & Rossi, S. Hypnotic realities. New York:
Irvington Publishers, 1976.
Hoffman, L. Foundations of family therapy. New York: Basic Books, 1981.
Madanes, C. Strategic family therapy. San Francisco: Jossey-Bass, 1981.
Palazzoli, M., Bascolo, L., Cecchin, G., & Prata, G. Paradox and counter
paradox. New York: Jason Aronson, 1978.
Papp, P. The Greek chorus and other techniques of family therapy.
Family Process, 1980, 19.
Shah, I. The exploits of the incomparable Mulla Nasrudin. New York: E. P.

Dutton, 1972.
Tesch, R. Phenomenological and transformative research: What they are and
how to do them. Santa Barbara, Calif.: Fielding Institute, 1980.
Watzlawick, P., Weakland, ]., & Fisch, R. Change. New York: W. W.
Norton, 1974.
Weeks, G. & L'Abate, L. A compilation of paradoxical methods.
American Journal of Family Therapy, 1979, 7, 61-76.
Weeks, G., & L'Abate, L. Paradoxical psychotherapy. New York: Brunner/
Mazel, 1982.
Weiner, N. The human use of human beings: Cybernetics and society. New
York: Avon, 1967.
Whitaker, C. Personal communication. March 1982.
Williamson, D. Personal authority in family experience via termination
of the intergenerational hierarchical boundary: Part III—Personal
authority defined, and the power of play in the change process.
Journal of Marital and Family Therapy, 1982, 8, 309-322.
Part Two
Approaches to
Paradoxical Intervention
2
Unpredictability and Change:
A Holographic Metaphor*
by Brian W. Cade, CSW
"It's time for you to answer now," the Queen said, looking a t her
watch: "open your mouth a little wider when you speak, a nd always
say 'your Majesty.” '
"I only w anted to see what the garden was like, your Majesty—''
"That's right," said the Queen, patting her on the head, which
Alice d id n't like a t all: "though, when you say 'g a rd e n '—I've seen
gardens, com pared with which this would be a wilderness."
Alice d id n't dare to argue the point, but went on . .a n d I
thought I'd try and find my way to the top of that hill—"
"When you say 'hill,'" the Queen interrupted, ” 1 could show you
hills in comparison with which you'd call that a valley."
"No I shouldn't," said Alice, surprised into contradicting her a t
last: "a hill ca n 't be a valley, you know. That would be nonsense—"
The Red Queen shook her head. "You may call it 'nonsense' if you
like," she said, "but I've heard nonsense, com pared with which that
would be as sensible as a dictionary!"
Lewis Carroll, Through the Looking Glass
*1 wish to acknowledge the members of my team without whom none of this

would have taken place: Bebe Speed, Philippa Seligman, and Philip Kingston.
The chapter is dedicated to the late Dr. Harvey Jones whose influence brought me
into this field.
28
UNPREDICTABILITY A ND CHANGE: A HOLOGRAPHIC METAPHOR 29
HISTORICAL INFLUENCES
A colleague and I were once asked by Lynn Hoffm an how we had
arrived at our Marx Brothers style of therapy. M y answer was that I
was on the road to Damascus w hen a green book called Strategies o f
Psychotherapy fell out of the sky and hit m e. As is probably true of
many in our field, it is hard to overestimate the influence of Jay
Haley on my professional development, not the least of which was
his introduction to me of the work of M ilton H. Erickson (Haley,
1963, 1973). The other m ajor influences on my development have
been the work of the Brief Therapy Centre, M .R .I., California
(Watzlawick et al., 1967,1974; Weakland et al., 1974), and the work
of Mara Selvini Palazzoli and her colleagues at the then Centro per
lo Studio della Famiglia, Milan (Palazzoli et al., 1975,1978). During
recent years, I have also developed a growing fascination with the
indefinable and unpredictable work of Carl W hitaker (see Neil &
Kniskem , 1982).
M y early interpretations of Haley's work led me to construe
therapy as a kind of contest in which paradoxical techniques were
power tactics that could be used to prevent a client or family from
using symptomatic behaviour to control and define their relation
ships with me. Later, pardoxical work becam e more of an intellec
tual exercise, a kind of chess game. A ttem pts would be made,
meticulously, to construct interventions with all potential loop
holes blocked, thus forcing the client or family to move outside of
the rule-governed "sym ptom atic game without end, which up to
that moment, had no meta rules for the change of its own ru les"
(Watzlawick et al., 1967, p. 237).
At this time, m uch effort was devoted to trying to answer the
question, " Is this a true paradox?" and interventions were
measured against external yardsticks from the fields of m athem at
ics and logic. However, it becam e increasingly clear to my col
leagues and me that a central ingredient in our "paradoxical"
work was the unexpectedness of the position taken by the thera
pist (or therapy team) or of the interventions given. We began to
talk less and less about paradoxes (Is it? Isn 't it?) and more about
introducing difference or new information into system s through
unpredictability. As Palazzoli has declared, "P eo p le are most in
fluenced w hen they expect a certain m essage and receive instead
a m essage at a totally different le v e l.. . .anything predictable is
therapeutically inefficient" (1981, p. 45).
Interventions began to be designed and measured by yard
sticks more integral to the process of therapy. More time was
spent attempting to ascertain how the client or family were ex
30 CHAPTER 2
pecting, or apparently w anting us to be, and then taking a posi

tion, or making a response, including our interventions (the more
formal closing statem ents or prescriptions), that was isomorphic
enough to connect with the client's/family's expectations (see de
Shazer, 1982[a]). Yet, with sufficient, unexpected, som etim es ab
surd, even shocking elem ents com e difference. A s Erickson and
Rossi (1979, p. 5) observe, 'A n y experience of shock or surprise
momentarily fixates attention and interrupts the previous pattern
of association. Any experience of the unrealistic, the unusual, or
the fantastic, provides an opportunity for altered m odes of appre
hension." However, m any of the approaches, and interventions—
or aspects of the interventions—that we have developed will fit
the criteria for the definition of paradoxical, as elaborated, for ex
ample, by Weeks and L'Abate (1979, 1982) or Rohrbaugh et al.
(1977).
THE CLINICAL SETTING

The work to be described took place in a small family therapy
agency set up in 1971 in Cardiff, South Wales. This was the first
such agency in the United Kingdom. Individuals or families are
either self-referred or referred by the myriad of social, medical, or
psychiatric services in the area; they present with the whole range
of emotional, behavioural, or psychological problem s. The Family
Institute offers free services and is funded primarily by a m ajor
British charity, Dr. Bam ardo's.
As with m ost m ajor British cities, Cardiff is fairly cosmopolitan:
Referrals to the institute come from all social classes and reflect a
range of cultural strands. Breunlin et al. (1983) exam ined some
facets of British culture that have influenced our tendency toward
the more indirect approaches to therapy. The British tend to be
much more cautious about change, suspicious of "tech n ical" so
lutions to family and other social problem s, and pessimistic about
the potential benefits of talking as a way of approaching prob
lems.
The American tendency to "let it all hang out" and to form per
sonal relationships quickly is regarded by the British as a sign of
superficiality, transience, and egotism, and even as a breach of eti
quette. The British value "keeping oneself to oneself." When a
problem exists, it should be handled within the family, as it is im
portant "not to wash one's dirty laundry in public."
The British are far less direct about their interpersonal relation
ships, seldom making direct statements about how they function or
UNPREDICTABILITY AND CHANGE: A HOLOGRAPHIC METAPHOR 31
how they wish them to be different. They maintain a "proper" dis

tance when relating, and avoid strong displays of affect. (Breunlin
et al., 1983, p. 99)
Although the above represents a rather drastic oversimplifica
tion of British societal values, it can offer a useful backdrop against
w hich to view the attitudes of the average British family towards
the idea of psychotherapy. There will, of course, be differences
among various subcultures. Yet m ost families can generally be ex
pected to dem onstrate this backdrop of culutral injunctions against
the openness required in the process of psychotherapy.
Judicious care is taken not to imply premature familiarity. The
therapist avoids the suggestion that therapy may provoke rapid or
extensive change. . . . The therapist reassures the family that
change should not imply throwing out the baby with the bath wa
ter, that is, unwittingly removing qualities of life valued by the
family, or creating new problems by oversimple formulations and
hasty action Levels of intensity are adroitly monitored, because
of the societal emphasis on avoiding undue proximity, affect-laden
revelations, or statements that directly define or redefine relation
ships. (Breunlin et al., 1983, p. 100)
A TEAM APPROACH
M uch has been written about the use of paradoxical techniques
by a single therapist. This chapter will concentrate on a team ap
proach to therapy in which one m em ber acts as therapist while
the others observe from behind a one-w ay mirror. The observers
can intervene in a variety of ways: by telephoning in, calling the
therapist out, sending m essages via the therapist, and occasion
ally entering the room (see Breunlin & Cade, 1981; Speed et al.,
1982).
O ur use of team s evolved partly from our training programs,
through which we discovered the potential richness of live consul
tation as a way of increasing therapist maneuverability and stimu
lating creativity. The team approach was also particularly helpful
with those individuals and families adept at incorporating a thera
pist into their system s—of both thought and action—thus quickly
rendering him im potent. A family could achieve this end by rais
ing the therapist's level of anxiety; making him increasingly angry
or frustrated; making him care for or overprotect them ; making
him competitive (either with them or other involved profes
sionals, past or present); uncritically "co op eratin g " with him but
32 CHAPTER 2
never changing; seducing him into an interesting and stimulating

social relationship or intellectual exercise; inviting him to join the
"reasonable and m otivated" versus the "unreasonable and un
m otivated" factions within their ranks; or sometimes simply over
whelming him. With such families, the team attem pted to reduce
the predictability of the social transactions around therapy, thus
throwing the family off balance and in various ways avoiding
such pitfalls.
At first, this unpredictability was seen primarily as a way of
freeing the therapist, allowing him to get on with the therapy.
However, we increasingly found that the unpredictability itself
was a powerful promoter of change. It is the latter phenom enon
that this chapter will explore and elucidate.
Initially, the observers saw themselves as advisors, w hose role
was to help the therapist intervene to change the family as quickly
and effectively as possible. Gradually, we began to see more
clearly how a family and a therapist would construct, conjointly, a
"reality," in ways such as those elaborated above, which would
often inhibit the process of therapy. More and more, the observ
ing team found itself intervening on the family's and the thera
pists' system . Frequently, the therapists would resist the family's
inputs, sometimes consciously but more often quite unaware that
they were doing so (e.g., by adopting a posture, gestures, facial
expressions, or a tone of voice that subtly underm ined or contra
dicted the input).
We also found that the team tended to be drawn into a num ber
of families' "gam e without end," with the result that we would
begin feeling bogged down, frustrated, angry, overanxious, over
whelmed, and so forth. Som etim es, the team would squabble un-
productively, different members identifying strongly with differ
ent individuals or factions within the family. At other times, we
would becom e bored and "sw itch off," both emotionally and in
tellectually, leaving the therapist to sink or swim. (The latter
would mainly happen w hen the therapist seem ed stubbornly
aligned with the family and against the team .)
It soon became clear that ju st as the therapist could not be an
objective observer of the family, the team could not be an objec
tive observer of the therapy. The family, the therapist, and the
observing team were all involved in the process of constructing a
reality, which could, through the process of reification, inhibit the
process of change. As observed elsewhere:
Gradually the implications began to appear broader than in the ear

lier explorations with triangulation and paradox. The context of
therapy, through the medium of the screen, had become more com
plex again. Therapeutic change was increasingly conceived as a
phenomenon that grew out of the evolving nature of the relation
ship between therapist, team and family, and the "definition" of
each only "existed" in terms of each element's "ideas" about the
nature of the others and of the relationship between them. Varia
tion in any one element could bring about change in the other two.
(Cade & Cornwell, 1983, p. 78)
We began to experiment with changing the reality of the
therapy context. Thus it becam e much more difficult for families
to define the rules of the "gam e." Information, opinions, news of
events in the team, shifting alliances, criticisms or approval, and
so on could be transm itted to the family in a variety of ways. "As
with Alice's mirror in Through the Looking Glass, a new world of
possibilities, sometimes vivid, simetimes strange or illogical, had
unfolded. Though never invited to pass, like Alice, through the
mirror, such families could be offered glim pses which, com bined
with what each family m ember m ight project from his or her own
beliefs, could begin to disrupt their habitual patterns of thought
and behaviour" (Cade & Cornwell, 1983, p. 78). It was during
these team experiences that I found m yself concerned more with
the unexpected and unpredictable than with the "paradoxical,"
seeing the latter as one aspect or class of the former.
A META-THEORETICAL FRAMEWORK
Karl Pribram (1976) of the Stanford Medical School has pro
posed that the brain works along holographic principles. Before
considering the implications of this theory for therapy, I will
briefly describe holography.
In 1947 Denis Gabor invented a new photographic process for
which he received a Nobel prize in 1971. A photographic plate is
exposed to two sources of light, one reaching it directly from a
source, the other reflected off the object(s) to be photographed.
The resultant interference pattern caused by the m eeting of the
two light sources is recorded on the plate. W hen, subsequently,
the plate is exposed to a light source of equal intensity to, and
from exactly the same direction as, the original direct source, a
three-dim ensional image of the object(s) is "recovered" and ap
pears, floating in space in the position of the original object(s).
Unlike the normal photographic negative, there is no direct cor
respondence betw een the shape of the object(s) and the image on
the holographic plate, which appears as an apparently m eaning
34 CHAPTER 2
less pattern of swirls. Each part of the plate carries inform ation
about every part of the object(s) and therefore each part can be
used separately to form an image of the whole object(s) by shin
ing a light onto it (though it will lack som e of the sharpness of
detail that can be recovered from the whole plate).
A nother im portant feature of the holographic plate is that it can
be used to store a vast num ber of different im ages. Changing the
frequency of the light source and the direction of the beam will
allow a different image to be laid down and subsequently recov
ered by a recreation of the original conditions (i.e., frequency and
direction of beam ). Also, if the two light sources are reflected si
m ultaneously from two different objects (rather than one source
reaching the plate directly), the interference pattern laid down
will relate to both and reilluminating the plate with light reflected
from either one of the objects will recover the image of the other.
Each image will be perm anently "associated ” with the other.
Thus, the hologram represents a sophisticated inform ation stor
age system , second only to the hum an brain. Recreating the origi
nal conditions will recover any image or group of im ages.
The “Holographic Brain”

The hum an brain can be seen as sharing a num ber of im portant
features with the process of holography. First, it appears that each
memory may be distributed evenly throughout the brain rather
than localized in any one part. Earlier in this century, Lashley's
experiments showed that removing successively larger parts of
the brains of experimental animals, though impairing perform
ance, did not eradicate m emories. Russell has proposed (1979, p.
154), "Any one memory would be encoded as a pattern of chem i
cal changes over trillions of synapses—and possibly glia cells as
well—and each synapse would be involved in billions of different
m em ories."
A nother vital feature of the brain is its ability to associate expe
riences and to store patterns of association. Thus, access to part of
a pattern leads to an almost instantaneous com pletion of that pat
tern based on learned associations. As with a holographic plate,
but in a far more complex and sophisticated way, w hen the origi
nal conditions under which a m emory or group of memories were
laid down are recreated, the experience, or significant parts of it,
can be recovered. The process of association also m eans that with
access to a small part of the original conditions the brain can re
cover m uch or all of the m emory or group of memories.
In any given situation potentially limitless facets of the total
gestalt of externally and internally derived experiences exist that

can serve to recreate the original conditions. These, in turn, can
cause a whole range of m emories and associations to be recov
ered. W hich of these facets of experience becom e highlighted at
any particular time, given the huge backdrop of potential associa
tions, depends on the particular conscious (or less conscious) con
cerns occupying us at that point in tim e. A s W ellwood com m ents
(1982, p. 130), "A p plyin g a fram e to the implicit (i.e., the whole
gestalt) is som ew hat analogous to deblurring a blurred photo
graph by highlighting the m ajor contours or spatial frequencies,
so that particular shapes can em erge from the blur." In other
words, our particular concerns—the specific focuses derived from
our frameworks for applying m eaning—will, through intensifica
tion, separate out certain features or aspects of experience from
the implicit or potentially rich associations possible. As our pat
terns of association becom e established in a particular way, they
will tend to influence the processing of subsequent experiences.
As De Bono has suggested (1971, p. 124), "Patterns are picked out
of the environm ent solely on the basis of familiarity, and through
such selection becom e ever more familiar." In this way we develop
belief frameworks or mental " s e ts " that determ ine how we see
ourselves and our world, and how we ascribe m eaning, and thus
respond to, those experiences. In our relationships with others,
we then develop patterns of behaving together, that both reflect
our mental sets and those of the people with w hom we interact,
and tend by repetition to be confirm ed—though such patterns
rarely develop consciously.
Wellwood talks of the relationship betw een the conscious and
the unconscious as follows:
We are continually processing many kinds of interactions or inter
ference patterns and can only pay attention to a very few. The or-
ganismic processing that we do not attend to becomes part of an
unconscious background, holographic blur. This background blur,
which has an implicit structure to it, is surely what the concept of
the unconscious refers to. However, the traditional model of the
unconscious in depth psychology makes it appear as though the
unconscious has an explicit structure to it, as though drives,
wishes, repressions, or archetypes exist in explicit form, as though
the unconscious were a kind of autonomous alter-ego What is
unconscious are holistic patternings, which may be explicated in
many different ways and at many different levels of the organism/
environment interrelationship. (Wellwood, 1982, p. 133)
Thus, we can be seen as becoming like holographic plates on
which the same range of lights (i.e., same range of frequencies,
36 CHAPTER 2
directions, and intensities of beams) tend repeatedly to be shone,

thus recovering the same range of images and associated images and
leaving unrecovered a myriad of other possibilities. Those lights
shone with a greater intensity (because of our current concerns) will
lead to particular images and their associations will stand out more
sharply in a figure/ground relationship to the others.
Completing this metaphor, the process of therapy can be seen as
analagous to changing the lighting directed at a holographic plate by
(a) increasing the number of potential light sources, and (b) varying
their direction, frequency, and intensities, such that a greater range
of images and associations can be recovered and different facets
highlighted, allowing for different figure/ground relationships.
Health and Dysfunction

From this framework, health can be considered the availability
and possibility of complex potential association patterns, and of po
tential framings for interpreting the "reality" of experience that
leads to a wide range of possible affective and behavioral responses.
Dysfunction will thus be seen as the converse. Association pat
terns become limited and belief frameworks increasingly reified
and rigidified, allowing for ever more limited possibilities for in
terpreting reality, which lead to a more restricted range of affec
tive and behavioural responses.
However, the terms healthy and dysfunctional are not used here
as predicates. We do not argue that certain people or groups are
healthy or dysfunctional, but that in specific contexts and with
respect to particular functions, either pattern of thought and be
haviour can develop.
Sym ptom s are those affective, behavioural, or physiological re
sponses that develop and becom e maintained through the rigid
application, in any area of functioning, of reified frameworks for
defining and responding to reality. They are particularly likely to
develop or intensify where it becom es necessary for the individ
ual, family or other system significantly to change or adapt pat
terns of belief and response due to developmental or other
changes either internal or external to the system.
Because symptoms tend to becom e the focus of considerable
attention, they may serve to draw attention away from other areas
of distress and dysfunction. This phenom enon may lead to an ap
parent vested interest in sym ptom s being m aintained. Yet I think
it an error to assume that they thus have " a pu rpose" (though as
a way of framing certain interventions, this formulation has cer
tain uses). I regard symptoms as outcomes of certain patterns,
which subsequently may apparently come to serve a purpose.

Why, in any given situation, a particular class of sym ptom s oc
curs, and how the bearer is "selected ," usually depends on a wide
range of variables. O ne analogy is that, as with a balloon, the
symptom reflects a weak spot where, under pressure, the break
down occurs. Such weak spots can include tendencies towards
physical ill health in one member, potentials for stress between
particular m embers, one m ember's involvement in particular peer
groups (e.g., gangs of youth on the fringes of delinquency, drink
ing fraternities), tensions betw een families of origin, and many
others. In any gestalt, there may be tensions and potential diffi
culties at various levels of functioning, and an increased focus on
a particular issue that, at a particular point, stands out from the
others. These tensions can lead to the evolution of a "p rob lem "
surrounding that issue relative to which other issues remain in
the background.
This process of problem formation is described most eloquently
by Watzlawick et al. (1974). Wrong, or inappropriate solutions
(w hether acts of com m ission or omission), that derive from partic
ular belief frameworks applied to certain difficulties can result in
no change or an exacerbation of those difficulties. A problem may
then evolve as "m ore of the sam e" attempted solutions (or class
of solutions) are repeatedly or increasingly applied or intensified.
This can lead to more of the same problem, and so on.
Symptoms and Holograms

Sym ptom s, w hether seen as residing in the individual, the
family, or any other system , tend to be the focus of considerable
preoccupation by the sym ptom bearer, his or her intimates, and
often of other system s—legal, m edical, psychotherapeutic,
school, neighbourhood, work, and so forth. Such preoccupations
will consist of frameworks for thinking about the sym ptom and
its meanings, affective responses, and behaviours for dealing
with the sym ptom and its various effects. As the sym ptom b e
com es entrenched, one can see the presence of a holographic g es
talt in w hich certain beliefs, attitudes, and responses are contin
ually highlighted and repeated and thus reinforced. Through the
process of association, the highlighting of any one part of this
gestalt will tend to recover the other parts; as the process be
comes self-perpetuating, other patterns of thinking and respond
ing become less and less available.
In holography, if one keeps shining the same frequencies of
light at the same intensities and from the same direction, one will
38 CHAPTER 2
repeatedly recover the same im ages and associated images with

the same figure/ground relationships, even though the potential
range of possible images and associations on the plate may be
considerably wider. Similarly, and perhaps through a related
process, if the same range of attitudes and behaviours are re
peatedly focused on a problem and the processes that surround
it, the same interlocking pattern of responses will tend to be re
covered from those involved in the process and the sam e associa
tion patterns will be highlighted.
Therapeutic inputs can becom e part of such a process if the
therapist's orientation, both personal and professional, leads to a
particular way of thinking about and approaching a problem . The
more of the same approach or class of techniques, deriving from
the therapist's particular frameworks or m odels, can lead to more
of the same problem, and so on. It is as though s/he is shining the
same range of lights, as described above, onto the situation, thus
continually recovering the same im ages and associated images
with the same figure/ground relationships.
The following case example should clearly dem onstrate this
process. A thirteen-year-old girl was described by her father as
disobedient, uncouth, and out of control. The m an was in his
m idfifties and was described by his wife and by the therapist in
volved as rigid, Victorian in his ideas about discipline, and totally
resistant and lacking in motivation. The m an saw all psychiatrists
and social workers as "w orse than u seless." A rgum ents in the
family were frequent and the mother, who was in her late thirties,
identified with her daughter, frequently defending her from the
father's unreasonable attitudes and expectations. The therapist
considered the girl to be quite healthy, but driven by her father's
rigidity to act out her frustrations and rebel. She had been unable
to get the man either to " s e e " his contribution to this process or
to cooperate with her attem pts to persuade him to be less severe.
The therapy becam e deadlocked.
O ne can see how each m ember of this group acted out of beliefs
held about the others, and how those actions had the effect of
recovering specific responses from the others. Such a system
served to continue the "g a m e " and to confirm each m em ber's be
liefs.
The father's attem pts to control his daughter would lead imme
diately to her fighting back. His wife would "a tta c k " him and de
fend the girl, and the therapist would be brought in to try to get
him to "o w n " his responsibility and " s e e " that he needed to
change his ways.
The m other's attem pts to moderate her husband's views and
encourage her daughter's development would serve to make the

father angry with both of them , at which point he would try to
im pose further restrictions. The girl would fight back and the
therapist would be called in to "d eal with him ."
The girl's defiant attem pts to find some space for herself would
anger her father and her m other would have to defend her; the
therapist would again attem pt to get the m an to change his atti
tudes.
The therapist's efforts to have the father see that he was far too
severe would anger him and he would blame his daughter and
wife. The girl would fight back, the m other would defend, and so
on.
Although this familiar process can be viewed and "exp lain ed "
through a range of different frameworks, one can see how each
person's inputs into the process tended both to recover and be
recovered by the inputs of all the others. Thus, a reality had been
created which, through the process of reification, inhibited the
possibility of change.
A THEORY OF CHANGE:
PREDICTABILITY AND
UNPREDICTABILITY
O ne im portant feature of "stu ck n ess" in any context is the pre
dictability of responses on the part of the participants. Obviously,
a degree of predictability is im portant in any context. O nly thus
can we negotiate and establish the m any patterns that m ust go
into the form ing of any stable relationship or set of relationships.
However, adaptability to changing circum stances requires the
possibility of renegotiating reality, of responding in an unpredict
able way; that unpredictability m ust then lead to a renegotiation
of how we view ourselves, our relationships, and the m eaning of
experiences.
This process of creating "realities" can be represented in a cir
cular fashion as shown on the following page.
As with any circle, this process has no beginning or end. It
should be possible for change to be elicted by a significant variation
at any point in the process. O f course, w hat will represent a signifi
cant variation will change enorm ously from situation to situation,
as will the location in the circle of the point at which introduction of
a variation m ight be m ost influential. The im portant factors will be
(a) the nature and flexibility of the various attitudes and belief sy s-
40 CHAPTER 2
tem s being focused onto the situation, and (b) the num ber of differ
ent systems or themes that can be seen as having influence on, or
investment in, the continuation of how things have been. Any con
text can include variables involving repetitive patterns among sig
nificant participants that limit the ease with which an individual or
group can change behaviours and belief frameworks. Such limita
tions on change can occur where, for example, parental overin
volvement im pedes an adolescent's ability to differentiate; a heavy
investment in their immediate extended families makes it difficult
for a couple to negotiate an improved marital relationship; and a
family with children deemed "a t risk " remains under constant sur
veillance by relatives, neighbors or professional agencies, in spite
of its attempts to change. Feelings of im potence and inadequacy
are thus perpetuated—the same feelings that may have led to the
emergence of the original problem.
As suggested earlier, symptoms can be seen as arising from the
rigid application of reified frameworks for defining and respond
ing to reality. In such frameworks, there will be a high level of
predictability with respect to the ways participants focus upon
and respond to the problems. As shown in the earlier case exam
ple, therapy became stuck w hen the therapist becam e predictable
in his approach to the problem. Through a reciprocal process of
verbal and particularly nonverbal negotiations, a reality becom es

constructed that reflects the attitudes, beliefs, and responses of all
participants, but serves to inhibit change.
Trying to change a system while oneself remaining the same, in
terms of how one views a situation and thus responds to it, seem s
(if the difficulty does not becom e quickly resolved), only to re
cover more of the same responses from the other members of the
system (however correct or logical a position appears to be one's
own viewpoint). The implication here seem s to be that to change
or increase the adaptability of a system , it is necessary to be able
to change oneself and the position one takes—to becom e, as it
were, unpredictable. In terms of the holography metaphor, adapt
ability seem s to require the possibility of varying what is focused
(i.e., both ways of thinking and the actions that reflect and com
municate the thinking) onto any situation in order that new im
ages, associations, and figure/ground relationships can be recov
ered from a background that will already contain potentially limit
less possibilities for reconstructing realities.
For example, in the earlier case example, the approach taken
was to completely change the focus that had been adopted by the
previous therapist. I com m ented to the father on how difficult it is
to bring up children in these more permissive times, and on how
many of the older values of respect and self-discipline seem to
have been lost. I declared my belief that parents, not children,
should decide what is appropriate behaviour in their family and
that youngsters need the greater experience of their parents—
even though they may see them as old-fashioned. " O f course," I
ended, "good parents will obviously becom e more flexible and
negotiate more as their children grow up." The man's attitude
changed almost immediately. And after a short period of thought
fulness, he began to insist that his daughter was really "n o t a bad
kid," that perhaps he was being too hard on her and needed to
change his approach.
The man had doubtless come to the therapy anticipating a fur
ther attack on his position. Initially, he seated him self with his
back turned towards me; his counterargum ents had doubtless
been well rehearsed. Expecting an approach inconsistent with his
apparent beliefs and attitudes, he was met instead by an ap
proach that validated his position and he thus seem ed imme
diately able to accept the suggestion, "G o o d parents will ob
viously becom e more flexible and negotiate more as their children
grow u p." In the face of my continued expressions of caution and
reluctance to let him blam e him self, he increasingly generated his
42 CHAPTER 2
own arguments as to why he should becom e more tolerant. He

agreed to a short series of sessions: The outcom e of the case was a
considerable improvement in the situation.
The previous therapist had concentrated totally on trying to
change the man's position. I concentrated on changing my posi
tion, in order that new possibilities could be recovered from po
tentials that already existed in the family.
The main features of this intervention were:
1. The position I took was sufficiently isomorphic to that of the
father for him to feel validated and accept the suggestion that
good parents becom e more flexible.
2. The father's attitudes were given a different meaning and rela
beled as virtues.
3. I had taken a totally unexpected position in the father/therapist
process, a process that had becom e nodal in the previous
stuckness.
These three features—sufficient isomorphism, different m ean
ings, and the unexpected position on a nodal issue—are central to
this approach and will be expanded on in the following section.
Thinking differently about things does not appear to require
insight into what previously was "w ro n g ." In fact, as a thera
peutic approach, giving insight tends to focus attention back to
what was or is rather than to w hat m ight be, and thus tends to
recover responses that relate to past or present patterns rather
than future possibilities. M y own opinion is that where insight
seem s to work, it does so through the process of reframing or
giving new m eanings to experience, which leads to the client or
family seeing and experiencing things differently, in turn re
sponding differently, and so on.
It is interesting to compare this approach with that of Carl W hi
taker who never appears to directly try to influence families. He
seem s to concern him self primarily with his own growth during
the therapy experience and refuses to let the family define his po
sition on any issue. He behaves unpredictably in that he always
reflects back to the family a different and som etim es surprising or
absurd perspective on the nodal them es with which they are
struggling.
“Unpredictable” Therapy
Over the door of the Pasteur Institute in Paris is the inscription,
"L u ck Com es to the Prepared M ind." To be creatively but rele
vantly unpredictable requires that a num ber of areas be suffi
ciently assessed. It is not enough ju st to be unpredictable; it is

necessary that the unpredictability relates to nodal issues and
them es, both in the family and the therapy process. W ith increas
ing experience, more and more of the assessm ent process may be
made unconsciously. Therapists such as W hitaker can act from in
tuition largely because, with many years of experience to draw
on, they can trust that they are doing their basic processing, from
which their "in tu itio n s" em anate.
Having elicited clear descriptions of problem s in term s of actual
behaviours and events rather than predicates, and information
about how often, w hen, where, and in relation to w hom they
m ost occur, it is also im portant to discover w hat attem pts the cli
ent, family, or other involved parties have made to solve the prob
lem s. From this information it should be possible to begin making
assessm ents about the interlocking patterns of belief underpin
ning the attitudes and responses of the various participants—
those elem ents that perpetuate recovery of more of the same
problem behaviours. Sym ptom s can be seen as em bedded in a
hierarchy of them es and contexts and it is necessary to identify—
from the participant's various responses and from the words and
metaphors they use—the nodal issues and processes that may be
contributing m ost to the stuckness. Decisions about who and
what should be included in the therapist's frame of reference will
partly reflect them es and processes im m inent in the problem con
text, and partly the frameworks (individual, family, network, etc.)
from which the therapist functions and thus selects and orders
information.
It is also im portant to assess the position the client or family
(and sometimes the referring professional) seem s to expect, want,
or seem s to be m aneuvering the therapist to take, particularly
with respect to those nodal issues or processes or towards the
symptom. If there have been previous therapies, expectations will
obviously be based on those experiences, and a therapist can
sometimes be faced with considerable upset or anger if s/he does
not behave like a previous therapist, even though that therapy
may have been unsuccessful in solving the problem s.
In both the position that the therapist or team adopts and the
interventions offered, significant facets of the participant's beliefs,
attitudes, responses, or the sym ptom atic behaviour itself should
be reframed or relabeled, usually so that w hat previously has
been seen negatively is presented positively. New connections can
be made betw een particular aspects or processes, and different
meanings can lead to different experiences and responses.
For an intervention to impact usefully, it m ust be sufficiently
44 CHAPTER 2
isomorphic to the belief patterns and attitudes of the individual or

family or subgroup chosen as the focus of the intervention. That
is, there should be sufficient direct or analogic correspondence be
tween important facets of the position taken in the intervention
and nodal beliefs, them es, events, or structures in the family, in
dividual, or subgroup. If the intervention can make sufficient con
nections, both cognitively and affectively, that can be recognized,
identified with, and responded to, then the therapist and individ
ual or family can becom e engaged together, albeit briefly, in a
reality. These subsequent reframings and other unexpected as
pects of the intervention can have maximum effect in that they
evolve, as it were, from "fam iliar territory" or from perceived
similarity in beliefs and attitudes.
The intervention must then include an unexpected attitude, ex
planation, request, or injunction with respect to a nodal issue or
theme, so that a different light is focused (either directly or
through analogy) on an area of particular significance and pre
occupation for the client or family. This unexpected aspect need
not be dramatic or shocking; it can be gentle and relatively unpro
vocative.
It is im portant continually to monitor the feelings, attitudes,
and responses elicited from the therapist or team by the client's or
family's beliefs, attitudes, and responses to the therapy. These
will reflect the frameworks underpinning what the therapist or
team will give to the situation. Therapists can becom e quickly
committed to a particular diagnosis, especially once they have
made an emotional investment. The diagnosis can thus becom e
reified so that, even in the absence of change, the same thera
peutic approaches are applied and more of the same problem be
h avio u rs) are recovered.
For most of us, w hen therapy becom es stuck our training leads
us to look harder and harder at the individual or family. This ap
proach suggests that the opposite is as im portant: W hen stuck,
the therapist or team should look to the framework from which
the approach being used has evolved. It is often not enough to
appear to change position tactically while in reality remaining
committed to a particular diagnostic form ulation. It is arguably
impossible for therapists, over any extended period of time, to
disguise our basic diagnosis; it will inevitably be betrayed, albeit
at an unconscious level, via the m any nonverbal behaviours
through which information can be exchanged. This is especially
true where we hold a strongly negative or blaming view of a par
ticular individual or group.
The choice, at any point in therapy, of which behaviour or
theme, which individual or subgroup on which to focus an inter

vention is a matter of therapeutic judgm ent, as is the choice of the
unexpected com ponent, and the degree of attention paid to issues
of engagem ent and m aintenance. W hile standardized assessm ent
procedures can give useful guidelines, particularly about what po
sitions to avoid, it is also im portant that a therapist or a team
member be able to trust his/her intuitions. As Milton Erickson ob
served, " I always trust my unconscious. Now, too many psycho
therapists try to plan what thinking they will do instead of waiting
to see what the stimulus they receive is and then letting their un
conscious mind RESPOND to that stim ulus" (Gordon & M eyers-
Anderson, 1981, p. 17).
WHAT IS PARADOXICAL THERAPY?

Weeks and L'Abate (1982) refer to the problem of defining para
doxical interventions. They com m ent, "T h e definitions either
seem so broad and abstract as to be m eaningless or not broad
enough to include all the techniques asserted to be paradoxical in
nature" (p. 94). It may be that the problem lies in the attempt to
measure what happens in therapy against criteria from the fields
of mathematics and logic.
From this framework, paradoxical interventions can be seen as
one class of unpredictable positions or approaches that can be
taken by a therapist or a therapy team in w hich, given a client's or
family's expectations, a reversal of the usual approach to a sym p
tom or to related processes is proposed. This is not to deny that it
is possible to dem onstrate how certain techniques can answer the
formal definitions of paradox, nor to suggest that paradox cannot
be a useful m etaphor and that the various com pilations on para
doxical techniques do not offer valuable guidelines for making
therapeutic choices and decisions. My argument is that the holo
graphic metaphor can offer an alternative framework for consider
ing stuckness and the process of therapy and change based on a
theory of how the brain processes and responds to experience—
albeit a theory that has by no m eans found universal approval.
De Shazer has asserted, "A n y intervention, paradoxical or not,
can be seen as built on a m irror im age of the fam ily patterns; thus
it is erroneous to consider paradoxical interventions as som ething
in a class by them selves. This view perm its us to discard another
notion that has not been too useful, namely, that it is the thing
paradox that is the change ag ent" (1982 [b], p. 82).
It is easy to see how the best known and perhaps most com
46 CHAPTER 2
m on paradoxical technique, sym ptom prescription, involves, in

its various form s, a reversal of an expected position or approach.
W hether com pliance or defiance based approaches are being used
(Rohrbaugh et al., 1977; 1981), the client or family expects to be
helped to decrease or be rid of a sym ptom and is instructed or ad
vised to increase it. In this and other approaches, what usually is
perceived as negative and harmful can be framed as positive and
helpful, even vital. "S lo w dow n" can be recom m ended rather
than "im prove." The therapist is expected to be helpful and an
expert. Yet s/he may define the situation as hopeless or declare
im potence rather than encourage change. O r the therapist may
accept or exaggerate a position taken by the client or family on a
symptom or related processes.
It is not my intention here to elaborate on the various kinds of
paradoxical techniques or to translate them into this framework: If
necessary, the reader can do this quite easily for him - or herself.
Such compilations can be of great help, particularly for beginners
to an approach. My reservation, however, with lists of techniques,
together with indications and contraindications for their use, is
that they can becom e reified and lead to clients or families being
classified and 'fitted ' into preconceived categories. As Gordon
and M eyers-Anderson (1981) observe, "Very often the kind of
changes that a particular therapist will pursue with clients are
those that are consistent with the therapist's model of the world
(professional training and personal experiences) rather than a
function of, and related to, the client's model of the world. The
point is that our private and professional beliefs/standards/rules
do not encom pass w hat is possible, but instead LIMIT what is
possible." They go on to point out that this tendency " . . .is de
pendent upon the presuppositions that what is effective for one
person can be effective for another, that problem situations that
share a com m on name and experiential description are structur
ally isom orphic" (p. 12).
I have used unpredictable reversals of expected therapeutic po
sitions in a whole range of situations, including many that have
appeared in various listings of contraindications, and achieved
positive results. O ne such intervention with a chaotic and poten
tially violent family (see contraindications listed by Fisher et al.,
1981) will be described in the next section. A careful analysis of
any situation, with particular reference to the assessm ent fram e
work described earlier, should help ensure that interventions fit
with a client's or family's beliefs, anxieties, fears, and expecta
tions, and so on. As Erickson observed, " In dealing with people
you try not to fit them into your concept of what they should be
. . .you try to discover w hat their concept of them selves happens

to b e " (G ordon & M eyers-Anderson, 1981, p. 34).
Team Interventions
A num ber of papers have addressed the use of team s in ther
apy and have elaborated on a variety of techniques, m any based
on the transm ission of "ex p ert" opinions, or the utilization of the
potentials for triangulation inherent in the approach (Breunlin &
Cade, 1981; Cade, 1980; Cade & Cornwell, 1983; de Shazer,
1982[a], 1983; Hoffm an, 1981; Palazzoli et al., 1978; Papp, 1980;
Speed et al., 1982). In this section, I will concentrate on those in
terventions in which the team reports or dem onstrates changes in
its position or dynam ics, with the therapist usually adopting a
neutral position or, on some occasions, identifying with the cli
ent's or family's position. A reality is projected, as it were,
through the one-way mirror, which in part mirrors facets of a
client's/family's beliefs or processes, and often addresses aspects
of the therapy process.
A woman sought help with her fifteen-year-old daughter who
she said was out of control and an inveterate liar. The m other
claimed to hate the daughter who, in turn, claimed to hate her
mother. The only thing both could agree upon was that the
m other's cohabitant of some four years duration and a younger,
thirteen-year-old daughter were in no way involved in the prob
lem and should in no way be involved in the therapy
In the first interview, the couple fought bitterly and viciously
over every issue, often screaming at each other, leaving the thera
pist confused and, unable to steer the interview into a more con
structive channel, feeling totally im potent. The therapist vainly
sought some positives in the relationship, but to no avail: The
couple often ignored her attem pts to calm things down and to
seek further information.
After about one-half hour the team called the therapist out for a
consultation and evolved the following intervention. The therapist
returned looking angry and confused, slam ming the interview
room door behind her. She apologized for com ing back feeling
somewhat "ra tty " and reported that the consultation had been
m ost unhelpful because there had been a lot of disagreement
among the team.
Some of my team feel that I have not listened carefully enough
to you nor understood the depth of feeling going on between you,
and how deep-seated the feelings of hatred are. They feel I've been
trying to look for positive feelings between you where they, in fact,
48 CHAPTER 2
do not exist. My team has left me totally confused. One of my col

leagues, with a lot of experience working with potentially violent or
violent families, has told me that it is often surprisingly the two
members of a family who are, deep down, the closest to each other
who seem to get locked into battles such as you've described. He
says there's nothing you can do but go on struggling with each
other. That sounds very confusing to me but that is what they said.
The m other immediately asked w hether it would be helpful if
she gave a com plete history from the time of her daughter's birth
(something the therapist had tried to take earlier without suc
cess). M other and daughter then began to m use together over
how very close they had been through the many difficulties they
had shared, first as unsupported m other and child and then
bringing up the younger girl together. They smiled frequently at
each other and cooperated to give the therapist a comprehensive
picture of how good things were until four years ago, each build
ing on the other's contributions. As they discussed more recent
events, they again started to bicker. Immediately the team called
through to the therapist and advised her to look angry, to repeat
into the telephone that she had not the slightest idea of what the
team was talking about, and then to leave the room.
After a gap of some ten m inutes the therapist returned, again
looking angry and confused and slamming the door.
Therapist: I don't know what's got into my team today; they are totally
at sixes and sevens.
Mother: [With a wry smile] I think they need a bit of therapy.
Therapist: Perhaps they do; they're certainly giving me lots of mixed
messages and have succeeded in confusing me more. In fact I feel
that the most help I've had during the last hour or so has been from
you two, and not from them. Because we now have no time left, all I
can do is share with you my confused impressions of what they have
given me, and then arrange a further appointment. So I will be left
with my confusion and must try to clarify and make sense of things.
One colleague keeps on insisting that you two must have loved
each other very deeply to be behaving as you are now .. . . There's a
very strong feeling in the team, and this is confusing to me, that. . .
e r...th e y seem to have some catastrophic expectation that some
thing dreadful would happen if things changed. They seem to feel
that if you two were not having terrible fights, hating each other the
way you do, that things would go wrong for other members of the
family, perhaps between you [indicating the mother] and your younger
daughter, or between you and your man, or between you [indicating
the daughter] and your grandmother [with whom she was apparently
very close]. I need to go away and think about this and try to sort
things out. I'll see you both at your next appointment.
The therapist then quickly ushered the puzzled couple out of

the room, the woman saying she would discuss this with her co
habitant, the girl saying she would discuss it with her sister.
At the following session, four weeks later, all four members of
the family arrived and reported dramatic improvements. The
mother explained these changes as probably being the result of
the daughter's sudden decision about four weeks ago to join a
youth club and to take a Saturday job to supplem ent her pocket
money. The family in no way related the changes to the therapy
nor was any responsibility claimed by the therapist who remained
cautiously pleased for them that things seemed to have sorted
themselves out.
Two years later, a follow-up visit by a researcher found that the
improvements had been maintained, there had been no recur
rence of the problems betw een them , nor had any other problems
developed. The girl was now in full-time em ployment and her re
lationship with her m other was described as good. W hen asked
about the therapy team, the woman smiled and said that while
she felt they could doubtless have been of help to many families,
" I think we just had them completely confused." This highlights a
major problem with the follow up of cases in which such inter
ventions are used. The client or family is usually unaware that an
intervention has been made and changes are rarely attributed to
the therapy. Perhaps all that can be shown with certainty is that
sometimes, after many years of symptomatic behaviour, things
begin to improve "coincidentally" just after a particular kind of
intervention has been used.
A chaotic family with multiagency involvements, who had
been the subject of regular case conferences, was brought for a
series of consultations by their social worker. Sh e felt totally
stuck, knew she had becom e overinvolved, but was unable, be
cause of official concerns, to disengage herself.
The family consisted of Enid, a 33-year-old "depressed and
severely agoraphobic" m other with three children, William, 14,
Sally, 11, and Cristopher, 4. Sally had becom e a truant from
school and there were suspicions that she was also being sexually
promiscuous: In the sessions she certainly behaved in a provoca
tive manner. The m other had an older sister, Joanna, 35, an alcho-
holic who slept in rough derelict houses but regularly visited the
family, often extremely drunk: At such times, Enid would carefully
bath her and put her to bed. There was also a younger brother of
30 who was mentally subnormal, prone to violent outbursts, and
a patient in a local psychiatric hospital situated immediately be
hind the family's house. This brother would come over the wall
50 CHAPTER 2
nearly every night and often caused havoc. The family lived in
considerable fear that he m ight harm one of the children because
of his extreme jealousy of the attention they received from Enid.
The hospital claimed they could not restrain him because he was a
voluntary patient (though they sent a burly orderly with him for
the first two therapy sessions " in case he were to becom e vio
le n t"). A lodger also lived in the house, a 50-year-old subnormal
m an who was suspected of having some sexual contact with the
children, but was otherw ise seen as "qu iet and harm less."
The social worker had m anaged, thus far, to keep this chaotic
family together, despite considerable opinion that the children
should be removed from the house for their physical and moral
safety.
Though in the initial sessions reasonable contact was made
with the various members of this family, the therapy hour was
always chaotic. The noise level was high and increased dramati
cally whenever an intervention was attem pted, family members
often arguing together and occasionally striking out at each other.
The team learned gradually that Enid's older sister had been
very close to their father and that her brother had been their
mother's favorite, in spite of the trouble he caused her through his
aggressive behaviour. Enid had, in many ways, been the Cinderella
of the family; she had worked hard to care for them all, yet had
never been or felt recognized or loved. The father had died many
years earlier, at which point Joanna had begun drinking more and
more heavily. The m other had died two and one-half years earlier,
at which point Enid had becom e increasingly symptomatic.
The family's attitude towards therapy alternated rapidly be
tween anger that so many professionals were attem pting to inter
fere and tell them how to run their lives, and desperate appeals—
particularly by Enid—for som eone to help. The social worker
would receive frequent em ergency telephone calls from or about
the family: Upon arriving at their house, she would find them
resistant or puzzled as to why she was there.
At the end of the fourth consultation session with the family, I
entered the room after a period of consultation with my col
leagues. I looked distressed and reported that my team had had a
big argument. O ne side had asserted that the family members
were doing quite well and did not need the continual interference
of so m any social agencies. The other side felt that Enid particu
larly needed considerable help to carry out an im portant task she
had undertaken on behalf of the whole family since the death of
her mother. Enid had recognized how m uch the family needed
and m issed her m other (who herself had rarely gone out of the
house) and, even though Enid had to sacrifice m any aspects of

her own life, she had decided to becom e her m other— to stay at
home and always to be available w hen needed. Since this was an
extremely difficult task, Enid required the constant help and sup
port of the social worker to carry it out. Her brother and sister
and, particularly, daughter Sally were viewed by this side of the
team as sensitive to the whole family's needs, and their trouble
some behaviours were, in a way, a help in that they kept Enid
continually worried and depressed and thus unlikely to think
about going out and not be constantly available to the household.
Enid soon began to reject this idea and to say she did not want
the responsibility of being this m other figure. I agreed with her
and claimed I had tried to tell my team the same thing but they
had insisted that it was im portant, not only that she continue to
take her m other's place for her sister and brother, but that she
also assume a grandparent role for her children, all of whom had
been very close to and m issed their grandmother.
" I cannot step into my m other's shoes," Enid protested, "M y
m other was an old w om an!"
"T h at's what I told them ," I said, "b u t they insisted that it was
vital that you continue to fill these roles—m other to your brother
and sister and grandm other to your children."
Enid began to protest more vehemently, insisting that she
should be able to have a life of her own. The more Enid expressed
determ ination about not wanting to take on this responsibility,
the more I claimed that I had tried explaining this to my team, but
that they had persisted in em phasizing the im portance of her re
maining responsible, and that she thus needed the help of the
social worker to continue with this difficult but necessary task.
Finally, I promised Enid that I would try to sort things out with
my team. I stated repeatedly that I was feeling confused as to
what they expected and thus was unable to give any advice. I said
I would be in touch with the social worker if I succeeded in mak
ing any sense of w hat had been said. As she prepared to leave the
interviewing room, using the one-way mirror to adjust the head
scarf she was putting on, Enid, as though she suddenly recog
nized how wearing it the way she did made her look m uch older,
started to laugh and said, "W ell, I suppose I m ight as well put on
this old scarf."
A few m onths later, the social worker wrote to report that Enid
had becom e far more independent. She had undertaken her own
negotiations with the Social Security departm ent, was going out
much more and looking far less dowdy, and had coped with a
particularly trying period during which two of her children had
52 CHAPTER 2
been hospitalized. Though under great stress, Enid regularly vis

ited both children in their separate hospitals; one such visit re
quired a fifty-mile bus journey. The social worker wrote, " I have a
feeling that Enid's newfound independence is increasing her con
fidence and she seem s at the m om ent to be making all positive
attempts to resolve her own situation." Enid had written to both
her member of parliament and a local counselor about the psy
chiatric hospital's unhelpful attitude toward the family's fears
about her brother's disruptive behaviour and potential for vio
lence. The social worker was careful to point out to me that her
"on ly involvement in this was to supply her [Enid] with a pen."
In another case, a couple was referred by their general practi
tioner for marital therapy. The husband was sim ultaneously re
ceiving outpatient treatment for "schizophrenia," following a se
ries of psychotic breakdowns over the previous ten years during
each of which he had been hospitalized and treated primarily
with drugs.
In an early session, the team had framed his illness as his at
tempt, on behalf of his family of origin, to keep alive the memory
of a sister who had died several years before his birth. The man
had not only been born on his sister's birthday but at the same
time of day, a coincidence that had always been seen by his par
ents as somehow having special significance. As far back as he
could remember, the man's birthday had always been, for his
family, a day of mourning.
This framing produced an initial improvement but, during the
fourth session, the man began to exhibit many of the symptoms
described earlier as the first signs of an impending and usually
rapid breakdown. He was becom ing increasingly preoccupied
with the meaning of life as revealed to him during his previous
psychotic episodes. He was excitedly sharing these revelations
with the therapist in an increasingly agitated manner, talking vol
ubly over her frustrated attem pts to seek further information
about the couple's wider family context.
As the man's thought processes became more and more disor
dered, his wife, a nurse by profession, looked on with calm con
cern. Finally, with a look of anguish, he said despairingly that
anybody who is not a schizophrenic could never understand what
he was trying to com m unicate. Shortly after this interaction, the
therapist came out for a consultation: The man continued talking
excitedly to his wife.
The therapist was advised to return to the session with a look
of agitated concern and report that one of her colleagues had ju st
begun to understand what the man had been saying and had be
come quite disturbed. The therapist was further advised to say

that her colleague felt it extremely im portant that she ask some
questions concerning the couple's wider family context. The man
becam e instantly calm and for the rest of the session cooperated
in answering all of the therapist's questions without any further
sign of his "d istu rbed " behaviour.
During this second part of the session it was ascertained that in
the man's family of origin and his wife's family, there were rela
tionship problem s about which he often worried. During his
breakdowns each family would visit him and express caring and
concern. W hen "sa n e," however, both families rejected him, and
any expressions of concern by him for their various problems
would lead to angry denials and claims that his worries stemmed
from his psychiatric condition, a claim echoed by his consultant
psychiatrist.
At the end of the session, the following intervention was de
vised. The therapist returned, again looking concerned, and re
ported that the colleague w ho had begun to understand w hat the
man had been expressing had, during the consultation, attem pted
to convince the rest of the team of the im portance of the man
remaining unsuccessful and a psychiatric patient. He had warned
them that, if the man let go of his dead sister's m emory or became
successful in leading a normal family life, serious problem s could
possibly arise in either his or his wife's family. Sh e further re
ported that the more this colleague tried to convince the team, the
more they becam e upset and angry and rejected his warnings.
The colleague becam e increasingly agitated as he continued
trying to convey his worries to the team, who in turn becam e in
creasingly angry with him, defining his ideas as crazy. "At this
point," said the therapist, " I called the consultation to a halt, and
so I can only leave you with this confused account of what has
happened in my team and I will see your in four weeks tim e." The
couple politely thanked the therapist and left without comment.
In the next session, the couple reported considerable improve
m ents in both the man's state of mind and their relationship.
Therapy was discontinued after two more sessions. Two years la
ter the couple's family practitioner reported continued improve
m ents. The man had been working, and would call into the sur
gery occasionally for a chat and som etim es for a mild tranquilizer
to help him through a particular period of stress.
In all three examples of unexpected team interventions, the
reality presented as existing behind the one-way mirror had repli
cated and evoked, analogically, nodal issues or struggles in the
families or the process of therapy. These realities were based on
54 CHAPTER 2
the team 's observations of how family m embers repeatedly dealt

with each other and the therapy, and its hypotheses about the
underlying rules and beliefs through which family m em bers inter
preted their experiences. The therapist then expressed and acted
out the feelings that had been evoked in him or her by the way(s)
in which the family had thus far been dealing with the therapy.
As suggested earlier, families can often evoke strong responses
in a therapist by their behaviours: O ften more of the same behav
iours by family members will recover more of the same responses
from the therapist and vice versa. These responses, however, m ust
in no way be blamed on the family but rather on the team 's internal
struggles, and no overt or impled linkup should be made for the
family betw een these realities and their own attitudes and behav
iours, which must in no way be negatively connoted.
Through reframing or relabeling, new m eanings were placed
on nodal issues. The therapists, usually by rem aining neutral
with respect to the team 's interventions, or som etim es by identi
fying with the family's or a family m em ber's position, stepped out
of the framework that had been eliciting the more of the same
responses. The sorting-out of the issues evoked was presented as
the responsibility of the therapists and their team ; and it was im
plied that, for the time being, therapy would com e to a halt until
these issues had been resolved.
As I have com m ented elsewhere:
At the following session, regardless of how effective the interven

tion appears to have been or not to have been, it is usually best to
make no mention of the "struggle" unless specifically asked. In the
author's experience, it is rare for a family to inquire. If, however, an
inquiry is made, it is best that the "struggle" be reported as not
having been resolved, even though the "feelings" may, if neces
sary, be reported as having diminished or settled. This is because
to resolve the "struggle" in favour of one side is likely, by implica
tion, to negatively connote the other. (Cade, 1980, p. 261)
Interestingly, as highlighted at the follow-up interview with the

m other in the first example, the families will be unaware that any
therapeutic intervention has been made; thus, even though con
siderable change may occur following their use, family members
can only assum e they have brought about such change them
selves. At follow-up interviews, w hen asked about their experi
ences of the team behind the one-way mirror, families have never,
in our experience, "rem em bered " such interventions.
These interventions, we have found, tend only to be effective
where an im passe has been reached in the therapy. W hen we
have used them as techniques to shortcut the process of therapy,
they have had little effect. It is as though the potency of the inter
ventions arises out of the experiences of deadlock from which the
team derives the necessary, and perhaps often unconscious, infor
mation for their construction.
Som e of the basic "u n exp ected " m aneuvers upon w hich such
interventions can be devised and built have been elaborated else
where (Cade & Cornwell, 1985). Briefly, they include:
a. The criticisms by the team of som e aspect(s) of the therapist's
behaviours, and/or apparent beliefs and attitudes that can be
seen as underpinning his/her approach to the family. Such
criticisms will reflect and express directly, or by analogy, as
pects of the family's more resistant responses to and apparent
feelings about the therapy. Thus, development of a symmetri
cal struggle betw een family and therapist can be diverted or
blocked and the interview often moved in a more constructive
direction, the family's position having been understood and
also powerfully expressed by the team.
b. The reporting by the therapist of opinions, advice, cautions,
and so on sent through by the team about which s/he remains
neutral, by claiming, for example, incom prehension, puzzle
m ent, or total confusion. The therapist can, in this way, avoid
being pulled into a struggle over the content of an interven
tion. S/he can claim to be unable to explain the m essage, yet
remain sym pathetic to the fam ily's struggles to understand
and deal with what has been said.
c. A declaration of im potence by the therapist or by the team.
No blame m ust be attached—either directly or by implica
tion—to family m em bers, and no hint of challenge, disap
proval, or sarcasm m ust be betrayed.
M aneuvers such as these tend to reverse the flow of a session
in which the therapist's previous approach has been recovering
increasing levels of resistance to the incorporation of new infor
mation on the part of the family or particular family m embers.
ETHICS
There will always be critics for w hom such techniques will be
seen as manipulative, controlling, and dishonest, particularly the
presentation of m anufactured realities, as described above. How
ever, this framework raises an interesting ethical issue. If stuckness
in families and therapy can be seen, in part, as a function of a
reified diagnostic framework repeatedly focused on the situation,
so that the same associations and responses continue to be recov
56 CHAPTER 2
ered, is it ethical to adhere to that framework in the face of no

change occurring, or even to find a way of "b lam in g " the family by
defining them as lacking in insight, resistant, or untreatable?
I will always remember the case of a forty-five-year-old man
who had been told by a psychiatrist that no help was possible at
his age because he would need five years of intensive analysis
before it would be possible even to begin addressing his sexual
problem s. N o doubt th e psychiatrist concerned was being
"h o n e st" from his framework. Unfortunately, this honesty pre
vented the man from seeking further help. He had believed the
diagnosis and it had becom e part of his own belief framework that
he was untreatable.
I am in agreement with the view of Watzlawick et al. (1974) and
Haley (1976) that it is impossible not to m anipulate. The question
can never be w hether or not to manipulate, but rather how m uch
and in what way will be most helpful in any particular case. The
myth that it is possible to avoid m anipulation is at best a utopian
self-deception; at worst, and particularly w hen coupled with the
belief that there is a correct view of reality, it can add considerably
to the distress of individuals and families as they are forced,
sometimes subtly, sometimes through the power of medical or le
gal authority, to accept the therapist's view of reality, especially
w hen defined as unmotivated, resistant, defiant, untreatable, in
adequate, and so on.
M y view is that it is possible to see these interventions as intro
ducing, through analogy, a range of alternative perspectives that
will refer to more significant truths for the family in that they are
usually framed with respect to underlying and often denied
themes and struggles. If there is deceit, it is only in the trappings
surrounding the transm ission of the essential framework of ideas.
The effectiveness of any intervention will depend on the validity
of the offered perspectives and their relevance to the individual or
family, rather than on the means by which they are introduced.
Weeks and L'Abate (1982) have highlighted the im portance of
not using such techniques as gimmicks, or out of frustration
w hen therapy seem s flat or families uncooperative. They under
line the importance of making responsible decisions based not
only on intuition but also on careful analysis of each case.
CONCLUSION
Although, throughout this chapter, I have been using principles
from holography as a metaphor to help consider the processes of
stuckness and ch an ge, there are those, particularly Karl Pribram

(1976), w ho believe that our brains do work on such principles.
A recent paper in New Scientist (Zuccarelli, 1983) has proposed
that, in hearing, the ear not only receives sound but also trans
m its it and then analyzes the resulting interference pattern or
"acou stic hologram ." Som e scientists, notably the physicist David
Bohm (1980), have gone further and proposed that the w hole u ni
verse is constructed on holographic principles.
REFERENCES
Bohm, D., Wholeness and the implicate order. London: Routledge &
Kegan Paul, 1980.
Breunlin, D. C., & Cade, B. W. Intervening in family systems using
observer messages. Journal of Marital and Family Therapy, 1981, 7,
453-460.
Breunlin, D. C., Cornwell, M., & Cade, B. W. International trade in
family therapy: Parallels between societal and therapeutic values. In
C. J. Falico’v (Ed.), Cultural perspectives in family therapy. M d.: Aspen
Systems Corp. 1983.
Cade, B. W. Resolving therapeutic deadlocks using a contrived team
conflict. International Journal of Family Therapy, 1980, 2, 253-262.
Cade, B. W., & Cornwell, M. The evolution of the one-way screen. The
Australian Journal of Family Therapy, 1983, 4, 73-80.
Cade, B. W., & Cornwell, M. New realities for old: Some uses of teams
and one-way screens in therapy. In D. Campbell & R. Draper
(Eds.), Applications of systemic family therapy. London: Academic
Press, 1984.
de Bono, E. The mechanism of mind. Middlesex, Eng.: Pelican Books,
1971.
de Shazer, S. Patterns o f brief family therapy: An ecosystemic approach.
New York: Guilford Press, 1982. (a)
de Shazer, S. Som e conceptual distinctions are more useful than
others. Family Process, 1982, 21, 71-84. (b)
de Shazer, S. Som e bonuses of using a team approach to family
therapy. In L. R. Wolberg & M. L. Aronson (Eds.), Group and
Family Therapy, 1982. New York: Brunner/Mazel, 1983.
Erickson, M. H ., & Rossi, E. L. Hypnotherapy: An exploratory casebook.
New York: Irvington Publishers, 1979.
Fisher, L., Anderson, A ., & Jones, J. E. Types of paradoxical
intervention and indications and contraindications for use in clinical
practice. Fam ily Process, 1981, 20, 2 5 -3 5 .
58 CHAPTER 2
Gordon, D., & Meyers-Anderson, M. Phoenix: Therapeutic patterns of

Milton H. Erickson. California: Meta Publications, 1981.
Haley, J. Strategies of psychotherapy. New York: Grune & Stratton, 1963.
Haley, J. Uncommon therapy: The psychiatric techniques of Milton H.
Erickson, M.D., New York: W. W. Norton, 1973.
Haley, J. Problem solving therapy. New York: Jossey-Bass, 1976.
Hoffman, L. Foundations of family therapy. New York: Basic Books, 1981.
Neil, J. R., & Kniskern, D. P. From psyche to system: The evolving therapy
of Carl Whitaker. New York: Guilford Press, 1982.
Palazzoli, M. S., Boscolo, L., Cecchin, G., & Prata, G. Paradox and
counter-paradox: A new model for the therapy of the family in
schizophrenic transition. Paper presented to the 5th International
Symposium on Psychotherapy of Schizophrenia, Oslo, 1975.
Palazzoli, M. S., Boscolo, L., Cecchin, G., & Prata, G. Paradox and
counter-paradox. New York: Jason Aronson, 1978.
Palazzoli, M. S. Comments on Paul Dell's Some irreverent thoughts on
paradox. Family Process, 1981, 20, 37-51.
Family Process, 1930, 19, 45-58.
Pribram, K. Consciousness and the brain. New York: Plenum, 1976.
Rohrbaugh, M., Tennen, H., Press, S., White, L., Raskin, P., &
Pickering, M. Paradoxical Strategies in Psychotherapy. Paper presented
at the American Psychological Association, San Francisco, Calif.
1977.
Rohrbaugh, M., Tennen, H., Press, S., & White, L. Compliance,
defiance and therapeutic paradox. American Journal of
Orthopsychiatry, 1981, 51, 454-467.
Russell, P. The brain book. London: Routledge & Kegan Paul, 1979.
Speed, B., Seligman, P., Kingston, P., & Cade, B. W. A team approach
to therapy. The Journal of Family Therapy, 1982, 4, 271-284.
Watzlawick, P., Beavin, J., & Jackson, D. D., Pragmatics of human
communication. New York: W. W. Norton, 1967.
Watzlawick, P., Weakland, J. H., & Fisch, R. Change: Principles of
problem formation and problem resolution. New York: W. W. Norton,
1974.
Weakland, J. H., Fisch, R., Watzlawick, P., & Bodin, A. Brief therapy:
Focused problem resolution. Family Process, 1974, 13, 141-168.
Weeks, G., & L'Abate, L. A compilation of paradoxical methods,
Weeks, G., & L'Abate, L. Paradoxical psychotherapy: Theory and practice
with individuals, couples and families. New York: Brunner/Mazel,
1982.
Wellwood, J. The holographic paradigm and the structure of

experience. In K. Wilber (Ed.), The Holographic Paradigm: and Other
Paradoxes, Boulder, Colo.: Shambhala, 1982.
Zuccarelli, H. Ears hear by making sounds, New Scientist, 1983, 100,
(1383), 438-440.
3
Beyond Paradox and
Counterparadox
by Klaus G. Deissler. Diplompsychologe
Now this going with the wind or the current, plus the intelligence
pattern of the human organism, is the whole art of sailing—of
keeping wind in your sails while tacking in a contrary direction.
Alan Watts
FOREWORD
M uch has been thought and written about paradoxes—indeed a
whole type of therapy has been named after it. In my opinion, so
much has now been written about therapeutic paradoxes that I
get the feeling I am dealing with a "gnaw ed and old b o n e" about
the state of which there is very little left to write. The m ost impor
tant articles and books have already been written (for detailed bib
liographical notes, see Weeks & L'Abate, 1982).
The question is then: W hat is there left to say?
I. HISTORICAL INFLUENCES
A. People Who Have Directly Influenced

My Thinking
W hen I maintain that paradoxical com m unication is an every
day phenom enon (Sluzki & Ransom , 1976), this in itself says just
60
BEYOND PARADOX A N D CO UNTER PARADOX 61
as much or as little as M ilton Erickson's statem ent that trance is an

everyday phenom enon. It is only w hen we come to define such
statem ents more closely that it becom es possible to draw conclu
sions about their usefulness (on trance as an everyday phenom e
non, see, e.g ., Erickson & Rossi, 1979).
I was first introduced to paradoxes by my maternal grand
father. He was regarded as a kind man in our family, but he could
also do mysterious things, for example, take away the sting from
bum s. It was this grandfather, too, who would offer my sisters
and me one deutsche mark if we had hiccups and could deliber
ately produce at least 10 repetitions. O ur efforts were usually in
vain.
O n other occasions he gave us "rid d les": he would ask, "W hat
is it? It hangs on the wall and goes tick-tock, and if it falls down
the clock is broken." The answer to this riddle is so obviously im
plied—it is a clock. W hat I did not notice as a child, though, was
that my grandfather created an interrogatory attitude or soft con
fusion. The answer to the riddle is so obvious that the person
trying to solve it has to ask him self subconsciously, "W hat's the
point of this silly riddle?" or "W hat answer does this person want
by asking a question such as th at?" The trick in this case is to
evoke an inquisitive or searching attitude. O f course, as a child I
missed the point of the next suggestion from my grandfather. For
example, he would say, "C om e on, finish your dinner, we want to
go to tow n." However, it gets worse: The paradox with which he
confronted me that influenced me most was the fact that he could
make me inquisitive about all the mysterious acts he could per
form. W hen as a teenager I asked him to explain, for example,
how he could take away the pain from burns, he would not give
me an answer. This had a tremendous effect on me: I becam e even
more inquisitive and, disappointed, I resolved to find my own an
swers. In any case, it was my grandfather who aroused my in
terest in indirect interpersonal influence, and who more or less
directly suggested that I go into psychology and later study in
direct suggestion. I had an "ah a experience" w hen I read Haley's
book on Erickson in 1974 (Haley, 1973).
B. Works Which Have Influenced My Thinking

I had long forgotten about the stories with my grandfather.
However, after I obtained my degree in psychology in 1974, hav
ing found the course rather boring, I realized I was dissatisfied
with what I had learned up to then. Toward the end of my degree
work I became acquainted with an older doctor from whom I was
62 CHAPTER 3
learning autogenic training (see Schultz, 1966; Thom as, 1972).

This doctor apparently also m astered other hypnotherapeutic
techniques, but was not willing to teach them to me as he wanted
to reserve them for other medical doctors. M y further attem pts at
finding out about hypnotherapeutic m ethods in Germ an-speak-
ing countries were rather disappointing: The procedures were
authoritarian; they had the hallmark of a "w hite-coat m entality";
and they were heavily ritualized. W hat I was looking for were
indirect, more flexible procedures that stood out through their
artistry.
Just before I took my degree I read Watzlawick's classic, Prag
matics o f Human Communication (Watzlawick, Beavin, & Jackson,
1967). livis book was for me the starting point for intensive studies
that went in two particular directions: (1) the practice of psycho
therapy—effect patterns that circulate in therapeutic com m unica
tion; and (2) the theory of hum an system s—higher connections or
process patterns that are able to provide the instrum ents of analy
sis in therapeutic processes.
I came across a further puzzle, or "k o an ," w hen dealing with
pathogenic and also constructive or creative double binds (see,
e .g ., Haley, 1963; Bateson, 1969; Berger, 1978; Erickson & Rossi,
1975). This "k o a n " concerned me until about 1979, the year in
which I met both Mara Selvini and her team , and Gregory Bate
son. At that point in my therapeutic work, I becam e a systemic
family therapist; at the same time, my work has been com m itted
to Ericksonian hypnotherapy since 1976. The interim end o f my
therapeutic development consisted of com bining Erickson's prac
tical genius with Bateson's theoretical genius: Perhaps we can call
the result "contextual psychotherapy."
II. PARADOXICAL THERAPY-A CONCISE

TERM FOR SELF-REFLEXIVE
THERAPEUTIC CONFUSION?
A. The Double Bind Theory

The double bind theory was originally conceived by Bateson et
al. (1956) as a way of freeing the term schizophrenia of its biological
and intrapsychic implications and finding an explanatory model
for schizophrenic behavior with a more social base. W hether this
function has been fulfilled is open to discussion. In psychother
apy, however, the double bind theory did turn out to be a
BEYOND PARADOX A N D COUNTERPARADO X 63
"branching criterion" or parting of the ways for the therapeutic

mind. Som e therapists dealt with the double bind theory without
being able to m aster it; they rem ained loyal to their old m echanis
tic epistemology. O thers left this theory behind and turned to sys
temic and cybernetic thinking and action models (see Bateson,
1969). The m ost influential development and adaption of Bate
son's ideas were carried out in M ilan (see Selvini-Palazzoli, 1978).
However, let us stay for the m om ent with the double bind the
ory. Watzlawick et al. (1967) and Erickson & Rossi (1975) in partic
ular used the double bind theory explicitly for therapeutic pur
poses by constructing situations that helped their clients out of
the context of negative alternatives and put them in the position
of not being able to lose. I need not stress the fact that construc
tion of such double binds requires practice, skill, and precise
knowledge of the client's problem s. If a therapist m anages to per
suade a couple who have not been on holiday for 20 years that
they have a choice of going to Spain or the Alps, this is a classical
form of double bind—even if the couple finally decides to go to
Italy.
B. Criticism of the Double Bind Theory: “Paradox”

Discussion about expressions, concepts, and theories in psy
chotherapy are often, in m y opinion, hairsplitting or a "w ar with
w ords" (see Shands, 1971). An example of this war with words
was, for me, the discussion of Dell's (Dell, 1982) and Keeney's
(Keeney, 1982) articles in the journal, Family Process. The attempt
to bring about conceptual clarity only results in more confusion
than there was before. (In principle, articles like these are relation
ship offers or definitions them selves. Perhaps a few clear state
m ents offer clarity, w hich may be used as suggestions by some
readers (see Erickson, 1964). In the light of these remarks what
follows should not be taken too literally or seriously.
1. The Need to Differentiate. W hen hum an beings are not in a

state of meditation, trance, or confusion, they are dependent on
making decisions in order to com e to term s with their world.
Among the m ost im portant differentiations they m ust make are
figure and ground, text and context, and elem ent and class. These
differentiations can be scrambled or resigned; for example, figure
and class, text and ground, and elem ent and context. Som etim es
such reassignm ents are useful; however, they change nothing of
the basic need to make differentiations. A s is implicit in the above
considerations, one of the m ost im portant possibilities for making
64 CHAPTER 3
differentiations is to define units that are parts of higher units. O f

course paradox can be defined as the situation in which a unit is a
part of itself. Bateson has formulated this thought in various con
nections following W hitehead and Russell (see, e .g ., Bateson,
1972).
If in this connection we prevent an elem ent from being a part of
itself, we thwart the possibility of a creative solution. Unfortu
nately, the theory of "logical typ es" recom m ends this prevention
and most thinkers who have dealt with paradoxes have accepted
it uncritically. O n the other hand, it is equally foolish to replace
the interdiction of confusing elem ert and higher unit with a new
one, that is, to forbid differentiating betw een different context lev
els or extensions. In so doing, one robs oneself of the newly won
freedom by replacing an old restriction with a new one.
2. The Need to Consider "T im e." I will restrict m yself to re

marking that the problem of the "self-including u n it" could be
claimed as partially resolved through the factor time. Astute think
ers might argue as follows: The classical form of paradox implies
two presuppositions: (1) There is a differentiation betw een class
and element; and (2) A defined order category can be class and
element of itself at the same time.
The second presupposition thus relies on simultaneity. If one
were to allow time to enter the consideration, one could disen
tangle confusion or avoid paradox by avoiding simultaneity. First
a defined unit can be considered as a class that includes a num ber
of elem ents. Then this class can be considered as a unit which is
an element of a higher class; thus this defined unit can be class
and element depending on what order category one selects succes
sively. If one were to now oscillate betw een the two functions of
this defined unit (class and elem ent), one could postulate that this
was the class that includes itself or the elem ent that was its own
class. This simple but exhausting trick is also often used w hen the
linearity of language is applied to describe patterns of action
which are recursive. In this case, the linear-causal if-then chain of
logic is attached to or equated with the linear-successive first-then
chain of time. If we proceed in this way, however, we rob our
selves of a perception, which was suggested by cybernetics
through the term feedback which m eans recursive processes (re is a
Latin prefix meaning back).
A way of better understanding the hum an " s e lf" (understand
ing oneself) is thus to consider recursiveness. The perception self
implies of course recursivenes: W hat I am doing feeds back (re
curses) to myself. Thus considering time can help to avoid confu
sion by oscillating betw een the two functions of a defined u n i t -

class and elem ent—but this oscillating process is rather strenuous
and tim e-consum ing. Additionally, the perception of the unit
which includes itself is a rather solid and static one. So it is neces
sary to postulate more fluid patterns, processes influencing them
selves through recursiveness.
3. The Need to Consider Self-reference, Recursiveness, or Au

tonomy. Living beings are characterized by som ething which
has been described in different term s: self-reference, self-moni-
toring, self-organization, and so on—in brief, autonomy. Ma-
turana and Varela have looked into this question in great detail
(see, e .g ., M aturana, 1982). Varela (1975) took Spencer-Brown's
indication logics further and developed the "calculus for self-ref
erence." At the same time, he added the "n eed to consider that of
recursiveness" to the "n eed to consider difference."
In therapeutic term s this raises the question of the significance
o f recursiveness for living beings. Basically, we can differentiate
betw een two possible types of recursiveness:
a. A planned action can be com pleted independently of contex
tual feedback. Living beings that act in this m anner behave as if
they were "closed, hard-programmed, autonomous system s."
b. Results or interim results have an effect along with other con
text variables on one's own actions and those of others. Living
beings that act in this m anner behave as if they were "op en ,
soft-programmed, autonom ous system s."
Maturana described the first case of recursiveness—the "m achine
m od el" of autonom ous living beings. A n example of such behav
ior would be a racing driver who does not care about the lives of
others, who drives on, relentless toward victory. This kind of be
havior is usually considered as "re-lentless [re (Latin) = back],
straightforward, and powerful." It can be called "biological auton
o m y ": All options of behavior are biologically programmed and
easy to predict.
I should like to call the second type of behavior "contextual
autonomy." Here, the social context of the action is taken into
consideration. This kind of behavior is generally called "re-spect-
ful, re-sponsible" and possibly even system ic hum ility" (see Bate
son, 1972).
4. Conclusions for a Therapeutic Model. Contextual autonomy,

described until now, implicitly focuses on one person, the context
itself not being further described. Every family therapist knows,
66 CHAPTER 3
however, that there is a vital difference in speaking about individ

uals, relationships of individuals, or system s (of relationships of
individuals). If we accept the need to differentiate, we can, w hen
including recursiveness or autonomy, differentiate betw een differ
ent contextual extensions, that is, individual, relational, and sys
temic (com posed of three or more persons).
It is also possible to differentiate betw een different contextual
extensions and sim ultaneously attribute relative autonom y to
them . Elsewhere I have called this process "recursive contextual-
ization" (see Deissler, 1983). See Figure 1.
Each of these contextual extensions—individual, relational, or
systemic—is taken in its own right, that is to say recursively, self-
referentially, or self-inclusively. This way there is no longer para
dox in the classical sense because the positing class that includes
itself is essential for autonomy.
The next step consists of processes, which run through time in
the model. Hoffman (1982) has suggested a model which she calls
"Tim e C able." This model includes time processes. If we try to
transfer this model to the therapy process, we can introduce, for
example, change of interlocutor as a "relationship param eter" in
order to illustrate systemic processes in time. Figure 2 is a model
of this kind of therapy session.
C. Conclusions
Room for confusion remains: an autonom ous subsystem (e.g.,
member of family) is not identical with the higher autonomous
system (e.g., family) of which it is a constituent subprocess. Even
if we allow the possibility of transcendence, both types of process
m ust be kept separate to avoid confusion. At the same time the
following can be said of paradoxes in the classical sense:
1. The system that includes itself as a subprocess is by definition
autonom ous. Such system s are the rule rather than the excep
tion in the field of living processes. System s which do not
have this recursiveness are "d ead ." System s which are in ad
dition self-productive or self-m aintaining are called autopoietic
by Maturana and Valera.
W hen several autopoietic system s com bine to form higher
level system s, we can speak of self-organizing system s (see
Jantsch, 1982).
2. If we then place the autopoietic system and self-organization
of several such system s at the same level, we confuse the de
fined hypothetical differentiation limits betw een at least two
system levels, which is useful in encouraging meditative exer
cises, induction of trance or, ecstatic states. O n the other
FIGURE 1
The Recursive Contextuallzatlon of Natural Processes*
Systemic Any 3-person (or more)
process level system
4 process extensions:
systemic (1: ABC)
relational (3: AB, BC, CA)
psychological (3: A, B, C)
somatic (3: A', B', C ’)
examples: family or couple
therapy
Relational Any 2-person subsystem

process level 3 process extensions:
relational (1:AB)
psychological (2: A, B)
somatic (2: A ’, B')
examples: couple. Individual
therapy
Psychological Any 1-person subsystem

process level 2 process extensions:
psychological (1: A)
somatic (1: A')
examples: m em ber of family,
m em ber of therapeutic
team
Somatic Any somatic subsystem

process level 1 process extension:
somatic (1: A')
examples: w om an, man
‘ G raphical Illustration within synchronic process perspectives. I.e., tem poral

processes a re ignored. A diachronic Illustration including time would have resulted
In a multiple com plex helix.
Source: Illustration from K. G. Deissler, "Die rekursive Kontextualislerung naturllcher
Prozesse," F a m ilie n d y n a m ik 1983, 8, p. 145.
hand, a differentiation is useful if, for example, it is a question

of understanding sym ptom s of autonom ous individuals in
self-organizing eco-groups (e.g. families).
3. W hether we call this kind of confusion (of at least two sys
tem s) a paradox depends on w hether one allows oneself to be
68 CHAPTER 3
FIGURE 2
The Person-Space-Tlme-Model of an Ongoing Self-Organizing

Human System*
"As distinct from the synchronic Illustration of the recursive co n te xtu a llza tlo n (see
Figure 1), the d ia c h ro n ic illustration considers tim e a n d s p a c e additionally. This
means that the recursion model is extended to form a m u ltip le helix.
I call this helix the p e rso n -sp ace -tim e-m od e t (PST) or co-relational-space-tim e-
m odel of an ongoing self-organizing human system. The a d v a n ta g e of this m odel Is
that it allows specification of system ic e ffe c t p a ttern s (which form a certain com mu
nication). s p a c e (where this com munication takes p lace), a n d tim e (when this com
munication Is enacted). These variables a re considered necessary to define a hu
m an system.
BEYOND PARADOX A N D CO UNTERPARADOX 69
confused or not. In my opinion, confusion is useful in the

therapeutic process, for example, to encourage creative solu
tions, so the term paradox is no longer im portant. I prefer to
follow Erickson and speak of utilizing systemic processes in a
therapeutic context.
D. Systemic Utilization—An Alternative Model

Erickson is known as the person who introduced the utilization
approach into psychotherapy—especially into hypnotherapy. If
you want to help a client with his problem s, it is necessary to
know the exact nature of his problem s, that is, the things he com
plains of. Moreover, it is also advantageous to know som ething of
his positive resources, for example, the things he does not com
plain of. If you know both levels, it is possible to suggest two
ways of getting the client to help him self: (1) The positive points
can be used to displace those described as negative; and (2) Those
described as negative can be "u s e d " in such a way in therapy that
the client is triggered to displace him self.
The second part is one which I should like to define here as the
utilization approach and is closest to what is m eant by paradoxical
intervention. The following example may serve as an explanation:
A 10-year-old boy who lives with his m other and 2-year-old sis
ter steals from his m other by taking m oney from her wallet and
then spending it on toys and candy. The m other has tried unsuc
cessfully several ways of stopping the boy from stealing. The
therapist decides to im plement the utilization approach by using
what is available to get the persons concerned to solve their prob
lem. He observes the following conditions: (1) the relationship be
tween m other and son; (2) The symptomatic relationship pattern:
The son steals from the m other who tries in vain to solve the
problem.
The therapist tells the m other to get a "th eft bag." At the begin
ning of each week she is to put in five deutschem arks in one-mark
coins. The son is told to steal a maximum of two marks a week
from the theft bag. If he fulfills this condition, he receives an addi
tional two marks pocket m oney as a reward. But if he takes more
than two marks he gets no pocket money. This procedure should
be continued for at least four weeks under the m other's supervi
sion. In order to understand the task, it is necessary to know the
following implications:
The task has the m other make the theft possible. The mother
has in addition the responsibility for carrying out the exer
70 CHAPTER 3
cise. Moreover, she has to get the theft bag. Finally, she has
to put the m oney in the bag.
The stealing is redefined and by implication desirable.
The boy either steals the prescribed am ount of a maximum two
marks and is then rewarded by the m other with an additional
two marks or he takes the m oney and forgoes the additional
pocket m oney from his m other by rewarding him self for
stealing.
M other and son report at the next m eeting that the son has taken
no more than two marks per week and has regularly received two
marks pocket money. At the same time, the son reports that it is no
longer fun to take the money. In the next m eeting, m other and son
report that stealing is no longer an issue. They agreed in front of
the therapist that the boy would get two m arks a week pocket
m oney and that the other two marks would be put into a savings
account for larger purchases (for example, a bicycle). Thus this
relationship betw een m other and son, with the therapist part of the
therapeutic system , showed relational autonom y by parent and
child deciding among themselves that the boy would no longer
steal from his mother. Moreover, with the therapist's help m other
and son found a new definition of their relationship.
This brief example describes a systemic process in which three
persons are involved: mother, son, and therapist (see Figure 1).
The therapist utilizes the relational process betw een m other and
son by:
Accepting their definition of their relationship.
Prescribing symptomatic behavior, thus defining the relational
process betw een m other and son as desirable, so that stealing
is still described as such, but by implication a new definition
of the relationship betw een m other and son is created.
In this way the son's sym ptom is utilized to redefine the relation
ship between m other and son. This is done by putting the m other
in a benevolent, affirmative role towards her son, which allows
him to do that which was previously forbidden or for which he
was punished. This "p erm issio n " removes the son's opportunity
to protest by stealing, and an opportunity to protest against this
redefinition of the relationship is given by stopping the child from
stealing. Through this therapeutic process, individual, sym pto
matic behavior is extended to a relational recursive context.
The m ost frequently-used term for this process is reframing. I
prefer to call it recontextualization, since this term can be defined
more precisely (see below).
The systemic utilization technique is com posed of:
BEYOND PARADOX A N D COUNTERPARADOX 71
1. Acceptance o f the family's definition o f their problem, w hether the

problem is seen as an individual sym ptom , a relationship, or a
m ultiperson conflict (see Figure 1).
2. Redefinition of one part of or the entire stated problem in a
systemic-recursive context, by declaring it as positive, desir
able, sensible, or useful in the frame of a recontextualization
(positive connotation).
3. Implicit or explicit recontextualization by: (1) changing the " re
cursive con text" (contextual extension, how m any people are
involved in the problem ?); (2) changing the tem poral context,
if necessary; and (3) changing the spatial context, if necessary.
4. Prescribing in part or as a whole the recursive patterns which
affect the problem or a further additional part.
E. Summary
1. Effective interventions always seem paradoxical w hen:
a. Different contextual extensions are confused and thus we
cannot differentiate betw een them .
b. The behavior defined as a problem and the interventions
aimed at this behavior are explained without considering
time.
c. The recursiveness of the different contextual extensions
and thus their respective relative autonom y are disre
garded; in other words, interventions seem paradoxical
when linear explanatory models are used.
2. Systemic utilization provides an alternative. In the process, it is
taken into account that sym ptom s and hum an behavior in
general can be described on identical levels. From a thera
peutic point of view, this involves choosing a process level for
intervention which will m ost readily trigger an autonom ous
solution to the problem . These may range from habitual cog
nitive patterns of an individual to system ic effect patterns of a
multiperson system . Thus possible interventions range from
indirect hypnotherapy (see Erickson, 1980) to those m ethods
derived from system ic family therapy (see Selvini-Palazzoli et
al., 1978, 1980; Hoffman, 1981; Penn, 1982; de Shazer, 1982).
It is the therapist's duty to accept the person-space-tim e con
tinuum of which he has becom e a part and the subprocess to
which it belongs—to learn to understand its essential systemic
patterns. Using these ongoing processes in such a way that
they can remove or solve them selves, the therapist or the ther
apy team contributes to the autosystemic transformation of the
family (see Deissler & Gester, 1983).
3. Thus the purpose of system ic utilization is to allow the group
72 CHAPTER 3
with the problems (see Haley, 1976), after a stage of systemic

confusion to solve those problem s through autonomous reor
ganization, in which the systemic effect patterns are reorgan
ized. From this perspective, systemic utilizations intervene in
the self-organizing of the hum an system by triggering solu
tions, evoking, suggesting, or preventing them which—para
doxically—strengthens the autonomy. Solutions that the fam
ily finds themselves are best, as each family is its own expert.
F. Final Metaphor
In situations in which parents push their children away, the
children are particularly apt to cling to them . If parents, however,
reciprocate (to excess) to the clinging of their children, the child
will remember that it does not want to be clung to. S/he will strive
to be free. Parents who love their children and value their own
and their children's autonomy, give a little counterpressure at the
right time or let them free themselves.
III. PSYCHOLOGICAL WELL-BEING

AND PSYCHOLOGICAL
SUFFERING
Somatic illness aside, the question arises as to w hether psycho
logical suffering is, like physical suffering, an illness—perceivable,
objectifiable, measurable. Thomas Szasz, The M yth o f Mental Ill
ness (1974), is the person who has m ost em phatically pointed out
the metaphorical nature of the term mental or psychological illness.
However, if psychological suffering is a social and thus systemic
phenom enon, relationships and system s of relationship are the
focal points on which the observer—including oneself—m ust con
centrate his attention (see Ruesch & Bateson, 1951). If the ob
server includes him self in his observations, he automatically de
fines him self as an autonomous subprocess of a larger observa
tion unit. Thus the question of recursiveness once again enters
our discussion.
The vital question in this context is: W hat is the epistem ology
of the person who calls him self a therapist with reference to the
social phenom enon psychological suffering? (Thomas Szasz raises
im portant questions here as well.) I should like to summarize my
position using the following six basic points:
1. Psychological suffering is a systemic phenom enon involving
at least the persons sharing the problem or those who consti
tute it. We refer here especially to the closer ecological group

whose member bears the sym ptom . W hen such a group en
ters a therapeutic relationship, the therapeutic organization
shares the problem (see Haley, 1976). This group and the
therapeutic team form a larger unit. This unit is the equivalent
of an eco-system ic approach to hum an problem s (see also
Keeney, 1979; de Shazer, 1982; or Deissler, 1981).
2. According to the recursive contextualization model, those
contextual extensions to social processes can be determined
which constitute a psychological problem, maintain it, or can
contribute to its solution. It is useful for therapeutic purposes
to choose that contextual extension which is best able to pro
vide an autonomous transformation of the systemic effect or
relationship pattern. In m ost cases, this will be the contextual
extension that includes close partners, that is, the family or
eco-group.
3. W herever possible, the therapist should avoid a dualistic posi
tion in which s/he appears as a m ember of a system of healthy
relationships as opposed to a system of sick relationships,
called upon from a position of health to “ cu re" the "illn ess"
of the other system . Instead therapist and team should see
themselves as a seam in the social interlacing with other eco-
systemic subsystem s. This m ethod enables the therapist to be
come an autonomous part of the "consensual sp ace" shared
with a family seeking help (see M aturana, 1982). This way
both parties—therapist and family—contribute to the solution
of complex social problems, which avoids such juxtapositions
as: therapist good, healthy, responsible—client bad, ill, guilty.
4. Psychotherapy, in particular system ic therapy, does not serve
to enforce social control or power and thus adherence to social
conventions, but to encourage novelty, that is, the creation of
new information in relationships. Thus therapy may serve—
indeed only within a small context—social variety. The old
view that psychotherapy should serve to maintain social con
ventions and thus prevent transformational processes comes
from linear cause-and-effect thinking, especially as used in
custodial psychiatry.
5. The hypothetical construction "pow er in relationships" is
seen by some authors as being central to the understanding
and maintenance of psychological problem s (see especially
the work of Haley, e .g ., 1959 or 1980).
O n the other hand some authors (see Bateson, 1972), sug
gested that it was not power itself but the belief in or episte-
mology of power that causes certain groups of people to be
come caught up in this model (see Stierlin, 1982). The thera
74 CHAPTER 3
pist's job is to manipulate these "pow er games/' to encourage

or prescribe them in such a way that they becom e superflu
ous. Thus the family is helped to abandon its epistem ology of
power by implication in favor of a system ic epistem ology
which in turn can strengthen the family's autonomy.
Therapists who accept this point of view are beyond the
power construct: If families behave according to the power con
struct, the therapist is accepting, but works with the family to
transcend this construct by planning the power gam e, thus
making it superfluous for the family.
6. The m ost abstract and general form ulation of w hat accompa
nies psychological suffering in hum an relationships is that re
lationship system s are fluid (ongoing), system ic patterns that
effect an accelerated evolution, and two conditions restrict the
evolutionary nature of this effect pattern process. We can dif
ferentiate betw een these two diametrically opposed processes.
a. The system that is in a perm anent state of development,
change, and reorganization has slowed its development
rate due to specific events or the course of specific proc
esses. We can call this changed speed of development
anachronic. An example of this phenom enon is the family
that cannot let its children becom e autonom ous.
b. The reverse situation in w hich development is too great is
equally possible. For example, the severing of a child from
the family happens too quickly or the m embers of the
family have no reliable basis for a relationship, thus sys
temic effect patterns change more rapidly than those in
volved can tolerate. In such cases as above, family m em
bers com pensate with sym ptom s. This kind of accelerated
development can be called metachronic.
c. In both cases, those involved have good reasons to accept
the anachronic or metachronic developmental rate of
change. It is the therapist's task to find these reasons—
through an appropriate exploratory process—in order to
use them effectively in therapy (an exploratory process
w hich is a further development of "circular questioning"
is described by Deissler, 1983).
IV. THERAPEUTIC METHODS

The following case is presented to give the reader closer insight
into therapeutic process as practiced and taught by th e Institut fur
Familientherapie M arburg. The exam ples can be called typical, in
that they are based on a detailed analysis of "problem -form ing

effect patterns" and further describe the systemic utilization tech
nique. To m ake our m ethods easier to understand, I should first
like to define more closely the basic therapeutic framework.
A. Therapeutic Framework
The Institut fur Familientherapie M arburg is a private institu
tion not financed by any public body. It is thus free from regula
tions or restrictions that could hinder its therapeutic m ethods.
The Institut's aims consist of: (1) investigating "problem -form ing
systemic effect pattern s" in naturally occurring hum an groups—
w hether fam ilies, com m unes, groups of residents, or other
groups—in order to develop and apply the m ost efficient, appro
priate, therapeutic strategies to these natural form s of hum an self
organization; and (2) examining these strategies for short- and
long-term effects.
The therapeutic process is as follows: All the basic data for the
problem group are available before the first interview; these data
are noted during the first telephone contact with a m em ber of the
problem group (see basic data form, Appendix I).
All therapy m eetings are prepared via team work and con
ducted under live supervision. A fter a m inim um of 45 m inutes,
the therapeutic session is interrupted for further deliberation by
the team. These interim deliberations take at least 15 m inutes, but
in difficult cases last up to 90 m inutes. During these interim delib
erations, we attem pt to develop a hypothetical explanatory model
of the problem -form ing effect pattern and, as far as possible,
evolve an appropriate intervention. If, following the interim delib
eration, the team does not agree about the problem -form ing effect
patterns, further explorations are made according to the different
hypotheses or the family is asked without explanation to come
back for the next session.
If the therapeutic team agrees on the problem -form ing effect
patterns, the family is asked to return to the next session and the
intervention is applied immediately or disclosed as a written proc
ess diagnosis with or without prescription within the following
days. A nother possibility is for the therapeutic team to announce
its conclusions at the end of the following session (this form of
intervention gives the family an opportunity to m ake its own deci
sions about the internal search processes and possibly to prescribe
som ething itself). The interventions are planned and carried out
based on the system ic utilization technique. At the end of the ses
sion the therapeutic team discusses the anticipated effects with a
76 CHAPTER 3
view toward intervention. The entire therapeutic process (prepa

ration, application, intervention, anticipated effects) is noted on a
specially developed therapy process form. This procedure is also
used with respect to the effects actually perceived in the following
session (see checklist for therapy process form, Appendix II).
The therapeutic technique used in an individual case depends
on the course of the therapy and has to be redecided from case to
case; too much space is required to describe this procedure in de
tail here. A checldist (see Appendix III) may be used to decide on
the therapeutic procedure. It should be pointed out, however,
that before each intervention, the hypothetical explanation model
has to be developed and consolidated (where possible) by circular
questioning. The intervals betw een each session average three to
six weeks in order to give the commentary, tasks, and so on an
opportunity to take effect (see Selvini-Palazzoli, 1980).
There is no official upper limit to the num ber of sessions: Thus,
the family is under no pressure to improve or not improve in a
fixed num ber of sessions. The aims of the therapy are stated by
those seeking help, and the therapy team accepts them . However,
the team reformulates these aims by converting them to systemic
process level. This can, for example, m ean that the team sees the
family's problem as one of organization, whereas the family
merely sees a daughter having a somatic problem . The therapists
do not declare their systemic construction of the problem unless
doing so is specifically planned as a therapeutic measure. Neither
does the team discuss the way the therapy works with the family
unless they come to the conclusion that, for research reasons,
such a discussion will be useful. For example, to discover the neg
ative effects of a particular procedure, the team encourages it.
B. A Case of a “Pathological Craving

for Marriage”
As the title suggests, I am exceeding the bounds of the remarks
above about psychological suffering and poking fun at psycho-
pathological categorizations. However, I would like to point out
the following: If the therapy team had used this diagnosis as a
basis for treatment, the therapy would have been unsuccessful.
This kind of categorization contains three decisive errors:
We pretend that the problem-forming effect pattern is caused
by one person.
We pretend that this cause is a negative entity within the per
son said to be ill.
We define the problem more or less explicitly as being insoluble—
either for the therapist or the system itself (see also Haley, 1980).
There is, however, a problem in describing clinical cases which

have been successfully treated. If we omit "declarations of ill
n ess" from a collegial point of view, we quickly find critics of our
own work who claim, "T h e case you're telling us about has noth
ing to do with illness. That's why it's so easy to treat." Still, let us
look at our concrete exam ple.1
A young, unmarried couple, Lars, a 33-year-old lawyer, and
Marie-Ann, a 29-year-old social worker, have known each other
for about five years and have been living together for three. Lars,
who had been married for eight years (until shortly before he
started therapy), had been separated from his former wife for five
years; he has a 12-year-old son by this wife. His contacts with his
former wife are sporadic and contingent on his contact with his
son. However, his form er wife has contact with his parents, espe
cially his mother. He is fairly close to his own family and has regu
lar contact with them . Marie-Ann hardly has any contact with her
family whom she partly rejects. W hile living with Lars, she often
expressed the wish to get married. Lars usually rejected this; he
was still married. Then, about one and one-half years ago, Marie-
A nn presented Lars with an ultimatum: "E ith er you divorce your
present wife or we split u p ." Lars capitulated and got divorced.
Shortly after the divorce, Marie-Ann confronted Lars with an
other, more or less ultimative request: "E ither we get married or
we split up." Lars rejected this second ultimatum. As the situation
deteriorated and the couple becam e increasingly bogged down in
a vicious circle of quarrels based on dem ands, refusals, accusa
tions, and counteraccusations, they came to us.
The therapy team constructed the problem -form ing effect pat
tern in the first interview ju st as it was described. We acquired the
additional information that Marie-Ann described the relationship
between Lars and his m other as very close; she also felt his par
ents had a negative attitude toward her. Lars stated that he would
not consider marriage after his recent divorce, and denied a close
relationship to his mother.
In the interim deliberation, the team considered what could be
done to encourage the couple to find an autonom ous solution to
their problems. We had little difficulty with the positive connota
tion of the problem, w hen a good correspondence betw een the team
and both the partners becam e apparent during the session.
In order to make a solution possible, it becam e necessary to put
the problem into a new context. We decided to transfer the prob-
'in this case the therapeutic team consisted of: Klaus G. Deissler, therapist;
Peter W. Gester, live supervisor; and Dorle Engel, minute taker. Therapeutic in
terventions are always a result of joint efforts.
78 CHAPTER 3
lem from the partners to their parental families; that is, to make
the problem one of their need to sever them selves from their
parental families. Thus we implicitly gave the problem of mar
riage—which was of primary im portance to the couple—a secon
dary position.
In addition, we accepted M arie-Ann's desire to get married as
the "official" problem and finally, we gave consideration to the
fact that Lars frequently discussed his problem s with his parental
family. To avoid antagonizing the partners, we decided to give
them a task which would take both parts into consideration. Thus
we were able to prescribe the problem -form ing effect pattern. This
session was concluded with the following intervention:
1. The therapist led the partners to understand analogically (with

voice tone, gestures, etc.) that the team accepted the couple's
relationship and supported them in finding a solution to their
problem s (analogical or implicit positive connotation).
2. He told both partners that the quarrels about w hether to get
married or not were affecting their relationship (acceptance o f
problem definition), but that the therapeutic team saw the major
problem as their not having settled the question of their rela
tionships to their parental families and the marriage issue as a
way to avoid sorting out these family problem s. If there was
to be a solution, the contexts would have to be clarified (recon-
textualization).
3. For the above-m entioned reasons, the following task seemed
appropriate: They were to write a joint letter to their relatives
in which each would give his/her point of view about mar
riage; each was to give three of their argum ents. They were
both to ask for help in solving the problem and request that
the relatives respond within a reasonable time (utilization of
the marriage problem, especially in relation to M arie-Ann,
and utilization of the family ties in relation to Lars) to trigger
multiple systemic effects. Aim: Autosystemic transformation of
effect patterns in part by the mutual writing of a letter to the
relatives, so that Lars and M arie-Ann would be regarded im
plicitly as a couple. They both accepted the com m entary and
the task: The therapy team had the im pression that Lars was
affected because he quietly asked further questions about the
task, whereas M arie-Ann sighed and thought aloud what ef
fects of carrying out the task m ight be. All in all, both Lars
and M arie-Ann seem ed satisfied with the advice they were
given.
A few weeks later, Marie-Ann called us and said they didn't need
to keep their next appointment; the problem had been solved. Som e
family turbulence had led to Lars severing himself more from his
mother and his ex-wife and becoming closer to Marie-Ann. They
were both happy with the result: Marie-Ann didn't want to marry
Lars now that she realized he was closer to her.
We asked both partners to let us know about reactions to the
letter. We have copies of Lars's and M arie-Ann's letter as well as the
replies. Together, they constitute 20 handw ritten pages. This is the
text of the letter that M arie-Ann and Lars sent to their relatives:
Dear
You will certainly be surprised to get a letter from us. We are
writing you to ask your advice on a problem that has been bother
ing us for some time.
Marie-Ann wants to get married and Lars doesn't.
This is Marie-Ann's point of view:
I am convinced tht Lars is the "right" man for me. I am also for
marriage because it's an official and recognized step for living to
gether. I also think that marriage puts the seal on growing up be
cause children finally leave their parental family in order to found a
new one. I think the ceremony of marriage is important and neces
sary in this context.
Lars rejects marriage for the following reasons:
I don't see why the state should give its O.K. to our relationship
and I don't want the state's and attendant expectations and rules
associated with it.
I don't think that marriage would bring about the changes
Marie-Ann wants—also regarding her own family. What's more, I
don't want to make the mistake again of saying "yes" out of con
sideration to a step which I can't stand by.
Many discussions have not helped us to find a solution or to
come to an agreement. So we would like to ask you to write to us
b y . . .with your advice. As we would like to hear your own point of
view, please write without consulting other people. We are looking
forward to hearing from you.
The answers—as you can im agine—were varied. I will present
only the gist of the responses here to give the reader an im pres
sion of the "system ic effects" of the letter:
Lars's m other wrote two letters in quick succession: She told
Lars that she could understand his doubts, suggested separa
tion, and added that each should look for a new partner. In a
further letter, she supported marriage suggesting sterilization
for one partner to facilitate a com prom ise of " n o family."
80 CHAPTER 3
Lars's aunt suggested living together as man and wife as legal

marriage caused enough problems.
M arie-Ann’s mother, answering also for her father, supported
M arie-Ann's wish for legalizing marriage. At the same time,
she apologized for her two "la z y " sons who are not going to
reply.
Lars's grandmother believed both partners would find a solution
and was sorry she couldn't help any further.
Lars's father didn't believe in marriage; he wished them both a
good time together.
Lars's sister, already twice divorced, advised against spoiling the
good relationship through a half-hearted marriage.
Lars's ex-wife—herself a therapist—wrote in great detail about
the "reactionary nature" of marriage. Sh e tried to set some
things straight about her marriage to Lars, for example, she
sent Lars the bill for the divorce. As far as Lars and Marie-
A nn were concerned, she thought they were both aiming for
a separation.
Marie-Ann told us in writing that there had been an argument
between Lars and his m other after the letter in which Lars em pha
sized his autonomy and took M arie-Ann's side. Marie-Ann said
that was all she wanted; marriage was no longer im portant to her.
Furthermore, her relationship to Lars had changed. He did more
housework and had assumed responsibility for contraception.
Lars confirmed these changes with the qualification that the argu
m ent with his m other was less dramatic than described by M arie-
A nn.
C. A Case of “Psychotic Loss of Identity”

This neologism shows that the inventivenss for psychopatholo-
gizing knows no bounds. I do hope, though, that not all readers
will take offense at this further irony.
The family to be described here was sent to us by a school psy
chologist who had heard of our work. Sh e worked in a town
about 140 kilometers from Marburg, where the family came from.
The father of the family got in touch with us to make an appoint
m ent. He was asked to tell us about the problem in writing before
the start of the therapy. A good m onth before the therapy started,
we received the following letter:1
'T h e letter is very hard to read and contains many orthographical and gram
matical mistakes. To make it easier to understand, a corrected version is repro
duced here.
Esteemed Mr. Gester,3

We have already spoken on the telephone. I shall now try to
describe our problem.
We are family W.: Father Helmut, 49 years old (civil servant, fire
brigade); Mother Gertraud, 45 years old (housewife); Berthold, 15
years old (secondary school); Uwe, 13 years old (secondary school).
Married 1966.
My wife had to leave her home town and move to R. (had dif
ficulty in settling down, many quarrels during the first years of
marriage).
After a miscarriage, birth of Berthold, 1968—small apartment, no
nursery.
Berthold was a restless child, cried a lot. I often shouted at the
boy and smacked his bottom.
Berthold was timid towards other children. Two years later, Uwe
was born. At first, Berthold always wanted to push the baby away,
but then he got used to him and played with him—especially as my
wife and Berthold always had and still have a close relationship. As
a small boy, Berthold was always interested in electric wires, his
favorite occupation was switching lights on and off.
A little cushion, his "heia" was his constant companion; it then
became so unappetizing that I threw it away on the advice of the
grandmothers. This was a shock for Berthold. A few days later he
looked for a doll he could take to bed or on journeys and even now,
at 15 years of age, the doll always has to be there at bed time. Until
he was 12, he often spoke to the doll, especially before he went to
sleep.
When he was four, Berthold had to go into hospital for 4 days
for a throat operation—he went to his bed with much screaming
and struggling—visiting was not allowed in those days. When I
was able to collect him 4 days later, he was lying there quite apa
thetically. It was only when he saw his things that he dressed
quickly and left the hospital like a whirlwind. At home, he clung to
my wife's dress and followed her wherever she went.
1972—move to a larger apartment. The children got a room of
their own.
Berthold started kindergarten; Berti just sat quietly in a comer
and didn't play with the other children, then a year later, when
Uwe started kindergarten, things got better. On our many visits to
the grandparents, where there were a lot of children whom
Berthold knew, he played with them gladly and lively.
School: To start with he had great difficulty in getting on with
the teacher and other children. Difficulties with writing and draw
ing (illegible), even now an urge to use only the right-hand side of
a piece of paper. Favorite subjects—physics and biology—good
3In this case the therapeutic team consisted of: Peter W. Gester, therapist; Klaus G.
Deissler, live supervisor; and Andrea Bohnke, Dieter Heim, and Hilde Krott, minute
takers.
82 CHAPTER 3
grades there too. Between the ages of 7 and 13, we took Berthold to
the education advice centre, then to the schools' psychologist in R.
(conversations with the therapist helped us the parents, and work
ing in the group helped Berthold a lot).
Berthold's school results got better and he became more open
towards others.
A year ago, the group work finished. At the same time he got a
new teacher. Berthold related well to this teacher, a motherly, but
determined woman who understood him. The new teacher, young,
energetic, strict—and the class was reorganized (few old class
mates)—this put a lot of pressure on Berthold.
During the next school holidays I took my boys on a cycling trip
and noticed some strange behavior in Berthold: before going to bed
he would fold his trousers over and over again to get them into a
particular position. Before, when eating, he would sometimes
make movements with his hands as if to lift the food through the
gaps. When he felt he wasn't being watched, he made head move
ments as if he wanted to pull up an object. When he got on his
bicycle and rode off it looked as if he wanted to pull the object
behind him. After a lot of asking, Berthold said he had built up
little walls.
His school results deteriorated, his behavior was strange there
too: when he sat down or stood up, he would swing his legs over
the back of the chair as if he had to climb over a wall.
On his way home from school, Berthold walked as if on a
marked line. After many talks with Berthold and the teacher, who
takes a lot of trouble over him, and who we get on well with, he
gave this habit up, or substituted another one: the doors to rooms
had to be closed, open doors bothered him—and washing his
hands, 5-6 times in succession isn't unusual. You have to be insis
tent to stop him.
His interests: intense playing with Lego, building bridges, aerial
railways, gears, reading, asking questions about something that
has aroused his interest—in detail and intensively.
Esteemed Mr. Gester, I shall stop now, until our appointment
on ___
With best wishes, Helmut W.
In principle, this letter gives enough information about the
family to enable the team to work out an intervention from it,
although the father's description is centered strongly on the
sym ptom s of the son Berthold.
In the preliminary interview, the letter and the inform ation it
contained were used to establish investigative directions that
would lead to concrete information. The interview revealed the
following analogical information:
1. The father answ ers nearly all the questions, including those
not directed at him.
2. The m other is extremely retiring, almost shy. Sh e leaves it to
the father to answer the questions—including those directed
at her—by claiming to have forgotten som ething.
3. Berthold hardly answers any questions; Uwe answ ers but of
ten indirectly hands things over to the father.
4. Berthold som etim es acts as if he h asn 't even heard the ques
tions he is being asked; he stares ahead or looks blank.
5. The m other seem s to be suffering and the father becom es
active.
6. During the interview, Berthold makes several imaginary
m ovements, as if to carry out certain actions.
The interview produced the following additional inform ation:
1. Berthold had undergone play therapy for three years—from
1979 on—but his sym ptom s were still present. His father ex
plained the imaginary actions: Berthold wanted to build a
protective wall around him self.
2. The parents married relatively late (he 32, she 28), because the
wife didn't want to leave hom e—about 70 kilom eters from her
parents' hom e. (She was very attached to her parents.)
3. According to the father, M rs. W. got on better with Berthold
in the past than now. Mr. W. is equally fond of both sons.
4. Four years ago M rs. W .'s father died. M rs. W .'s m other then
lost the will to live: Her children and grandchildren couldn't
com fort her after her loss.
5. According to Mr. W., his wife suffered the m ost during her
father's illness; she worried that if he died she would go mad.
During the interim deliberation, the therapeutic team discussed

extensively the additional inform ation they had gathered. W ith
respect to the inform ation that the interview had essentially con
firmed, the problem seem ed clear: The team m embers did not dif
fer in their reconstruction of it. The team was concerned, how
ever, that the family m ight disqualify a directly-given verbal com
m ent. They th u s decided to send a w ritten, elaborated,
hypothetical, explanatory model as "p rocess diagnosis" to the
family. After doing so, they told the family that the next appoint
m ent would be in about seven weeks: O nly the parents were in
vited. After the family was told about the process diagnosis, they
were immediately dismissed.
The diagnosis elaborated and written by the therapeutic team
84 CHAPTER 3
was sent five days after the session. The letter is reproduced here
verbatim, after which the elem ents of the systemic utilization
technique used for this intervention are summarized. The ele
m ents can be further resolved with the aid of the checklist (see
Appendix III).
First we gave instructions with written interventions including
how, when, and by whom the letter should be read. O ur instruc
tions were as follows:
Dear Mr. W.,
As promised we are sending you our diagnosis for your family.
We ask you to proceed as follows: wait until Saturday evening
before handing each member of the family the sealed, individually
addressed envelopes. This should be done shortly before supper
when the whole family has gathered to eat. Only then should the
envelopes be opened. This applies to yours as well.
As soon as each member of the family has taken their letter out
of the envelope, read your letter aloud, slowly. Please make sure
that you are not interrupted. After reading the letter you can have
your supper.
Yours sincerely,
Peter W. Gester, for the therapy team
This is the text of the letter itself:

Dear W. Family,
As promised during your visit to Marburg, we are sending you
our diagnosis.
You, Mr. W., love your wife and family very much. You care for
them in big and little matters in a way that only very few fathers
do. You have been doing this since the start of your married life.
You, Mrs. W., are the same. Through your marriage, you were
separated from your family, and especially the separation from
your father (grandad) has caused you much pain. You made the
sacrifice of leaving your beloved parents for the sake of your hus
band's job.
You, Mrs. W., were always quiet, like your father (grandad) and
so as not to disturb your husband you never told him how much
you were suffering through the separation from your parents, espe
cially your father (grandad).
You, Berthold, noticed this even as a small child. Therefore you
decided early on to comfort your mother during this separation.
Doing this without considering yourself, you sacrificed your own
development.
When you, Mrs. W., suspected your father was going to die,
you were afraid of going mad, because you feared you wouldn't be
able to bear the grief of the bereavement. In addition, you were

worried whether your mother would survive your father's death.
After grandfather's death you decided to bear this stroke of fate
quietly.
You, Berthold, noticed all this exactly and took the only possible
decision, namely to comfort your mother and your grandmother:
this meant going mad instead of your mother and being like your
grandfather; adopting his characteristics and behavior (making
things, being quiet and withdrawn). Your constant hand and finger
movements also express the fact that you took on your grand
father's manual skill to comfort your mother and grandmother.
You, Uwe, decided to help your parents with their problems by
being the family sunshine.
The therapy team is deeply impressed by the mutual concern in
your family. We especially respect the sacrifice made by Berthold to
console mother and grandma.
Therefore we cannot ask that you, Berthold, change your behav
ior or even abandon it, but on the contrary: We want to set you the
task of maintaining your sacrifice, at least in your mother's and grand
mother's presence, behaving like your grandfather until no one can tell the
difference.
Yours sincerely,
Peter W. Gester, for the therapeutic team
The systemic utilization technique used here is com posed as

follows:
1. The instruction to read the letter out loud implies a ritualized
prescription that utilizes the pattern o f action (speaking/silence
pattern) of the family.
The father reads the letter, the others have to listen.
The following supper ensures that they all “ sw allow " the pre
scription together and then “ mutually d igest" it over a longer
period (metaphorical or symbolic utilization).
2. The entire com m unication pattern of the family is given a posi
tive connotation: mutual affection, father's activity, m other's
pain at separation, Berthold's sacrifice, Uwe's efforts, and so
on.
3. The problem as defined by the family is accepted, but recon-
textualized by including the m other's parental family—grand
parents, parents, Berthold.
4. Finally, the recontextualized problem-forming effect pattern is uti
lized by prescription: In the presence of at least his m other and
grandmother, Berthold is to behave like his grandfather “ until
no one can tell the difference." This implies an increased de
mand with the implicit continuation of Berthold's behavior.
86 CHAPTER 3
About a week later, the colleague who had sent the family to us
called and said that Mr. W. had been to see her and had shown
h er our letter. He had told her that the family was com pletely con
fused by it, especially Berthold. For example, Berthold was asking
if he was mad and would have to go into a m ental hospital. Mr.
W. was also surprised that only the parents should attend the
next session.
In our colleague's opinion we had hit the nail on the head. She
had had the sam e hypothesis for years but had been unable to
com m unicate it to the family. Sh e said she had asked Mr. W. not
to talk to her about the content of the letter, but only to us.
A s previously agreed, the couple came to the second session
alone. They didn't seem at all confused; on the contrary they
were rather balanced. Possibly the weeks in betw een had lessened
the effect of the letter. Their spoken contributions were more
evenly distributed. W ithout being asked directly, Mr. W. m en
tioned that Berthold had calm ed down, w asn 't fidgeting or
"m essing around" anymore (imaginary actions). The grand
mother, whom they were visiting at the time, had noticed this
too. However, Berthold was shocked that, according to the letter,
he was behaving like his grandfather, and asked on occasion
w hether he was mad.
M rs. W. confirm ed her husband's observations, but pointed out
that the improvements could be due to the present school holidays
and that Berthold's behavior could occur again w hen school
reopened. But, she concluded, Berthold really had becom e quieter.
Mr. W. added that he thought his wife had been shocked by
the letter too. M rs. W. confirmed this; the word mad had been so
"o d d ." Further them es of the session were: (1) how the partners
had got to know each other; (2) their present relationship to Mrs.
W/s m other; and (3) their sexual relationship.
The interview didn't bring any im portant new inform ation to
light: The hypotheses were confirm ed that M rs. W. was still sad
about leaving hom e. Mr. W. continued trying to ease his wife's
separation from her parents at considerable cost to his own psy
chological well-being, and they were virtually unable to discuss
their sexual relationship.
After the second session, the family received a further interven
tion. It was similar to the first, but em phasized more directly M rs.
W/s com forting function for her mother, as well as Berthold's for
his. In addition, relapses were predicted for Berthold. The family
w as also advised to discuss this letter with the school psycholo
gist and ask for advice.
The children were invited to attend the next session at which
time the effects elicited by the first letter were confirm ed. Therapy
ended after the fourth session. The school psychologist reported
further improvement in Berthold's behavior, w hich was con
firmed by several of his teachers.
In drawing a conclusion from this therapy, we can assume that
the first intervention released system ic confusion (compare the
school psychologist's reports). The therapy team further hypothe
sized that the confusion was started by an autonomous reorganizing o f
the patterns which effect in the family (autosystem ic transform ation).
V. FUTURE RESEARCH
In 1978 I carried out an investigation in a school, the object of
which was a quantitative check of paradoxical interventions on
speech anxiety before classm ates. For various reasons, I have yet
to publish the results, the m ost im portant being that to a certain
extent I concluded that quantitative investigatory m ethods are
com pletely unsuited to the precise evaluation of the system ic ef
fects of therapy. To explain this further, I should first like to sum
marize the m ost im portant conclusions that I, personally, have
drawn from the results:
1. Quantitative tests for m easuring gradual differences in the de
gree of fear param eters are easy to construct and, w hen well
prepared, provide accurate results.
2. Standardized paradoxical procedures used with groups of
people assem bled based on quantitative criteria (using statisti
cal parameters of comparable control groups) cannot be more
precisely analyzed because of their qualitative effects. The
quantitative effects of these standardized procedures deviate
widely, but do not show any significant mean differences.
Am ong other things, they do not take into consideration the
relevant qualitative conditions of the test persons, but merely
the degree of certain parameters.
3. Inexperienced therapists, even if they have been taught indi
vidual techniques and had several practice lessons in using
them , are not able to com pensate for the standardization ef
fect (see no. 2 above). O n the one hand, these therapists are
tied to the standard procedure, and on the other they are un
able to estimate the effect of other possible positive influences
that they m ight use constructively.
4. The m ost significant disadvantage of standardized paradoxical
procedures is the fact that the relational or system ic context is
88 CHAPTER 3
not considered and thus the concept "system ic patterns

which effect," whose essential com ponents are individual
symptoms, is neglected as an influencing network. These
contextual effect patterns need to be considered in paradoxical
procedures if one is to work efficiently: The systemic utiliza
tion technique is a procedure that m eets this requirement.
Thus I conclude that quantitative investigations conceived ac
cording to the "isolated variation in variables" model and at
tem pting to separate effect variables are, for therapeutic pur
poses, of little value.
If we accept the above conclusion as being correct, the question
of how these systemic effect patterns can be exactly evaluated
arises. Let us remember that we are evaluating fluid, ongoing re
lationship patterns. I believe that we are at the beginning of a sci
entific development within this field, but I suspect that the first
steps toward evaluating systemic effect patterns have already
been taken. I should simply like to cite som e models that I believe
will codetermine future research: Spencer-Brown (1979), Varela
(1975), Taylor (1979), and Pearce & Cronen (1980).
These models have one thing in com m on: They make it possi
ble to show isomorphic recursive process patterns, thereby help
ing to show ongoing relationship patterns in natural groups. The
above-mentioned models have not been sufficiently developed to
make them directly usable for practical purposes. Thus their fur
ther development is necessary.
Finally, I should like to point out that the therapist as well as
the participating observer m ust include themselves in the consid
eration of these models to achieve a circular or recursive consider
ation of ongoing effect patterns. This m eans, in principle, that the
therapist m ust direct his own behavior, that is, his therapeutic
skill toward the effect patterns of which he is a constituent part.
VI. THERAPEUTIC ETHICS

No therapeutic procedure has been or is so attacked from an
ethical point of view as the so-called paradoxical procedures.
W hy? The answer is simple: The problem is com plicated. The
more efficient the therapeutical procedures, the more they are
suspected of being manipulative. In Germ an-speaking countries,
people are especially cautious in this respect. The original lan
guage of psychoanalysis is Germ an. Thus in Germany, m any ana
lytically oriented psychotherapists see their thinking and actions
threatened w hen new, efficient procedures—like paradoxical
ones—start displacing traditional therapeutic m ethods. It is not

surprising that critics of paradoxical procedures attack them at
their most sensitive point—efficiency. Paradoxically, a decidedly
weak point of efficient procedures is their efficiency: Or, the
strength of these procedures is their weakness.
Let us take an example: O ne can use a knife to carve wood, cut
bread, carve meat, or—and nobody can deny it—kill people. Ther
apists concerned about efficiency are confronted sooner or later
with a central question, the significance of which only few people
have reflected upon. W hat type of person considers using him self
as a m eans for change—a trigger for change in other people—in
such a way that som eone seeking help really feels better after
therapy? Moreover, the uninitiated will ask what is occurring
w hen an intervention prescribes the very thing that is contrary to
one's thinking: This is immoral or underhanded, is it not?
My own therapeutic work has been called "technization," "cold
intellectualism," or "contem pt of hum an b ein g s"—even w hen cli
ents felt much better after therapy. The effectiveness of my thera
peutic work did not play any part in its criticism. M y values were
questioned: Is it permissible to intentionally m anipulate people
therapeutically so that they will feel better afterwards?
I am not sure how I would convince som eone observing my
therapeutic work and doubting my integrity that I respect natural
processes—and hum an life in particular. I try to be as unpreju
diced as possible—neutrally concerned—toward the clients who
come to me.
In my opinion the gist lies elsewhere. Thom as Szasz pointed
out in his book, The M yth o f Psychotherapy:
Franz Anton Mesmer nowadays may be considered as the father of
modern psychotherapy. This is a late acknowledgment of his work.
But Mesmer made some mistakes in his therapeutic endeavors—and
therapists in these days still do. Let me remind the reader of Mesmer
treating Maria Theresa Paradies. This girl—an excellent piano
player—suffered from "hysterical blindness." Mesmer treated her
successfully, but he did not respect two contextual aspects concern
ing this case:
— her family (the ecological context of his patient)
— his own colleagues (his own professional context)
Therefore the result of his success was—in short—that this case was
his greatest failure. His envious colleagues warned her family that
their daughter might lose the pension given to her by the empress,
the family got angry at him, the girl felt worse after having re
gained her sight, she played worse piano and a campaign against
Mesmer was started by his medical colleagues. Finally Mesmer left
Vienna.
90 CHAPTER 3
W hat does this Paradies-metaphor tell us? I think it is the fol

lowing: Therapists who try to gain power through their thera
peutic work, e .g ., their successes, are doomed ultimately to fail.
We can paraphrase Bateson and say that they d on't realize that
they them selves are only parts of higher-level evolving system s
that cannot be m anipulated unilaterally and linearly according to
the construct of power.
APPENDIX I: BASIC DATA

Date: Family doctor:
Referred by: Medical insurance:
I. Caller 2. Date of birth:

3. Profession:
1. Name:
2. Address: 4. Phone:
3. Reason for c a l l-
current problem:
VI. Parents of Partner
4. Date of birth: 1. Name:
5. Profession: 2. Date of birth:
6. Phone: 3. Profession:
4. Phone:
II. Partner
1. Name: VII. Siblings of Caller
2. Address: 1. Name:
3. Start of partnership: 2. Date of birth:
4. Date of birth: 3. Profession:
5. Profession: 4. Phone:
6. Phone:
7. Marriage: VIII. Siblings of Partner
8. Separation:
1. Name:
III. Children 2. Date of birth:
3. Profession:
1. Name: 4. Phone:
2. Date of birth:
3. Profession: IX. Pets
4. Phone:
X. Previous
IV. Further Partners Psychotherapeutic or
(Living Together) Psychopharmacological
1. Name: Treatment
2. Date of birth:
3. Profession: XI. Future Notes
4. Phone:
XII. End of Therapy
V. Parents of Caller
1. Name: XIII. Result
BEYOND PARADOX A N D CO UNTERPARADO X 91
APPENDIX II: CHECKLIST FOR

THE COMPLETION OF A
THERAPY RECORD
A. Planning of Therapy
1. Formal Data
a. Date.
b. Name of the family.
c. W hich people were required to attend.
d. M em bers of the therapy team present.
2. New Information before the Session
3. Hypotheses
a. System ic—the entire defined problem -bearing system .
b. Relational—various relationships within the system .
c. Psychological—the individuals.
d. Som atic—the individuals.
4. Therapeutic Measures
a. Joining.
b. Recursive creation of information.
c. Intervention at end of session.
B. Therapeutic Procedures
5. Formal Data
a. Date.
b. M em bers of family present/absent.
c. Length of session.
d. Video—audio—live.
6. Effects Triggered by the Therapeutic Intervention o f the Previous
Session—Observed or Reported
a. System ic (see above).
b. Relational.
c. Psychological.
d. Som atic.
7. How Is the Therapeutic System (Including Therapist) Tuned during
the Session?
b. Relational.
c. Psychological.
d. Som atic.
92 CHAPTER 3
8. Which Therapeutic Measures Were Carried out Compared with

Those Which Were Planned?
a. Joining. (How did the therapist actually join the family?)
b. Recursive creation of inform ation. (W hich questions
did the therapist actually put to the family; how did he
do this?)
c. W hat effects did the therapist evoke through a and b?
9. How Was the Interim Discussion?

a. Length.
b. Team in agreement?
c. W hich interventions did the team decide on?
No intervention (reasons).
Further exploration in next session (reasons, fields).
Verbal com m entary at end of session with or without
task (brief formed and context sketch).
Preparation of written intervention (brief content and
formal sketch).
10. End o f Session

a. Was the session terminated by the therapist as planned
by the team; did he make im portant changes?
b. Did the final intervention bring new inform ation to
light?
c. How did the family, in general, take the final interven
tion?
11. Next Appointment

a. Date.
b. Persons required to attend.
12. Mistakes and Difficulties Noted at the Session
13. Anticipated Long-Term Effects o f Intervention

b. Relational.
c. Psychic.
d. Somatic.
14. Prepatory Considerations for Planning Next Session

a. Are there any special points that were not dealt
with during the session?
b. Are there new directions for probing in the next
session?
APPENDIX III: CHECKLIST FOR

DESIGNING SYSTEMIC UTILIZATIONS
A. Phases in the Designing of

Systemic Utilizations
The formulation of systemic utilizations assum es that the thera
pists are familiar with the basic prem ises of systemic thinking and
practice. In addition formulating systemic utilizations can be use
ful for exercises or beginners to aid in differentiating betw een the
following phases of the team consultation:
1. The Cathartic Phase. In this phase, every member ot the

team should be allowed to express his feelings—w hether negative
or positive—even if they do not contribute to the solution of the
problem. Each team member thus has the opportunity to express
first impressions of the family. Since this phase is called "cathar
tic," we can understand the role of expression of feelings for
therapeutic interventions.
2. The Analytic Phase. This phase immediately follows the ca

thartic phase. It enables the team to collect analytical elem ents
(mosaic stones) and to develop new hypotheses about the prob
lem . Every m ember of the team can express his thoughts, as
sum ptions, constructions of the problem, and so on.
3. The Positive-Connotative Phase. The analytic phase often

produces the result that most analyzed elem ents of the problem
are seen negatively by the people who express them . It is often
amazing to observe how hard a team m ust work to wrest positive
aspects from a problem. However, serious aspects of problems
should not be softened, even in this phase. Usually, it is possible
to develop a positive connotation of the problem, which implicitly
evokes the family's cooperation and at the same time suggests so
lutions to the problem.
4. The Systemic Phase. Only w hen the team has developed a

positive aspect of the whole problem spectrum is it able to reach
the systemic level. In this phase, an understanding is developed
based on a review of connected effect patterns. The problem-
forming effect patterns can be constructed by the therapeutic
team, which defines itself as a "participating ecological partner."
94 CHAPTER 3
The team is aware that in so doing it is constructing reality and not

describing objective reality.
5. Planning an Intervention on the Basis of the Systemic Con

struction of the Problem: Systemic Utilization. In this phase, a
utilization of the problem -form ing effect pattern is put into action.
In so doing, the team is not interested in the reality o f things but in
Wirklichkeit. Here reality means that which is effected—as the G er
man word so precisely describes it—that which has an effect: ef
fect connections, effect patterns, relationship patterns, systemic
processes, and so on. The next section gives a few suggestions for
the construction of systemic utilizations.
B. Suggestions for the Development of Systemic

Utilizations
1. Positive Connotation and Systemic Approach
These are prerequisites for system ic utilization: They imply
that the family will find its own solution to the problem.
2. Redefinition or Redesignation o f the Problem-Forming Effect Patterns
The aim of a redefinition lies in the implicit opening up of
possible solutions or options for the family. We can differenti
ate betw een the following kinds of redefinition: (a) Relabeling
(exchange of labels, changing a name); (b) Recontextualization
(regarding persons, space, time); and (c) Refunctionlization (al
lotm ent of new effect or function of process patterns).
3. Utilization o f the Problem-Forming Effect Patterns
a. Solution-Oriented Prescriptions in Phases o f Simple Co
operation:
Sim ple prescriptions of the problem -form ing effect
patterns as they appear to the therapist.
Prescription of a worsening of the problem-forming ef
fect patterns (regarding frequency, length, intensity).
Prescription of the problem -form ing effect patterns by
modifying the context (compare with 2(b); persons,
space, time).
Prescription of the problem -form ing effect patterns by
division of intention and effect.
Prescription of problem -form ing effect patterns by di
vision into different effects.
Prescription of new, ritualized, formal patterns with
out link to contents.
b. Prescription by Preventing or Reversing Changes in Phases o f

Inverse Cooperation:
Em phasis of negative consequences of change.
Em phasis of the advantages of relating the problem-
form ing effect patterns.
Direct prevention or banning of changes.
Prediction or prescription of a relapse.
Taking back effected changes.
A nnouncing an intervention without ever giving it, so
that the family develops a prescription for itself.
c. Prescriptions by Indirect Utilization:
Problem -referential m etaphors which open up new
options (jokes, proverbs, stories).
M etaphorical (symbolic) tasks (tasks with formal analo
gies to the problem but nonetheless open up op
tions).
Direct or indirect suggestions interspersed at will into
conversation, making solutions possible.
Indirect com m unication, through proxy, w hen som e
body else is meant.
Creating confusion, then offering practical suggestions
that are gratefully accepted to lessen the confusion.
d. Written Interventions Which M ake Use o f the Techniques a .-c .
C. Criteria for the Prognosis of Success of

Therapeutic Utilizations
1. Do the therapeutic team and the family m eet at their corre
sponding levels of evolution (coevolutionary interface)?
2. Has the right contextual extension of the family been chosen?
3. Have the problem -form ing effect patterns been accurately de
scribed?
4. How well or how elegantly has the therapeutic team connoted
the positive problem -form ing effect patterns?
5. How well has the therapeutic team form ulated a redefinition
of the problem -form ing effect pattern implying a solution?
6. How well has the therapeutic team managed:
a. To construct the therapeutic double bind in such a way
that only a positive solution is possible? (This interven
tion does not allow the problem group any com m ent or
leaving the field; it is the equivalent of a discontinuous
change.)
96 CHAPTER 3
b. To construct the therapeutic double binds in such a way

that a continual glide into new possibilities for a solution
are initiated? (This procedure relies on com m ent, i.e .,
reaction from the problem group; it is the equivalent of
continuous change.)
7. Is the utilization close enough to the family's own epistemol-
ogy to be accepted by them?
8. Is the utilization far enough away from the family's own epis-
temology to be regarded as inventive and proficient, that is, as
opening up options?
REFERENCES
Bateson, G. The birth of the double bind. In M. Berger (Ed.) Beyond the
double bind. New York: Brunner/Mazel, 1972.
Bateson, G. Steps to an ecology of mind. New York: Ballantine, 1972. (b)
Bateson, G., Jackson, D. D., Haley, J., & Weakland, J. Toward a theory
of schizophrenia. Behavioral Science, 1956, 1, 251-264.
Berger, M. M. (Ed.). Beyond the double bind: Communication and family
systems, theories, and techniques with schizophrenics. New York:
Brunner/Mazel, 1978.
Deissler, K. G. Anmerkungen zur okosystemischen Sichtweise der
Psychotherapie. Familiendynamik, 1981, 6, 158-175.
Deissler, K. G. Die rekursive Kontextualisierung naturlicher Prozesse.
Familiendynamik, 1983, 8, 139-165. (a)
Deissler, K. G. Rekursive Informationsschopfung. Regeln zur
Entwicklung problemrelevanter Fragen im kokreativen ProzeB der
systemischen familientherapie. In preparation, 1983. (b)
Deissler, K. G., & Gester, P. W. Autosystemic transformation: Trance
phenomena, systemic transformation or utilization in family
therapy. Paper presented at the 2nd International Congress on
Ericksonian Approaches to Hypnosis and Psychotherapy. Phoenix,
Ariz., November 30-December 4, 1983.
Dell, P. F. Beyond Homeostasis: Toward a concept of coherence. Family
Process, 1982, 21, 21-41.
de Shazer, S. Patterns of brief family therapy: An ecosystemic
approach. New York: Guilford, 1982.
Erickson, M. H. The confusion technique in hypnosis. American Journal
of Clinical Hypnosis, 1964, 6, 183-207.
Erickson, M. H., & Rossi, E. L. Varieties of double bind. American
Journal of Clinical Hypnosis, 1975, 17, 143-157.
Erickson, M. H., & Rossi, E. L. Hypnotherapy: An exploratory casebook.

New York: Irvington Publishers, 1979.
Family Process, 1982, 21, 391-434.
Haley, J. The family of the schizophrenic: A model system. Journal of
Nervous and Mental Disease, 1959, 129, 357-374.
Erickson. New York: Ballantine, 1973.
Haley, J. Leaving home: The therapy of disturbed young people. New York:
McGraw-Hill, 1980.
Hoffman, L. Foundations of family therapy: A conceptual framework for
systems change. New York: Basic Books, 1981.
Hoffman, L. A co-evolutionary framework for systemic family therapy.
Unpublished manuscript, 1982.
Jantsch, E. Die Selbstorganisation des Universums. Vom Urknall zum
menschlichen Geist. Miinchen, Germany: dtv, 1982.
Keeney, B. P. Ecosystemic epistemology: Critical implications for the
aesthetics and pragmatics of family therapy. Family Process, 1982,
21, 1-19.
Keeney, B. R Ecosystemic epistemology: An alternate paradigm for
diagnosis. Family Process, 1979, 18, 117-129.
Maturana, H. E Erkennen: Die Organisation und Verkdrperung der
Wirklichkeit. Braunschweig, Germany: Vieweg, 1982.
Pearce, W. B., & Cronen, V. E. Communication, action, and meaning: The
creation of social realities. New York: Praeger Publishers, 1980.
Penn, P. Circular questioning. Family Process, 1982, 21, 267-280.
Rossi, E. L. The collected papers of Milton H. Erickson (Vols. I-IV). New
York: Irvington Publishers, 1980.
Ruesch, J., & Bateson, G. Individual, group and culture: A review of
the theory of human communication. In J. Ruesch & G. Bateson
(Eds.), Communication: The social matrix of psychiatry. New York:
Norton, 1951.
Schultz, I. H. Das Autogene Training. Versuch einer klinisch praktischen
Darstellung. Stuttgart: Thieme, 1966.
Selvini-Palazzoli, M. Why a long interval between sessions? The
therapeutic control of the family-therapist suprasystem. In M.
Andolfi & I. Zwerling (Eds.), Dimensions of Family Therapy. New
York: Guilford, 1980.
Selvini-Palazzoli, M., Boscolo, L., Cecchin, G., & Prata, J. Paradox and
counterparadox. New York: Jason Aronson, 1978.
98 CHAPTER 3
Selvini-Palazzoli, M., Boscolo, L., Cecchin, G., & Prata, J.

Hypothesizing-circularity-neutrality: Three guidelines for the
conductor of the session. Family Process, 1980, 19, 3-12.
Shands, H. C. The war with words. The Haag: Mouton, 1971.
Sluzki, C. E., & Ransom, D. C. (Eds.). Double bind: The foundation of the
communicational approach to the family. New York: Grune & Stratton,
1976.
Spencer-Brown, G. Laws of form. New York: E. P. Dutton, 1979.
Stierlin, H. Familientherapie—Wissenschaft oder Kunst? (Engl.: Family
therapy—science or art?) Opening lecture at the 4th Annual
Meeting of the German Association for Family Therapy (DAF),
Marburg, Germany, October 6-10, 1982. In Familiendynamik. 1983,
8, 364-377.
Szasz, T. The myth of mental illness. New York: Harper & Row, 1974.
Szasz, T. The myth of psychotherapy: Mental healing as religion, rhetoric and
repression. Oxford, Eng.: Oxford University Press, 1979.
Taylor, W. R. Using systems theory to organize confusion. Family
Process, 1979, 18, 479-488.
Thomas, K. T. Praxis der Selbsthypnose des Autogenen Trainings. Stuttgart:
Thieme, 1972.
Varela, F. A calculus for self reference. International Journal of General
Systems, 1975, 2, 5-24.
Watzlawick, P., Beavin, J., & Jackson, D. D. Pragmatics of human
communication. New York: Norton, 1967.
Weeks, G. R., & L'Abate, L. Paradoxical psychotherapy: Theory and
practice with individuals, couples, and families. New York: Brunner/
Mazel, 1982.
4
Paradoxical Intention
b y Viktor E. Frankl. M.D.. Ph.D.
The purpose of the following chapter is to present the details of

the logotherapy position regarding the use of paradox in therapy.
Although the material will not be new in the sense of offering
previously unpublished facts, it is our intention to provide a de
finitive statem ent regarding this therapeutic issue based upon a
review of the significant logotherapy literature in the area. A dis
cussion of relevant research will not be undertaken in this report
as the inform ation is available elsewhere (Solyom et al., 1972), in
addition to Chapter 8.
Paradox in logotherapy is largely subsum ed under "paradoxical
intention," a technique developed by Frankl in 1929 at the psychi
atric hospital of the University of Vienna M edical School and first
discussed by him in a Sw iss neuropsychiatric journal (Frankl,
1939). Frankl coined the term paradoxical intention in 1947 in a book
that he published in Germ an (Frankl, 1947).
Frankl explains this technique's therapeutic effectiveness by re
ferring to the phenom enon know n as anticipatory anxiety.* The
*Editors note: Logotherapy has been extensively used in the treatment of obsessions
and compulsion, phobias, and sexual neuroses. However, it can be used in the treatment
of many other disorders.
99
100 CHAPTER 4
erythrophobic individual, for example, who is afraid of blushing

w hen he enters a room and faces a group of people, will actually
blush at precisely that moment. A given symptom in a patient
evokes a phobia in the form of "fearful expectation" of its recur
rence; this phobia actually provokes the sym ptom 's recurrence;
and the recurrence of the symptom reinforces the phobia (Figure
D-
FIGURE 1
The First Circle

Formation: Phobias
In order to demonstrate on what theoretical grounds the prac

tice of paradoxical intention had been based, I would like to quote
the following passage from the book I wrote 37 years ago.
All psychoanalytically oriented psychotherapies are mainly con
cerned with uncovering the primary conditions of the "conditioned
reflex" as which neurosis may well be understood, namely, the sit
uation-outer or inner—in which a given neurotic symptom
emerged the first time. It is this author's contention, however, that
the fullfledged neurosis is caused not only by the primary condi
tions but also by secondary conditioning. This reinforcement, in
turn, is caused by the feedback mechanism called anticipatory anx
iety. Therefore, if we wish to recondition a conditioned reflex, we
must unhinge the vicious cycle formed by anticipatory anxiety, and
this is the very job done by our paradoxical intention technique.
(Frankl, 1947, p. 91)
There are cases in which the object of the fearful expectation is
fear itself. Our patients spontaneously speak of a "fear of fear."
Upon closer interrogation it turns out that they are afraid of the
consequences of their fear: fainting, coronaries, or strokes. As I
pointed out earlier, they react to their fear of fear by a "fligh t from
PARADOXICAL INTENTION 101
fear" (Frankl, 1953). In 1960 I had arrived at the conviction that

"Phobias are partially due to the endeavour to avoid the situation
in which anxiety arises" (Frankl, 1960).
Along with the phobic pattern which we may describe as a
flight from fear, there is another pattern, the obsessive-com pul
sive one, characterized by what one may call a "fig h t against ob
sessions and compulsions." Patients are afraid they might commit suicide1
or homicide or that the strange ideas haunting them might be the
precursors—if not symptoms—of a psychosis. In other words, they are
not afraid of fear but rather of themselves.
Again, a circle form ation is established. The more the patient
fights his obsessions and com pulsions, the stronger they becom e.
Pressure induces counterpressure, and counterpressure in turn
increases pressure (Figure 2).
FIGURE 2
The Second Circle Formation:

Obsessions and Compulsions
1Perhaps a special comment should be made with respect to the application of
paradoxical intention to situations involving suicidal ideas and impulses. It can

not be too strongly stressed that it is to be used only when suicide is the content
of a true obsession being resisted (and reinforced by this resistance) by the patient.
In a situation where a patient is prone to identify himself with the suicidal impulse
(as may be the case in endogenous depression), paradoxical intention will serve
to increase the danger and is therefore absolutely contraindicated. Let me sum
marize: Paradoxical intention should not be applied in psychotic depressions, but
only in cases in which an individual is haunted by the obsessive idea that he
might attempt suicide. The obsessive individual doesn't wish to commit suicide,
but rather fears to do so. That is the type of patient whom the therapist may
encourage (ironically of course) to attempt suicide.
102 CHAPTER 4
In order to unhinge all the vicious circles discussed, the first thing to do
is take the wind out of the anticipatory anxieties underlying them, and
this is precisely the work of paradoxical intention. It may be defined as a
procedure in whose framework patients are encouraged to do, or wish for,
the very things they fear—albeit with tongue in cheek. In fact, an integral
element in paradoxical intention is the deliberate evocation of humor, as
Lazarus (1971) justifiably points out. After all, a sense of humor is one of
the specifically human capacities, namely, the capacity of "self-detachm
ent" (Frankl, 1966). No other animal is capable of laughter.2
In paradoxical intention, the pathogenic fear is replaced by a

paradoxical wish, thereby unhinging the vicious circle of anticipa
tory anxiety. We will cite the following case as an example. A
woman, M ary B ., had undergone various m ethods of treatm ent
for 11 years yet her com plaints, rather than decreasing, had in
creased. She suffered from attacks of palpitation accom panied by
marked anxiety and anticipatory fears of a sudden collapse. After
the first attack, she began to fear it would recur and consequently
it did. The patient reported that whenever she had this fear, it
was followed by palpitations. Her chief concern, however, was
that she m ight collapse in the street. A fter her adm ission to the
neurological departm ent of the Polyclinic Hospital, Dr. Kurt Ko-
courek advised her to tell herself at such a m om ent, "M y heart
shall beat still faster! I would ju st love to collapse right here on the
sidew alk!" Furtherm ore, the patient was advised deliberately to
seek out places which she had experienced as disagreeable, or
even dangerous, instead of avoiding them . Two weeks later the
patient reported, " I am quite well now and feel scarcely any palpi
tations. The fear has com pletely disappeared." Som e weeks after
her discharge she reported, "O ccasionally mild palpitations oc
cur, but w hen they do I say to m yself: 'M y heart should beat even
faster,' and at that moment the palpitations cease."
Another woman, treated by Dr. M ichael Ascher, com plained of
a fear of contracting genital herpes following the appearance of a
feature article on this topic in a national m agazine. Sh e was a 32-
year-old secretary in a large law firm and lived by herself. She
began wearing gloves from her departure from hom e in the m orn
ing to her return in the evening. She stopped eating lunch with
her friends and ate at her desk, only consum ing food brought
2Hand et al. (1974), who treated chronic agoraphobia patients in groups, observed
that they spontaneously used humor as an impressive coping device: "When the whole
group was frightened, somebody would break the ice with a joke, which would be
greeted with the laughter of relief" (pp. 588-602). One might say these patients
reinvented paradoxical intention.
from hom e in sealed pouches, and only w hen she had thoroughly
cleansed and prepared the area. Her time outside of work was
spent mostly in her apartm ent with a few close friends or her par
ents. She never perm itted anybody to enter her apartm ent, and
never wore clothes in her hom e that she had worn outside; these
"contam in ated " articles were kept in a special closet. Sh e con
sulted a therapist w hen she began carrying a sheet which she
used to cover any chair she sat on outside of her apartm ent. Fi
nally, the client was encouraged to remove her gloves, throw the
sheet upon which she was sitting into a corner, and try to "ca tch "
herpes. The therapist m odeled and the client initiated "h erp es-
catching" behavior, that is, sitting on all chairs, touching various
places in the office (desks and various places on the floor), touch
ing her face and body with her ungloved, "co n tam in ated " hands.
Throughout the session, which was quite lengthy, the therapist
and eventually the client joked about their activities, about the
client's concern regarding catching herpes, and about herpes in
general (it was quite funny; the reader should have been there).
Naturally, the client was reluctant to participate at the beginning
of the exercise, but by its conclusion modeling was unnecessary
for she was initiating the required behavior. At the conclusion of
the session, the therapist suggested that the client continue to try
to "c a tc h " herpes throughout the next week. That is, she was not
to use her gloves or sheet; she was to wear the "contam in ated "
clothes throughout her apartm ent, use public lavatories whenever
possible, eat lunch with her friends in local restaurants, and in
general, do everything possible to "c a tc h " the disease. U pon her
return the following week, she reported that she had carried out
the therapist's instructions and felt a good deal better, but was
" s a d " to report that she had not contracted herpes. In fact, she
still had periods of discom fort and retained som e ritualistic behav
ior. However, she was com pletely com fortable in the therapist's
office and further "contam in ation" there was unnecessary. She
w as instructed to continue to try to catch herpes during the week
prior to her third session. W hen she returned, she reported that
she felt better than she had in m any years and that no rituals re
mained. Therapy continued with a focus on other issues; her con
cerns regarding contracting genital herpes did not reappear at a
one-year follow up.
Paradoxical intention may be effective even in severe cases. Let
us turn to an illustrative case history:
Mrs. Elfriede G., a 35-year-old woman, was a patient in the Neuro

logical Department of the Polyclinic Hospital when I presented her
104 CHAPTER 4
at one of my clinical lectures. She reported that as a child she was

meticulous and while her friends were playing in the park, she
stayed at home scrubbing and cleaning. For three years, the patient
had been virtually incapacitated by an extreme fear of bacteria;
hundreds of times a day she had washed her hands. Fearing con
tact with germs, she no longer left the house; fearing exposure
through outsiders, she excluded all visitors. She would not even
allow her husband to touch the children for fear he would transmit
germs to them. Finally she wanted a divorce because she felt she
had made her family unhappy.
She had been institutionalized because of several attempts at
suicide. After unsuccessful treatment in various clinics and hospi
tals, she had finally been taken to the neurological department of
the Polyclinic Hospital by ambulance, for she had become com
pletely helpless. In the lecture hall of the hospital, in the presence
of a class of students, I spoke to the patient for the first time. I
asked her: “Are you accustomed to check the door many times be
fore leaving home, or to check whether a letter has really fallen into
the mailbox or not, or to check several times whether the gas valve
is really closed before going to bed?" "Yes, that is my case/' she
said anxiously. I then proceeded by pointing out that this meant
she belonged to a certain type of character structure which in tradi
tional European psychiatry was conceived of as "anankastic/' and
that this meant immunity to psychoses. A sigh of relief was her
response, relief after long years of suffering from the fear of becom
ing psychotic.
Because of her fear that the obsessions had been psychotic
symptoms, the patient had fought them. By this very counterpres
sure, however, she had increased the pressure within herself. I
then remarked: "You have no reason for such a fear. Any normal
person can become psychotic, with the single exception of people
who are anankastic character types. I cannot help but tell you this
and destroy all your illusions in this respect. Therefore you need
not fight your obsessive ideas. You may as well joke with them."
Then I started paradoxical intention. I invited the patient to imitate
what I did. I scrubbed the floor of the lecture hall with my hands
and said: "After all, for the sake of a change, now, instead of
fearing infection, let's invite it." Stooping and rubbing my hands
on the floor, I continued: "See, I cannot get dirty enough; I can't
find enough bacteria!" Under my encouragement, the patient fol
lowed my example, and thus she began the treatment which, in
five days, removed ninety percent of her symptoms.
An incapacitating pattern of three years' standing was broker,
up in a matter of a few weeks. She spoke jokingly of all her former
symptoms. She asked her fellow patients whether any of them
could provide her with "some more bacteria." She cleaned postop
erative patients in the laryngological department. She was in
steady contact with bloody things. She washed her hands only
three times a day although she frequently handled putrid material.

She wanted "to make as much acquaintance with germs as possi
ble." She spontaneously declared: "I want to let the poor beings
live and not wash them away." On the sixth day she left the hospi
tal to buy wool to knit a pullover for her youngest child, to knit it
"here in an environment full of bacteria." "O n each loop of the
sweater," she said humorously, she wanted "one bacterium sit
ting." She was beaming with joy and felt completely healthy. When
she went home for Christmas her behavior was, for the first time,
as normal as it had been before the onset of her neurosis three
years before.
There was no longer any need to apply paradoxical intention.
The patient embraced the children, caressed them without the
slightest fear of infecting them. She resolved deliberately: "Now I
will transfer the bacteria onto my children." The washing compul
sion had disappeared. "I am the happiest person on earth," she
declared. She was able to do everything in the normal routine in
cluding tasks that formerly she had not been able to finish. She did
all of the housework and devoted herself to her children as she had
not been able to do since her neurosis became full-fledged. She
was able to devote herself to her youngest child for the first time in
his life! Some time later I asked her about the washing compulsion.
She replied: "I have to laugh at that now. It seems quite unreal to
me that I ever had to suffer from anything like that. Now at ten
o'clock in the morning my housework is finished. Before, I got up
at 3 o'clock in the morning and even by night my housework was
not completed." (Frankl, 1975, pp. 226-237)
Compare this type of intervention with the technique of symp

tom prescription. Sym ptom prescription and paradoxical inten
tion are two different things. W hen I apply sym ptom prescrip
tion, I want the patient to increase, say, anxiety. W hen I use para
doxical intention, however, I want the patient to do, or wish for,
that whereof h e is afraid. In other words, not fear itself but rather
its object is dealt with. Let me invoke a case published by the logo-
therapist Byung-Hak Ko (1981), a professor at the National Uni
versity of Korea. The patient had been suffering from fear of
death. Treating him by paradoxical intention, the psychiatrist did
in no way recommend him to increase the thanatophobia but, to
quote from the paper, the respective instruction read: "Try to be
more dizzy, have faster palpitations, and choke more. Try to die in
front of the people." And the next time, in fact, the patient en
tered the psychiatrist's "office cheerfully and reported success."
Or, to quote another example to demonstrate the difference be
tween paradoxical intention and sym ptom prescription: A profes
sor at the University of Nebraska, discussing paradoxical inten
106 CHAPTER 4
tion (in an unpublished paper), explained it as follows: "F o r ex

ample, a person who has an obsession to wash his hands ten
times a day will be invited to do so 30 tim es a day." I would say
such a procedure is unequivocally representative of sym ptom pre
scription: The patient was advised to exaggerate the symptom!
Along the lines of this technique, M rs. Elfriede G ., w hose case
history I presented above, would have been advised to wash her
hands not "hundreds of times a d ay " but rather thousands of
times a day; in other words, to exaggerate her bacteriophobia. In
stead, however, an attem pt was made—and, as we saw, success
fully—at converting her fear of germs into the ironical wish "to
make as much acquaintance with germs as possible." In contrast
to symptom prescription, paradoxical intention does not focus on
the symptom and its increase but rather on the fear underlying it
and—rather than its increase—its inversion.
The paradoxical intention technique lends itself to the treat
m ent of sleeplessness. The fear of sleeplessness results in a hy
perintention to sleep w hich prevents the patient from sleeping
since sleep presupposes the utm ost of relaxation. But how can
one remove the anticipatory anxiety that is the pathological basis
of hyperintention and brings about a vicious circle that increases
the disorder? The hyperintention to fall asleep m ust be replaced
by the paradoxical intention to stay awake.
In order to take the wind out of the sails of his fear of sleepless
ness, we advise the patient not to try to force sleep, since his
body will provide the m inimum am ount of sleep it really needs by
itself. Therefore, he can safely try to do ju st the opposite—stay
awake as long as possible. In other words, the hyperintention to
fall asleep, arising from the anticipatory anxiety of not being able
to fall asleep, should be replaced by the paradoxical intention not
to fall asleep, which soon will be followed by sleep.
The following report was written by a student at Duquesne
University and also illustrates the use of paradoxical intention.
For 17 years I stuttered very severely; at times I could not speak at

all. I saw many speech therapists, but had no success. One of my
instructors assigned your book, Man's Search for Meaning, to be read
for a course. So, I read the book and I decided to try paradoxical
intention by myself. The very first time I tried it, it worked fabu
lously—no stuttering. I then sought out other situations in which I
would normally stutter, and I applied paradoxical intention and it
successfully alleviated stuttering in those situations. There were a
couple of situations thereafter when I did not use paradoxical in
tention—and the stuttering quickly returned. This is a definite
proof that the alleviation of my stuttering problem was due to the
effective use of paradoxical intention. (Frankl, 1978, pp. 129-130.)
The principle underlying paradoxical intention is sometimes

used unwittingly and even unwillingly. This story concerns a cli
ent of my former student Uriel M eshoulam of Harvard University,
who reported it to me as follows:
The patient was called to the Australian army, and was sure he
would avoid the draft because of his stuttering. To make a long
story short, he tried three times to demonstrate his speech diffi
culty to the doctor, but could not. Ironically, he was released on
grounds of high blood pressure. The Australian army probably
does not believe him until today that he is a stutterer.
O f the three pathogenic patterns distinguished by logotherapy,

two have now been discussed: the phobic pattern, characterized
by flight from fear, and the obsessive-com pulsive pattern, distin
guished by the fight against obsessions and com pulsions. W hat
then is the third pattern? It is the sexual neurotic pattern, which
again is characterized by the patient's fight. Here, however, the
patient is not fighting against anything, but rather fo r sexual plea
sure: It is a tenet of logotherapy that the more one aims at plea
sure the more one m isses it.
W henever potency and orgasm are made a target of intention,
they are also made a target of attention (Frankl, 1952). In lo
gotherapy the term s we use are hyperintention and hyperreflection
(Frankl, 1962). The two phenom ena reinforce each other so that a
feedback m echanism is established. In order to secure potency
and orgasm, the patient pays attention to him self, to his own per
formance and experience. To the same extent, attention is with
drawn from the partner and whatever the partner has to offer in
terms of stimuli that m ight arouse the patient sexually. A s a con
sequence, potency and orgasm are in fact dim inished. This in turn
enhances the patient's hyperintention and the vicious circle is com
pleted (Figure 3).
In order to break it up, centrifugal forces m ust be brought into
play. Hyperreflection can be counteracted by the logotherapeutic
technique called dereflection (Frankl, 1955), that is to say, the pa
tients, instead of watching them selves, should forget themselves.
But they cannot forget them selves unless they give them selves.
Again and again, it turns out that the hyperintention of sexual
performance and experience is due to the patient's sexual achieve
m ent orientation and tendency to attach to sexual intercourse a
"dem and quality." To remove it is the very purpose of a logothera
peutic strategy that (in addition to the dereflection technique) I
described first in G erm an (Frankl, 1946). To illustrate the logother
apeutic approach to sexual neurosis, let me quote from the first
pertinent publication in English (Frankl, 1952):
108 CHAPTER 4
FIGURE 3
The Third Circle Formation:

Sexual Dysfunctions
The following trick was devised to remove the demand placed on

the patient by his partner. We advise the patient to inform his part
ner that he consulted a doctor about his difficulty, who said that his
case was not serious, and the prognosis favorable. Most important,
however, is that he tell his partner that the doctor also has abso
lutely forbidden coitus. His partner now expects no sexual activity
and the patient is "released." Through this release from the de
mands of his partner it is possible for his sexuality to be expressed
again, undisturbed and unblocked by the feeling that something is
demanded or expected from him. Often, in fact, his partner is not
only surprised when the potency of the man becomes apparent but
she goes so far as to reject him because of the doctor's orders.
When the patient has no other goal before him than a mutual sex
ual play of tenderness, then, and then only, in the process of such
play is the vicious circle broken, (pp. 129-130)
The report from which I am now going to quote concerns a
patient of mine suffering from frigidity (Frankl, 1962). The tech
nique used in treating this case was dereflection.
The patient, a young woman, came to me complaining of being
frigid. The case history showed that in her childhood she had been
sexually abused by her father. However, it was not this traumatic
experience in itself that had eventuated in her sexual neurosis. It
turned out that, as a result of reading popular psychoanalytic litera

ture, the patient lived all the time in fearful expectation of the toll
that her traumatic experience would someday take. This anticipa
tory anxiety resulted in both excessive intention to confirm her
femininity and excessive attention centered upon herself rather
than upon her partner. This was enough to incapacitate the patient
for the peak experience of sexual pleasure, since the orgasm was
made an object of intention and an object of attention as well. Al
though I knew that short-term logotherapy would do, I deliberately
told her that she had to be put on a waiting list for a couple of
months. For the time being, however, she should no longer be con
cerned about whether or not she was capable of orgasm, but
should concentrate on her partner, better to say whatever made
him lovable in her eyes. "Just promise me that you won't give a
damn for orgasm," I asked her. "This we'll take up discussing only
after a couple of months, when I start treating you." What I had
anticipated happened after a couple of days, not to say nights. She
returned to report that, for the first time not caring for orgasm, she
had experienced it the first time. (pp. 194-195)
Claude Farris once treated another type of sexual neurosis and
did not apply dereflection but rather paradoxical intention:
Mr. and Mrs. Y. were referred to me by Mrs. Y.'s gynecologist. Mrs.

Y. was experiencing pain during intercourse. Mr. and Mrs. Y. had
been married for three years and indicated that this had been a
problem from the beginning of their marriage. Mrs. Y. had been
raised in a Catholic convent by sisters, and sex was a taboo subject.
I then instructed her in paradoxical intention. She was instructed
not to try to relax her genital area but to actually tighten it as tight
as possible and to try to make it impossible for her husband to
penetrate her and he was instructed to try as hard as he could to
get in. They returned after one week and reported that they had
followed instructions and had enjoyed painless intercourse for the
very first time. Three more weekly sessions indicated no return
of the symptoms. Paradoxical intention has proved effective in
many cases in my experience, and at times almost works me out of
business.
The logotherapeutic technique of paradoxical intention lends it
self to the treatm ent of a variety of neurotic conditions, in particu
lar, those in which anticipatory anxiety (nowadays often referred
to as performance anxiety) plays a decisive role as a pathogenic
factor.
Last but not least, I would like to em phasize that opportunity
remains for both therapist and patient to develop creative modifi
cations of this technique: Such accom m odations to a given case
are even mandatory. For not only in logotherapy, but in psycho-
110 CHAPTER 4
therapy in general, it is imperative to individualize according to

the person with whom the therapist is confronted, and to impro
vise according to the situation with which the patient is faced.
REFERENCES
Frankl, V. E. Zur medikamentosen Unterstutzung der Psychotherapie
bei Neurosen. Schweizer Archiv fur Neurologie und Psychiatrie, 1939,
43.
Frankl, V. E. Arztliche Seelsorge. Vienna: Deuticke, 1946.
Frankl, V. E. Die Psychotherapie in der Praxis. Vienna: Deuticke, 1947.
Frankl, V. E. The pleasure principle and sexual neurosis. International
Journal of Sexology, 1952, 5.
Frankl, V. E. Angst und Zwang. Acta Psychotherapeutica, 1953, 1,
111 - 120 .
Frankl, V. E. The doctor and the soul: From psychotherapy to logotherapy.
New York: Knopf, 1955.
Frankl, V. E. Paradoxical intention: A logotherapeutic technique.
American Journal of Psychotherapy, 1960, 14, 520-535.
Frankl, V. E. Man's search for meaning: An introduction to logotherapy.
Boston: Beacon Press, 1962.
Frankl, V. E. Logotherapy and existential analysis: A review. American
Journal of Psychotherapy, 1966, 20, 552-560.
Frankl, V. E. Paradoxical intention and dereflection. Psychotherapy:
Theory, Research and Practice, 1975, 12.
Frankl, V. E. The unheard cry for meaning: Psychotherapy and humanism.
New York: Simon & Schuster, 1978.
Hand, I., Lamontagne, Y., & Marks, I. M. Group exposure in vivo for
agoraphobics. British Journal of Psychiatry, 1974, 14.
Ko, Byung-Hak. Application in Korea. The International Forum for
Logotherapy, 1981, 4, 89-93.
Lazarus, A. A. Behavior therapy and beyond. New York: McGraw-Hill,
1971.
Solyom, L., Garza-Perez, J., Ledwidge, B. L., & Solyom, C.
Paradoxical intention in the treatment of obsessive thoughts: A
pilot study. Comprehensive Psychiatry, 1972, 13.
5
Paradoxical Techniques:
One Level of Abstraction in
Family Therapy
by Luciano L'Abate, Ph.D.
This chapter presents paradoxical techniques as one level of ab

straction am ong at least four different therapeutic approaches.
Paradoxical techniques are also the first step in the process of
therapy. They function to reduce stress and conflict, provide
symptomatic relief and crisis reduction, and engage the family in
the process of learning to trust the therapist enough to find new
ways to negotiate problem s.
HISTORICAL BACKGROUND
The m ajor em phasis in a paradoxical approach to therapy stems
from a variety of sources (L'Abate, 1969). Those who have most
influenced a circular approach (in this context, circular is synony
mous with paradoxical) are Dewey and Bentley (1949), Ruesch and
Bateson1 (1951), and von Bertalanffy (1968). Incidentally, Rychlak
1 I consider J. Ruesch and A. F. Bentley the two unsung pioneers of how general
systems theory preceded family therapy theory. Perhaps someone will someday trace
their contribution more fully than I can.
111
112 CHAPTER 5
(1968) is one whom I credit with clarifying my thinking on dialec

tical versus demonstrative approaches. O ther influences are so
heterogeneous as to defy pinpointing.
M y particular position can be called positivistic in that I use a
constructive, positive approach in response to the negatives pre
sented at the outset by most families (L'Abate & Kearns, submitted
for publication). O ne could say that my thinking has been in
fluenced by families themselves, as they (mostly children in my
original practice of clinical child psychology) presented their refer
ral problems (L'Abate, 1973). Som e of my thinking has of course
been influenced by Rogers (1951)—positive regard; Ginott (1965)—
intimacy; and the Bible—specifically the "Serm o n on the M ount"
paradox.
Most clients responsible for my conceptualization of a paradox
ical approach (L'Abate, 1975) were seen in the context of my part-
time private practice, my supervision of graduate students in a
university clinic, and in my supervision of private practitioners
seeking further professional training to qualify for AAMFT m em
bership. Consequently, the cases subm itted to me by supervisees
or seen directly range from schizophrenics to normals. The most
common symptoms in my practice have been school under
achievement, learning disabilities, lack of motivation in studying
(L'Abate & Baggett, in press), and a variety of family problems
that can be subsumed under family depression (L'Abate, in
press). As a whole, my practice has not included extreme cases of
drug abuse (though I have dealt with some), alcoholism, or ex
treme psychiatric pathologies.
My thinking on paradox came about through dealing with the
original referral (L'Abate, 1975, pp. 63-75). I becam e literally sick
and tired of listening to families scapegoating via the child's un
derachievement, lack of motivation, learning problems, and the
whole host of negatives that were practically overwhelming the
child. So, over time, I found m yself shifting more and more in the
direction of contradicting the family about the reason for referral,
essentially finding a positive reason (love and care) instead of ac
cepting the family's negatives. I began to seek positive reasons for
the referral which I generaled to the whole of the negativity
(L'Abate & Kearns, submitted for publication) that families bring
with them to therapy sessions. In counterposition to the family's
negativity, I learned to develop a positive stance basically the
same as what various authors have called positive reframing, with
the same characteristic—a positive rather than a negative outlook
on family functioning (L'Abate, 1975). Because I felt effective in
using a positive approach to the referral symptom, I generalized
PARADOXICAL TECHNIQUES 113
this approach to anything and everything negative that a family

or a client brought to my office (Weeks, 1977).
Regarding the specifics of paradox, I found that paradoxical
techniques fit within a framework of therapy consisting of at least
four different levels of abstraction: (a) structural, (b) linear, (c) cir
cular, and (d) metaphorical. The paradox stem s first from an ap
preciation of the variety of levels of abstractions in therapy and a
dialectical view of behavior.
The structural approach, best represented by M inuchin (1974),
is understood to mean dealing with nonverbal aspects of family
living, that is physical arrangements for sleeping, eating, chores,
responsibilities, and routines. This is clearly the m ost concrete,
probably the primary, level at which a therapist needs to inter
vene. The linear approach is the straightforward one used by the
majority of family therapists. It consists of a gradual, step-by-step
sequence of confrontations, interpretations, homework assign
m ents, and exercises (L'Abate, subm itted for publication 1983 [a]),
all of which derive from a gradual, step-by-step view of behavior.
Circular approaches consist of more tangentially indirect, cryp
tically paradoxical reframing of symptomatic behavior in positive
terms. They are best illustrated by the work of Weeks and L'Abate
(1982), M adanes (1981), and de Shazer (1982), which will be eluci
dated in this chapter. M etaphoric approaches are perhaps best il
lustrated by the work of Carl W hitaker (Neill & K niskem , 1982)
and some members of the existential school (Kempler, 1981). They
are at the highest level of abstraction because they extract dim en
sions and issues from behaviors that are not clearly evident to the
naked eye and ear. They cannot be classified as either linear or
circular because they function in a com pletely different modal
ity—the metaphor. Hence, they need to be classified under a sepa
rate heading.
WHAT IS PARADOXICAL
PSYCHOTHERAPY?
The therapeutic paradox consists of (a) a variety of ways of in
tervening without appearing to intervene and of helping (without
appearing to control) a family learn to control what they are do
ing; and (b) contradicting a family's ideologies and mythologies
without appearing to contradict. The paradox, understood to be a
contradiction, lies in the therapist's contradicting what the family
does without appearing to contradict (lest the family feel dis
114 CHAPTER 5
counted). Therefore, the paradoxical stance supports a family as a

group of individual hum an beings; the therapist never criticizes or
judges what they do. The therapist needs to help them break dys
functional patterns of behavior without increasing their feelings
of inadequacy to the point that they flee from treatm ent, and
avoid confrontations that increase either "resistan ce" or feelings
of inadequacy to the point that they discontinue treatm ent.
The theory that underlies paradox is based on a view that be
havior is inherently variable—intra- as well as interpersonally. The
many facets of variability can be condensed into two aphorism s or
truisms: "O n e person's trash is another's treasure," which speaks
to interpersonal variability; and "O u r assets can be our worst lia
bilities," which speaks to intrapersonal variability. O nce we accept
the inherent variability of hum an existence, we can accept the
idea that psychopathology consists of wide swings from one ex
treme of contradiction to the other within and am ong individuals
in the family. The goal of therapy is to decrease such variability,
and to decrease contradictions and inconsistencies as m uch as
possible (the goal of reducing variability will, of course, be elabo
rated further).
The paradox can be an approach, a technique, or a m ethod,
depending on its replicability. As an approach, paradox is a meta-
theoretical view, one that relates mostly to variability: It repre
sents a philosophical position about hum an nature. Basically, a
technique is peculiar to the individual and relates to the style of the
therapist. A method can be replicated, regardless of the particular
style of the therapist (Haley, 1976; H ansen & L'Abate, 1982;
L'Abate et al., in press).
The intended effects of a paradox are of course to help a family
change by decreasing either the anxiety level (if you prefer, intra
psychic concepts) or the variability and turmoil that exist among
family m embers, helping them find new ways to relate after the
turmoil and crisis have been reduced (L'Abate, subm itted for publi
cation 1983[a]). Obviously, this goal is achieved in a variety of ways;
the paradox appears in as many forms as do other treatments.
HEALTH AND DYSFUNCTION

M y theory of family pathology has evolved over a period of
time (L'Abate, 1964, 1976, 1983 [a]) and is currently com posed of
four m ajor aspects: (a) styles, (b) com ponents, (c) levels, and (d)
power and negotiation.
Styles
The three styles in intim ate in terp erson al relationship s
(L'Abate, in preparation 1983 [b]) pertain to apathy (A), reactivity (R), and
conductivity (Q ).
Apathy consists of all kinds of behaviors that pertain to
abuses—atrophied, aggressive, avoidant, aversive—that is, de
structive relationships. Reactive relationships represent the polari
ties of negative complementarity, in which the dialectic is same-
opposite and of which repetition is the m ain characteristic: yes-
no, black-white, true-false, right-wrong—immediate reactions of
one individual to w hat another says or does. Reactive styles
characterize a great m any husband/wife, parent/child intimate
(close and prolonged) transactions. O ne individual reacts to the
other. Then, because the second individual reacts to the first,
counterreactions and escalations ensue.
The conductive style represents com m itm ent to constructive
change through change of self. It represents the ability to delay
reacting until m uch more inform ation (both internal and external)
is obtained and processed and a response can be constructive and
positive. Clearly, psychopathology is present in all abusive, apa
thetic, avoidant relationships. A slightly lesser degree of pathol
ogy (but still dysfunctionality) is present in reactive relationships,
while "h e a lth " is found in conductive relationships.
Components
Interpersonal relationships can be represented through a circu
lar inform ation-processing model based on five com ponents:
Emotionality, Rationality, Activity, Awareness, and Context.
These five term s stand for, in traditional psychological language,
structure, process, outcome, correction m echanism , and context.
In inform ation-processing language, these com ponents would be
called input, throughput, output, feedback, and context. These
five com ponents perform a variety of classificatory and diagnostic
functions, which have been elucidated elsewhere (L'Abate, 1981;
L'Abate & Frey, 1981; L'Abate et al., 1982; Ulrici et al., 1981).
This model indicates that, for health, all five com ponents need
to be present on an equal basis. Pathology represents an overre
liance on one of the five at the expense of the others. For instance,
an overreliance on em otionality may bring about a decrease in ra
tionality, producing a sequence that goes from em otionality to ac
tivity, thereby bypassing and short-circuiting rationality. By the
116 CHAPTER 5
same token, an overreliance on rationality may delay activity and

push emotionality into a secondary or tertiary position, one in
which emotionality is not used as a resource but is seen instead as
a liability. The same applies to awareness, defined as reflection on
one's behavior. The context could be denied, overemphasized, or
m isinterpreted. Balance among all the com ponents and equal,
flexible use bring about health; overreliance on one at the expense
of the others brings about pathology.
Levels
The third aspect of this theory consists of levels and patterns of
congruence in functioning (discussed in greater detail in L'Abate,
1964, 1976, 1983 [a]). There are two levels of interpretation—the
descriptive and the explanatory. The descriptive level is com posed
of two sublevels: The self-presentational, which refers to the fa
cade or public image; and the phenotypical, which describes how
we actually behave in our more private m om ents. The explanatory
level also consists of two sublevels: the genotypical, which refers
to the underlying, attributionally inferred, abstract constructs that
we use to explain (redundantly, one might add) individual or sys
temic behavior (e.g., anxiety, guilt, self-esteem ); and the histori
cal, which is the generational, life-cycle view of how behavior
came to be. Health represents a certain degree of consistency
among all levels; that is, one level is not overemphasized at the
expense of another.
Power and Negotiation

The fourth aspect of this theory considers power sharing and
negotiation. The power-sharing model defines power as who
makes the decisions and responsibility as who carries them out.
The two levels of task assignment are orchestration (i.e., large de
cisions, such as moving, changing jobs) and instrum entation (i.e.,
the picky, everyday decisions). Negotiation Potential (NP) is
based on (a) the degree of functionality (ill-health), (b) the motiva
tion to negotiate (will), and (c) the ability to negotiate (shall). This
particular aspect of the model is summarized by a formula: NP =
ill x skill x will. The negotiation model consists of the 'having,'
'being,' and 'doing' model (L'Abate et al., 1980), which speaks to
a family's modalities of living (what is to be negotiated). The con
tent of negotiation can be based on services and information (do
ing), goods and m oney (having), or love and status (being). C on
flicts in doing or having are pseudo issues that distract us from

our inability to deal with issues of being.
Theory of Personality
Health and pathology are based on the differentiation of rela
tionships. Change and growth result from our ability to be con
ductive. Change is based on three multiplicative requisites: (a) do
ing som ething positive and constructive, (b) doing som ething new
and different from what one has done in the past, and (c) doing it
frequently or intensely enough for it to withstand the impulse to
maintain the status quo. The three requisites of change, then, are
positivity, differentness, and strength. We fail to grow w hen we
behave abusively or reactively; that is, w hen we continue to do
either the same as or the opposite of the other persons in the fam
ily. In conductive relationships, two pluses equal a multiplication
(growth) in the relationship. In reactive relationships, there is one
plus and one minus (sam e-opposite), which equal zero and thus
repetition of the same relationship over time. In abusive relation
ships, two m inuses divide or split energy to the point of destruc
tion (L'Abate, 1976).
More helpful than the view of health and pathology as a con
tinuum is the bell-shaped curve view. That is, health is in the mid
dle of the curve with pathology on both its sides (one side being
too much and the other too little; one too frequent and the other
too infrequent; one too intense and the other not intense
enough). Health thus represents the middle of the road on the
four com ponents previously described (L'Abate, 1964, 1976;
L'Abate & Kochalka, in press).
Health is properly represented by the whole concept of conduc
tivity; it is our ability to be com m itted to change, to be creative
and constructive. Normalcy would be properly represented by
reactivity because, if we consider normalcy just a statistical con
cept of frequency, it appears that m ost hum an relationships are
reactive. Normal, however, would be neither health nor dysfunc
tion; dysfunction is represented by the apathetic relationships al
ready described.
Sym ptom s are usually produced by generational patterns in
which an inability to express hurt in conductive, appropriate, con
structive ways brings about pathology (L'Abate, 1977; L'Abate et
al., 1979). Sym ptom s then are produced by our inability to ex
press properly, constructively, and conductively our genotypic
feelings of hurt, despair, grief, loneliness, inadequacy, helpless
118 CHAPTER 5
ness, vulnerability, fallibility, and neediness, and by our inability

to resolve these feelings by sharing them with our intim ates, who
in turn reassure us of our w orth and our im portance to them
(L'Abate & Sloan, in press, [a], [b]). O f course, we first have to be
convinced of our own inherent w orth as hum an beings and aware
of our need to let those we love know about our hurts and our
fears of being hurt.
A sym ptom is any behavior that over time brings pain, hurt,
and grief to the family and decreases the fam ily's functioning to
the point of increasing dependence on outside others (L'Abate &
Kochalka, in press). The purpose of a sym ptom is to keep the
system unchanged, to freeze it in repetition or in increasing abu
siveness and apathy, which leads eventually to its com plete break
down. A system redresses the balance of power within a family:
The least powerful and m ost vulnerable individual, usually the
symptom bearer, holds the upper hand and has control and
power over the rest of the system . The sym ptom is usually meta
phorical of relationships, and allows the m ost vulnerable and sup
posedly weakest individual to achieve som e degree of supremacy
over the rest of the family system .
O ne of the m any characteristics of dysfunctional families
(L'Abate & Kearns, subm itted for publication) is acontextuality;
that is, m ost families are unable to see the sym ptom in context.
Thus they evidence linearity (concreteness in particular causes/ef
fects), digitality (at the expense of analogic thinking), uncontrolla
bility of the sym ptom , defeating behavior, conflicts in distance
regulation, lack of intimacy, and lack of laughter. The sym ptom ,
therefore, represents a double bind: The m ost vulnerable and un
controllable individual, who is the sym ptom bearer, through his
uncontrollability, achieves control of the system . The m ajor para
dox of the system —the achievement of power without appearing
to achieve power—is the inherent contradiction in m ost dysfunc
tional families. This contradiction also applies to the therapist's
behavior; that is, the therapist needs to achieve control without
appearing to control.
THEORY OF CHANGE
Change occurs w hen the sym ptom is restructured in a positive
context. W hen reframing is strong and frequent enough, the fam
ily can successfully work as a system . The m ost im portant condi
tion in the process of change is a cognitive restructuring—a re
framing of the sym ptom from a negative to a positive context.
Change occurs w hen a therapist helps a family achieve control of

the sym ptom through either a straightforward prescription of the
sym ptom , or positive reframing plus prescription of the sym p
tom: The sym ptom is ritually and systemically linked to the rest
of the family (Weeks & L'Abate, 1982). The client does not need
insight into why and how change has taken place. In fact, it can
be argued that in som e cases insight will be harm ful. Instead, the
therapist can use positive restructuring to explain how a sym ptom
works. A full explanation should be given in a variety of positive
attributions rather than in the one attribution acceptable to the
client (L'Abate, in press).
Cognitive restructuring in and of itself is not sufficient to effect
change. Even if thinking is altered in dysfunctional system s,
changes in thinking cannot be directed toward behavioral change
per se: by the very nature of reactive and apathetic behavior, one
person's more positive thinking will be contradicted by the more
negative thinking of others who are frightened by any possible
change in the system . Thus thinking or rationality would actually
be counterproductive. It is im portant, therefore, that positive re
framing, multiple or otherw ise, be followed by a precise prescrip
tion for altering the behavior of the family, a prescription that
links the sym ptom ritualistically—through repetitive m anifesta
tions of the dysfunctional behavior in w hich the whole family sys
tem is involved. W ith this particular framework (positivity, differ
entness, and frequency), change can occur. Eventually, of course,
the ultimate goal in family therapy is to help the system achieve
intimacy, an area that m ost paradoxical, structural, and system ic
therapists have ignored (L'Abate & Sloan, in press, [a], [b]).
How rapidly paradoxes will produce change is difficult to pre
dict. W hen a paradoxical direction, or injunction is given in terms
of multiple positive attributions and sym ptom repetition that is ritual
ized and systemically linked to the rest of the family, change can be
very rapid. This change, however, represents only the first part of
therapy (L'Abate, subm itted for publication [a]). I agree with
Alexander and Parson's (1982) view that therapy consists of three
separate phases. The first is sym ptom alleviation, crisis reduction,
turmoil decrease; the second is skill training; and the third is ter
mination. M ost paradoxical techniques are to be used in the be
ginning phases of treatm ent to help the system achieve a degree
of stability. A fter this goal has been achieved, a variety of linear
procedures (task assignm ents, structured enrichm ent) may be fol
lowed (L'Abate, subm itted for publication [a]). Therapy is not
based on a single, simple use of one technique or m ethod.
Families are interested in procedures that work; their orienta
120 CHAPTER 5
tion toward help is very pragmatic. The family orientation be

comes philosophical only in family system s that are too rationally
oriented, systems that use rationality to avoid emotionality, or
'having' and 'doing' to avoid 'being.' W hen the paradox is posi
tively administered, the reactions—in addition to what is called
the paradoxical face—are delight, pleasure, enthusiasm , relaxa
tion, and enjoym ent. These responses should be the outcome of
any therapeutic procedure and should especially occur w hen the
family learns to control and conquer the symptom.
It is useless to consider conscious-unconscious levels in family
functioning. In some cases paradoxical directives work without
any explanation at all; in others they function best w hen a full
explanation is given. W hether or not (and how m uch) the thera
pist should explain depends on the family: some families neither
want nor care for explanation; others crave, demand, and need a
full, logical explanation of whatever is being done. It is im portant
then that the therapist understand how the family functions best
in order to meet their expectations.
Som etim es am nesia occurs; that is, the family is com pletely un
aware of what happened during the first session. For instance in a
blended family that had been able to defeat a therapist after two
years of treatment, rapid change occurred w hen the 13-year-old
daughter's misbehaviors were relabeled as "p rotectio n " for her
mother's depression. The daughter was congratulated for all the
care she had shown in protecting her m other and being so very
sensitive to the m other's needs. These positive verbal interpreta
tions were all that was needed to bolster this girl's feelings of sag
ging self-esteem and failure and turn her into a delightful, happy,
and successful teenager, one who brought about a new level of
functioning in the rest of the family. O f course, other directives
and other interventions were needed. Yet the family, including
the former identified patient (I.P.), repeatedly indicated their com
plete unawareness of how this "m iracle" had been achieved.
In some cases, the family is perfectly aware of all the steps that
have taken place. A key to the hum an condition, w hen consider
ing behavior from a viewpoint of individual differences, is that it
is difficult to condense or to generalize the im portance of a single
dimension.
Change can only occur in context. It is almost impossible to
obtain an isolated change, one that does not affect the rest of the
system . I am not aware of (nor can I imagine) any situation in
which symptomatic relief was not accompanied by some change
in the rest of the system.
The therapist who wants to make a living needs to produce
change. I am aware that many individually oriented therapists

have successful practices by helping individuals " to feel good "
even though no external change seem s to be taking place, espe
cially in the client's family relationships. For a marital and family
therapist, however, behavioral change is the measure of effective
ness. The therapist need not behave in any specific ways, but
needs such a com bination of expertise and hum anness that expert
interventions can be adm inistered without appearing expert. Per
haps this is another paradox: The way a therapist needs to act is
perhaps im portant to the therapist's internal consistency; that is,
being an expert without appearing to be one.
The importance of always being hum an is in contradiction to
the "e x p e rt" position taken by the Milano Group (H ansen &
L'Abate, 1982), who like to maintain some distance between
themselves and the families. The paradox here is to achieve close
ness without losing distance—to be ju st close enough and just
distant enough to be objective and effective. Paradoxical interven
tions need to be imparted in a context of personal hum anness,
which is what makes almost every m ethod a technique, and tech
nique m ust always be grounded in some methodology. If one
equates m ethodology with expertise and techniques with hum an
ness, it is the com bination of the two that produces change. Over
reliance on method is as counterproductive as overreliance on hu
m anness. Effective therapy is always a com bination of the two.
Paradoxical therapy may be necessary at the beginning of ther
apy, but it is not sufficient to effect change. Effective therapy con
sists of a variety of structural, linear, circular, and metaphorical
procedures that need to be interwoven in the therapy process. In
some cases, paradoxical therapy may function as a placebo, in
that it decreases "a n x iety " thus decreasing variability and im
proving the stability of the system . If this is what the placebo
does, it could be that paradoxical therapy makes the system more
amenable to linear interventions. W hether it works as a placebo to
provide temporary relief or as a "real cu re" to provide long-range
changes, a therapist can often make hay out of either outcome
(L'Abate, submitted for publication [a]).
Change can be lasting w hen a positive spiral takes place; on the
other hand, there may be regressions. The therapist m ust always
take the position that the family is responsible for change (" I do
not want to be praised because I d on't want to be blam ed").
Thus, any intervention needs to be evaluated by the test of time.
As therapists, we cannot rest on our immediate observations and
accept positive change as perm anent. We need to question
w hether the change is indeed lasting and w hether it has pro
122 CHAPTER 5
duced a positive spiral of other changes and offshoots to indicate

that the interventions have been truly successful.
Paradox is not a religion: there is no question of believing or
not believing the paradox. The only belief necessary is the accept
ance of alternative positive explanations (i.e., positive reframing),
worded in a way that makes sense to the family. The family's be
lief may not be necessary; following the directives, regardless of
belief, is necessary. The reorganization in thinking that derives
from the positive reframing may take som e time. Furtherm ore,
different family m embers need different interpretations or they
will interpret differently regardless of w hat the therapist says.
W hat is im portant is that the family m embers behave in new ways
that are in agreement with w hat the therapist has said. They may
follow the directives and change may not occur, indicating that a
different paradox may be needed or that the therapist m issed the
boat at that particular time and needs to use other kinds of direc
tives. This is why the therapist m ust always qualify directives
with tentativeness, with uncertainty: " I'm not sure w hether this
will help; it may sound crazy "; " I d on't know w hether or not this
will work, but let's try "; " I d o n 't know anything else to do at this
point—but let's see w hat happens i f . . . " These qualifications
m ake the therapist more acceptable and the intervention m uch
more hum an than it m ight otherw ise be.
The best way to prevent a relapse is to predict one. The relapse
should always be interpreted as a natural regression; the therapist
needs to be ready for regressions at any level and stage of ther
apy. W ithout regressions, we might question w hether the system
is really alive and struggling. O ne should not only be very tenta
tive in giving prescriptions but should wonder w hether they will
work, w hether som ething else may be necessary. In fact, I rou
tinely say from the very beginning of therapy, "T h in g s are going
to get worse before they get better," or, " I f I'm doing you any
good, things are gonna have to get worse so that the system can
explode and existing, traditional ways of behaving can be given
up for new, different w ays." A ny kind of relapse or regression can
be positively interpreted as the system 's way o f returning to fa
miliar, past behaviors due to the difficulty of finding new ways of
relating to each other ("T h e devil I have is better than the devil I
m ay g e t").
W hen there is a sym ptom substitution, the same approach is
necessary; it may indicate that the therapist has erred in dealing
with the sym ptom to begin with or has m iscalculated the dys-
functionality of the system . There are m ultiple-problem families
w hose degree of dysfunctionality is such that all the therapist can
do is fix the leaks. W hen one leak is fixed, another springs up

elsewhere. This, however, is not an indication of sym ptom substi
tution but an indication of the degree of dysfunctionality. There
are families w hose hom eostasis is predicated on continuous up
roar and turmoil. These are system s that not only resist interven
tion but are not interested in intervention (apathy). There are sys
tem s that survive through turmoil, system s that in fact would not
know how to deal with each other because they would face such a
degree of internal depression that hom icide, suicide, or psychosis
could be the outcom e of stirring up such extreme underlying
(genotypical) dysfunctionality. In som e cases, these families
m ight be let alone until they break down completely. This may be
a heartless thing to say, but if help is not sought and no change is
forthcom ing, there is very little that a therapist can do.
INDICATIONS AND CONTRAINDICATIONS

Assessing the degree of dysfunctionality is im pressionistic and
intuitive, and is based chiefly on w hether the system will or will
not follow directives, hom ework assignm ents, tasks. The degree
of conformity with or opposition to a hom ework assignm ent is
one of the basic clues by which we can evaluate the appropriate
ness of the assessm ent. Psychiatric labels m ay som etim es help to
indicate the degree of severity of the dysfunctionality; however, if
the labels are used to deem phasize a systemic perspective, those
labels are not only useless but can be destructive to a systemic
understanding of a family. A ssessm ent, then, is m ostly subjec
tive, inferential, and attributional. How m uch the family can or
cannot take the interventions that are given during the initial ses
sions is one way to assess the degree of dysfunctionality or func
tionality. The sym ptom , the strength of the sym ptom , and the de
gree of rigidity of the family—all these characteristics are present.
A n informal checklist (L'Abate & Kearns, subm itted for publica
tion) plus a variety of other paper and pencil tests and question
naires (L'Abate, in preparation [b]; L'Abate & Wagner, in press)
about the characteristics of dysfunctional system s are used to
assess the system multidimensionally.
The first three sessions are thus devoted to assessm ent in
which the family evaluates the therapist(s). It is up to the family
to decide after the third session w hether they w ant to enter a
therapeutic contract. The first three sessions are strictly diagnostic
in term s of present and past situations. History and a genogram
are som etim es useful, even though they may be taken sketchily at
124 CHAPTER 5
the beginning and filled in during the course of therapy. Tasks are
used to check on the degree of functionality of the family system .
If a family is able to comply directly with structural changes and
complete homework assignm ents, there is no need to go on to a
circular or a metaphoric approach.
If tasks are not completed or not even undertaken, a more cir
cular approach may be necessary. W hich kind of intervention is
decided strictly on the basis of cooperation or inability to cooper
ate. Cooperation suggests use of structural and linear approaches;
lack of cooperation or inadequate cooperation suggests use of cir
cular or metaphoric approaches.
In the diagnostic assessm ent of a family, the theoretical frame
work presented earlier through the models of Apathy-Reactivity-
Conductivity (A-R-C), Emotionality-Rationality-Activity-Awareness-
Context (E-R-A-Aw-C), and power sharing and negotiation are all
used. The framework is diagnostic as well as therapeutic. The lev
els of abstraction in therapy serve as a framework for specific
treatment guidelines. Go structural if possible; if not, go linear; if
linear does not work, go circular; and if circular does not work,
try metaphorical. Som etim es, one reasons quickly and intuitively
that the metaphorical might be used. These are not absolute
guidelines; it is very difficult to give any guidelines that could not
be contradicted (see Weeks & L'Abate, 1982, for contraindications
for the use of paradox).
The paradox in and of itself is just one approach, and it has to
be mixed with all the other approaches. The more techniques and
methods the therapist knows, the more flexible and better off the
therapist is in meeting the various needs of families. Thus the
therapist shifts always from one level of abstraction to another.
The level of abstraction and flexibility of the therapist in meeting
the family's demands could range across all four different levels.
There is no one treatment of choice: In certain cases, paradox may
be the only treatment of choice; in some cases, it is never the
treatment of choice. There should be as many ways to meet fam i
lies as will allow us and them to succeed.
STRUCTURE AND PROCESS OF THERAPY

In terms of sequence of interventions, the level of abstraction
already outlined indicates how we can work with a com bination
of approaches. Som etim es, especially if one knows from the first
session that the symptom bearer is hostile and negative and will
oppose any task assignment, it is clearly im portant that the task
assigned from the beginning session be paradoxical. If the sym p

tom bearer and the rest of the family respond positively to the
therapist, dem onstrating very few signs of opposition, structural
and linear approaches are more appropriate.
The stage for a paradoxical intervention is set in term s of the
family's readiness to interpret reality in different, positive, and
strong ways, through multiple attributions (L'Abate, in press). In
preparing the stage, the therapist is extremely tentative about
how the intervention will work and very clear about if, w hen, and
how the intervention is to be given. That is, the therapist uses the
four basic aspects of paradoxical intervention—positive reframitig,
leading to a prescription that is ritualistically repeated and systemi
cally linked to the rest of the family.
Deciding which technique to use is mostly intuitive, in the
sense that it is done very quickly and the therapist m ust feel con
fident about the chosen intervention. But w hen there are many
indications of potential resistance and rigidity in the system , a
paradoxical approach clearly seem s in order. The first three diag
nostic sessions are used to decide which techniques to use, w hen
and how to use them .
No intervention is ever given unless there is a follow-up. A di
agnostic contract for three sessions needs to be reached before a
therapist can give the next assignm ent. An intervention is given
for the purpose of diagnosis rather than improvement of the sys
tem . The family is told that, except for specific directions given by
the therapist, the system is not to change but to continue as
usual.
W hen an intervention does not work, we have to decide
whether it was poorly structured, w hether there was som eone in
the system who sabotaged it, or w hether a more relevant inter
vention is needed. (The whole issue of defeats and failures in
therapy is considered in L'Abate & Baggett, in press; L'Abate &
Kochalka, in press.) If the family is truly bound by defeat, they
enjoy defeating each other and are set in defeating. If they know
how to relate only by defeating each other through discounting,
contradicting, reactivity, and apathy, the therapist m ust take hold
of the fear that s/he will be defeated. W hen the defeat is clear, it
needs to be positively reframed: "You need the defeat and you
need to defeat each other [by not doing assignm ents, or not doing
homework, or not carrying out whatever directives are given] to
be together and to keep the family close; It is im portant that you
continue defeating each other; let's see how many ways you can
find to defeat."
W hen defeat seem s to be the key them e of a family's function
126 CHAPTER 5
ing, it is im portant that the therapist prescribe the defeat, allow

ing the family latitude to continue to defeat each other but start
ing the defeats at certain tim es, in certain places, and in certain
ways (frequency and intensity), all of which allow the therapist to
achieve control over the defeats. Linearly, one can deal with the
defeats by having every family m ember list how each defeats oth
ers and how each is defeated by others. A fter this assignm ent, a
m aster list is developed by having each person share their individ
ual lists with the entire family. Everybody then knows how the
family succeeds through the defeats. The family is exhorted to
continue defeating each other, because if they do n ot do so they
may fall apart and break down. Thus, appreciation of how defeats
keep the family together is im portant. Although such an explana
tion may sound paradoxical to the family, it is not paradoxical
from the therapist's viewpoint. The paradox is that prescription of
the defeat allows the therapist to achieve control over self-defeat-
ing and other-defeating behaviors.
Prescriptions m ust be given with time, place, and frequency
constraints. For instance, siblings who are experiencing sibling
rivalry ("loving each o th er") are assigned to bicker and fuss at
each other—at certain times, in certain sequences, and in certain
places, with links to the rest of the family. If they are going to
fight, they should do it for 30 m inutes on Monday, Wednesday,
Friday, or for 15 m inutes on Tuesday, Thursday, Saturday. These
are part of the ritualization that helps the therapist achieve control
over the system . Ritualization makes the assignm ent specific, but
the therapist m ust keep always in mind that positive reframing
should come before the assignm ent is given to the system : "You
care enough to fight with each other. If you did not care for each
other, you would not fight. Fighting is, after all, a form of love-
m aking."
The goal of treatm ent is to help families function more produc
tively and constructively and to feel better about them selves and
each other. The goal of therapy is to help families negotiate is
sues; thus, no m atter what issues they w ant to negotiate, they
need the skills to negotiate (L'Abate et al., in press). This is an
area in which paradox is limited. In other words, paradox is lim
ited at the beginning of therapy w hen there is a crisis or an up
roar. Eventually, the initial phase of treatm ent, as m entioned be
fore, needs to be followed up by skill building. Skill building con
sists of giving families a paradigm by which they can learn
(through rules, regulations, and guidelines, including, eventually,
family conferences) how to negotiate (L'Abate, subm itted for pub
lication [a], [b]).
The goals of treatm ent—what they want to get out of therapy—

should be decided by the family. They may decide the specific
content. The process, however, is decided by the therapist; the
process then results from the structure provided by the therapist.
The specific issues that the family w ants to deal with are the con
tent. It is up to them to learn to negotiate those specific issues.
Som e of the process goals—to becom e m uch more conductive,
more balanced in functioning, and m uch more congruent at levels
of functioning, to learn how to share power, and to learn to nego
tiate—are appropriate for all fam ilies. I cannot think of any fam i
lies for w hom these goals are not useful. The specificity of the
content is the business of the family; the generality of the process
is the business of the therapist.
Paradoxes are fully explained to som e families, w hich allows
them to think differently about the system . The level of explana
tion, however, is limited strictly to how m uch a family can take. If
a family includes a six-year-old child, it is going to be very diffi
cult to explain certain aspects to that child. The level of explana
tion has to be geared to the lowest com m on denom inator of un
derstanding in the family, w hich may som etim es limit w hat a
family can understand. It is more im portant, however, for families
to change for the better than to understand.
Clients try to avoid som e practices by not following the task or
by indicating that they do not buy the positive reframing, thus
continuing to do what they have been doing. At this point, the
therapist needs to be aware of being defeated and use the defeat to
explain how the family needs to defeat each other. Termination is
com plete w hen the family feels they have achieved their goals—
that they feel better, that they are more functional, more relaxed,
more satisfied with each other, and that they have achieved som e
constructive m odus operandi to indicate that they are in charge
(e.g ., family conferences in which they negotiate issues more con
structively).
According to the therapeutic contract, cancellations m ust be
made 24 hours in advance or the family is charged (except for dire
em ergencies). Termination m ust take place in the therapist's of
fice, not by telephone. The agreem ent to term inate needs to in
clude everybody in the family. The whole issue of term ination is
an im portant one in ascertaining w hether the family is indeed
ready for term ination. Som etim es, one m em ber of the family does
not want to term inate because s/he remains dissatisfied. In such
an instance, how m uch freedom is that individual given to ex
press an opinion? An agreem ent that is satisfactory to the individ
ual who does not want to term inate can be made to terminate
128 CHAPTER 5
within a given num ber of sessions. Here, the therapist needs to be

sensitive to unfinished business, intimacy typically being one of
the most important (L'Abate, subm itted for publication [a]).
Success is determ ined by the family's happiness in being alive,
their feelings of joy and success in negotiating, com pleting issues,
being in charge, and their reported satisfaction with services
rendered.
TECHNIQUES
Paradoxical techniques should be organized as follows: (a )
"Things are gonna get worse before they get b etter"; (b) " I'm not
sure this is gonna work, but let's try it and s e e "; (c) positive re
framing through prescription of the sym ptom , which is (d) ritual
istically and (e) systemically linked for the whole family.
I continue to rely on written (I believe very strongly that letters
can be used with children) as well as verbal com m unications, and
I believe that more linear techniques should be developed before
one goes on to the paradox. If there are any com m on errors in the
use of these techniques, they lie in using them too stereotypically
and generally, without considering the specific case and the spe
cific context of specific families. Using paradox rigidly and uncriti
cally will, I think, produce severe disappointm ents for both fami
lies and therapists (L'Abate, 1977; Wagner et al., 1980).
CLINICAL ILLUSTRATIONS
Som e uses of the paradox in an inpatient children's setting
have been reported by Jessee and L'Abate (1980) and in other pub
lications (L'Abate & Farr, 1981; L'Abate & Sam ples, 1982; Soper &
L'Abate, 1977; Weeks & L'Abate, 1982). This section will describe
isolated cases in which I consulted on a one-shot basis and stayed
in touch with the therapist about the long-range results of the
intervention.
In one case, treated by Don Laird, I consulted over a period of
three years. A young man of 22 had a long history of hospitaliza
tion and incarceration for sudden, seemingly unprovoked and ex
plosive temper tantrums that ended in destructive or aggressive
outbursts against property and people, especially members of his
family. During previous hospitalizations, he had at various times
been diagnosed as paranoid schizophrenic and given other, simi
lar labels. He was seen in the fourth session during a treatm ent
that spanned three years. The aggressive outbursts were ex
plained to him as his way of expressing him self. As positives,
they were accepted as part of his style. Perhaps he m ight want to
achieve control over these outbursts (he did). But it would be very
difficult, if not impossible, not to have them because he used
them, instead of emotionality (which he avoided diligently), as
his mode of self-expression.
Consequently, a plan was laid out: He should have aggressive
outbursts at specific times with specific family m embers with
whom he was involved and in w hose presence he was in the habit
having temper tantrum s. He was also told to call the therapist
after each outburst to describe the outcome for the family; he
called to indicate that he did not feel like having a tem per tantrum
but that he would try again. The therapist expressed amazement
at his inability to follow instructions, especially in view of his long
history of aggressive outbursts. He was told to try again at the
specified times. He should have them in the houses of both his
grandmother and his parents. He did have one or two minor half
hearted, half-baked aggressive outbursts.
Eventually, the outbursts disappeared completely. He was able
from then on to start writing letters to members of the opposite
sex, with whom until then he had been unable to establish any
relationship. Ultimately, he was able to apply for a job, which for
the first time he maintained for longer than one year. He is now
contemplating marrying the girl he has dated for more than a
year, and he has kept his current job for two years.
The next case, which also concerns tem per tantrum s, was a 10-
year-old boy, from an abusive, incestuous family, who had been
farmed out to a foster family for possible adoption. Therapy was
conducted by Dr. M argarett S. Baggett; I was again the one-shot
consultant. The boy's tem per tantrum s were quite explosive and
threatening, especially to the foster m other; he also exhibited
them in school. The tantrum s were positively reframed as his own
method of expression. A specific place was chosen where he was
encouraged to have his tantrum s on a regular basis. The principle
here is that the undesirable, symptomatic behavior should be made to
occur zvhen and where it does not naturally or spontaneously occur
(L'Abate & Kearns, submitted for publication). In this case, a
basem ent room (which contained only empty boxes) was desig
nated for the tem per tantrum s. The foster father was encouraged
to get a couple of old inner tubes, fill them , and ask the child to
go to the basem ent on a regular basis to have his tantrum . We
agreed that Saturday morning, w hen the foster father would also
130 CHAPTER 5
be present, would be the best time. The boy was then to have a
tantrum for at least 15 m inutes. He was to report back to the ther
apist w hen h e had finished his assignm ent. The foster parents
were to encourage him to have the tem per tantrum from 10 A.M.
to 10:15 A.M. every Saturday m orning. The child reported to the
therapist (Dr. B .) that he had failed to do the assignm ent and that
he was not planning to have any more tem per tantrum s. The
therapist expressed surprise at his inability and unw illingness to
carry out the assignm ent and also noted that Dr. L. would proba
bly be very disappointed that he had not carried it out: He was
encouraged to continue having tem per tantrum s, especially at the
time and place specified. This case was followed up after a few
m onths, during which time the tem per tantrum s, especially at
hom e, had subsided to zero, even though the boy still had a few
tantrum s at school. Unfortuantely, however, the foster parents
changed their m inds about adopting the child because of the
stress he seem ed to place on their marriage. To decrease the pain
of loss w hen the boy left, Dr. Baggett organized a farewell party
to make the leavetaking a joyous occasion.
In spite of his considerable improvement and exem plary behav
ior, this child did continue to exhibit occasional fits of opposi
tional behavior, which displeased his foster parents a great deal.
Consequently, plans were made for his return to his family of ori
gin, who, in the meantime, had received therapy and counseling
from another therapist. Although the battle of the tem per tan
trums was won, the war—to keep this child in the foster hom e
and be adopted—was lost. N evertheless, the child did show con
siderable improvement in his self-control, even upon returning to
his family of origin.
A nother case was a couple, married for 25 years, referred to us
(Dr. L and M rs. L) by their son, who was concerned that they
stayed at hom e all the time. During those 25 years, the m other
had developed a com plete phobia about housew ork. The father
was employed full time but had learned to cook and do the
household chores, while the wife went shopping and spent her
time in nonproductive activities. She had becom e com pletely pho
bic about going outside (except for shopping) and enjoying life.
Theirs was a rather miserable existence, with no friends, no
travel, no vacation—all because of the wife's continuous need to
control all situations. Eventually, she was verbally congratulated
for the way she had achieved control over everybody (without, of
course, having appeared to achieve control) and for how m uch
she cared for the whole family, especially the husband. She kept
him busy all the time to help him (a refugee and victim of the
holocaust) avoid dealing with his depression and his own hurt.
After receiving this interpretation, the couple reported having
gone on a work-vacation holiday (the first in years). She had be
gun to do housew ork and cook (in spite of the husband's resist
ance!). They discontinued treatm ent after the seventh session be
cause they had not gotten anything out of therapy. I agreed with
them and offered to refund their money, an offer they refused.
CONCLUSION
Paradoxical techniques are but one level of abstraction in at
least four levels of therapy They are especially relevant in the ini
tial phase of therapy to induce sym ptom atic relief and reduce
stress and emotionality. These techniques are not the end of ther
apy but rather, the beginning. To achieve success, a therapist
needs a variety of linear approaches that will help families attain
greater intimacy and learn to negotiate problematic issues.
REFERENCES
Alexander, J., & Parsons, B. V. Functional family therapy. Monterey,
Calif.: Brooks/Cole, 1982.
Bertalanffy, L. von. General systems theory. New York: Braziller, 1968.
de Shazer, S. Patterns of brief family therapy: An ecosystemic approach.
New York: Guilford Press, 1982.
Dewey, J., & Bentley, A. F. Knowing and the known. Boston: Beacon
Press, 1949.
Haley, J. Problem-solving therapy. San Francisco: Jossey-Bass, 1976.
Hansen, J. C., & L'Abate, L. Approaches to family therapy. New York:
Macmillan, 1982.
Jessee, E., & L'Abate, L. The use of paradox with children in an
inpatient setting. Family Process, 1980, 19, 59-64.
Kempler, W. Existential family therapy. San Francisco, Calif: Jossey-Bass,
1981.
L'Abate, L. Principles of clinical psychology. New York: Grune & Stratton,
1964.
L'Abate, L. A communication-information model. In L. L'Abate (Ed.),
Models of clinical psychology. Atlanta, Ga.: Georgia State College
School of Arts and Sciences Research Papers, No. 22, 1969. 65-73.
132 CHAPTER 5
L'Abate, L. Psychodynamic interventions. A personal statement. In R.

H. Woody & J. D. Woody (Eds.), Sexual, marital, and family relations:
Therapeutic interventions for professional helping. Springfield, 111.:
Charles C. Thomas, 1973. 122-180.
L'Abate, L. A positive approach to marital and familial intervention. In
L. R. Wolberg & M. L. Aronson (Eds.), Group therapy 1975: An
overview. New York: Stratton Intercontinental Medical Book Corp.,
1975. 63-75.
L'Abate, L. Understanding and helping the individual in the family. New
York: Grune & Stratton, 1976.
L'Abate, L. Enrichment: Structured interventions with couples, families, and
groups. Washington, D.C.: University Press of America, 1977.
L'Abate, L. Toward a systematic classification of counseling and
therapy theorists, methods, processes and goals: The E-R-A model.
The Personnel and Guidance Journal, 1981, 59, 263-265.
L'Abate, L. (Ed.). Family psychology: Theory, therapy, and training.
Washington, D.C.: University Press of America, 1983. (a)
L'Abate, L. Styles in intimate relationships: The A-R-C model. The
Personnel and Guidance Journal, 1983, 61, 277-283. (b)
L'Abate, L. The paradoxical treatment of (marital) depression.
International Journal of Family Therapy, in press.
L'Abate, L. Toward a systeTnatic approach to family therapy. Manuscript
submitted for publication, 1983. (a)
L'Abate, L. Toward a technology of marriage and family evaluation and
intervention. Manuscript in preparation, 1983. (b)
L'Abate, L., & Baggett, M. Failure to keep a father in family therapy. In
B. Coleman (Ed.), Failures in family therapy. New York: Guilford
Press, in press.
L'Abate, L., & Farr, L. Coping with defeating patterns in family
therapy. Family Therapy, 1981, 8, 91-103.
L'Abate, L., & Frey, J., III. The E-R-A model: The role of feelings in
family therapy reconsidered: Implications for a classification of
theories of family therapy. Journal of Marital and Family Therapy,
1981, 7, 143-150.
L'Abate, L., Frey, J., Ill, & Wagner, V. Toward a classification of family
therapy theories: Further elaborations and implications of the
E-R-A-Aw-C model. Family Therapy, 1982, 9, 251-262.
L'Abate, L., Hansen, J. C., & Ganahl, G. Key concepts and methods in
family therapy. Englewood Cliffs, N. J.: Prentice-Hall, in press.
L'Abate, L., & Kearns, D. The tenets of paradoxical psychotherapy.
Manuscript submitted for publication, 1983.
L'Abate, L., & Kochalka, J. Clinical training in family psychology. In B.

Okun (Ed.), Training issues in marriage and family counseling. Ann
Arbor, Mich.: ACES, in press.
L'Abate, L., & Samples, G. T. Intimacy letters: Invariable prescription
for closeness-avoidant couples. Family Therapy, 1983, 10, 37-45.
L'Abate, L., & Sloan, S. Z. Intimacy. In L. L'Abate (Ed.), Handbook of
family psychology and psychotherapy. Homewood, 111.: Dow Jones-
Irwin, in press, (a)
L'Abate, L., & Sloan, S. Z. Intimacy workshops to increase intimacy in
couples. Family Relations, in press, (b)
L'Abate, L., Sloan, S., Wagner, V., & Malone, K. The differentiation of
resources. Family Therapy, 1980, 7, 237-246.
L'Abate, L., & Wagner, V. Theory-derived, family-oriented test-
batteries. In L. L'Abate (Ed.), Handbook of family psychology and
psychotherapy. Homewood, 111.: Dow Jones-Irwin, in press.
L'Abate, L., Weeks, G., & Weeks, K. Of scapegoats, strawmen, and
scarecrows. International Journal of Family Therapy, 1979, 1, 86-96.
Minuchin, S. Families and family therapy. Cambridge, Mass.: Harvard
University Press, 1974.
Neill, J. R., & Kniskern, D. P. (Eds.). From psyche to system: The evolving
therapy of Carl Whitaker. New York: Guilford Press, 1982.
Ruesch, ]., & Bateson, G. Communication: The social matrix of psychiatry.
New York: W. W. Norton, 1951.
Rychlak, J. A philosophy of science for personality theory. New York:
Houghton Mifflin, 1968.
Soper, P. H., & L'Abate, L. Paradox as a therapeutic technique: A
review. International Journal of Family Counseling, 1977, 5, 10-21.
Ulrici, D., Wagner, V., & L'Abate, L. The E-R-A model: A heuristic
framework for classification of skill training programs for couples
and families. Family Relations, 1981, 30, 307-315.
Wagner, V., Weeks, G„ & L'Abate, L. Enrichment and written
messages with couples. American Journal of Family Therapy, 1980, 8,
36-44.
Weeks, G. Toward a dialectical approach to intervention. Human
Development, 1977, 20, 277-292.
Weeks, G., & L'Abate, L. Paradoxical psychotherapy: Theory and practice
with individuals, couples, and families. New York: Brunner/Mazel,
1982.
6
Ericksonian Styles of
Paradoxical Treatment
by Stephen R. Lankton, MSW, and Carol H. Lankton, MA
We came upon a fortune cookie while writing this chapter that

read: "Always remember: if you tell the truth, you never have to
rem em ber anything." The m aster paradoxologist is the person
who says only the obvious, thus suggesting great m eaning. In
fact, Lao Tzu (c. 550 B.C.) wrote that "W ords that are strictly true
seem to be paradoxical" (1976, p. 168). In this chapter we will not
discuss the content of pardoxical truth but the im pact and use of
paradoxical interventions. More specifically, we will relate our un
derstanding of an approach that makes use of paradox both in
sym ptom prescription and in metaphor.
THE VALUE OF PARADOX

O ur interest in paradox resulted from contact with M ilton
Erickson, a phenom enal paradoxologist. Gregory Bateson, after
studying Erickson, defined paradox as a contradiction in conclu
sions that were correctly argued from consistent prem ises (Bate
son, 1972, p. 223). He observed that som e clients responding to
134
ERIKSONIAN STYLES O F PARADOXICAL TREATMENT 135
Erickson would produce trance phenom ena to resolve problems

posed by contradictory commands which could not be discussed.
He cited this observation as an illustration of a double bind situa
tion being resolved via a shift in logical types (1972, p. 223). Web
ster's Third N ew International Dictionary defines paradox as: " 1 . a
tenet or proposition contrary to received opinion. 2a. a statem ent
of sentim ent that is seemingly contradictory or opposed to com
m on sense and yet perhaps true." Bateson's definition appears at
2(b): " a statem ent that is actually self-contradictory and hence
false even though its true character is not immediately apparent"
(1976, p. 1636). It is this usage that we refer to in the following
case transcript, which includes paradox delivered explicitly in
words (paradoxical binds), in the logic of indirect suggestions, in
the context of a metaphoric story line, in sym ptom prescriptions,
and in paradoxical intention for interview m anagem ent.
Following his contact with Erickson, Bateson introduced para
doxical interventions to Haley and the MRI group who were influ
ential in studying and popularizing this approach. Richard Fisch,
director of Brief Therapy at the M ental Research Institute (MRI)
informed us recently, however, that a trend am ong the MRI group
is to abandon the notion of paradox in therapeutic intervention
(Fisch, Decem ber 2, 1982). This trend results from the idea that
the interventions formerly called paradoxical are not actually para
doxes at all. Instead, the MRI group views them as perfectly logi
cal given the family system in which they are delivered. It seem s
to us, however, that the descriptive term paradox (and related con
cepts) need not be abandoned sim ply because its use in a disor
dered family system is often em inently logical. Obviously, we
w ouldn't be using these interventions unless there were some
logic in doing so, even if the logic is only apparent to the family or
individual receiving them .
Perhaps the ultimate value in paradox rests in the understand
ing that ideas have cycles of existence. This is true for the m om en
tary thoughts we each entertain and for groups of individuals.
W hen groups discuss a topic, the topic is given longer existence.
But the consensus beliefs and sets of beliefs associated to core be
liefs all have a duration and a cycle, both in the culture and in
history. The value of paradox may result from the role it plays in
reminding us, albeit sophistically, that these cycles exist. W ith the
paradox, the listener is rem inded of the wisdom illustrating that
w hat we do now is part of the cycle of w hat we do later, or, at the
very least, shall we say: W hat we think and believe now is cyclical,
and in som e way each seem ingly polarized part allows for the ex
istence of its opposite. We all know that w inter makes way for
136 CHAPTER 6
summer; day makes way for night; the moon sets and the sun
rises; we can get angry and later be more loving; we can decide
affirmatively and negatively about the same incident in a matter of
years, m onths, even shorter periods. Each of us recognizes these
truths to contain a certain natural reminder, but we often fail to
put the wisdom of that knowledge into practical application. Para
doxically, it is true that the moon rises so that we can enjoy sun
light, winter comes so that summer will find us, and so on.
We will take a short tangent to illustrate this point. Part of the
value of this natural wisdom is the colloquial (day-to-day) applica
tion of its principle. For instance, we often hesitate to act on a
decision to follow an impulse until we examine that impulse from
other perspectives or otherwise think it over, give some time to
find out how we feel about it over time, and so on. Each of us has
hesitated on a purchase, held back som ething we might have said
prematurely, restrained our vocal tone, gone som ewhere to "th in k
over" a concern, and so on. Such behavior is preventative prob
lem solving, and often preventative mental health. As a case in
point, Fritz Peris introduced a Gestalt therapy exercise that con
sisted of examining the exact opposite of one's urge, wish, im
pulse, fantasy, feeling. In so doing, the individual learned to
maintain a "creative precom m itm ent." "T h e ability to achieve and
maintain an interested impartiality betw een imagined opposites,
however absurd one side may seem, is essential for any new crea
tive solution of problem s" (Peris et al., 1951, p. 53).
But in the case where a problem already exists, the application
of the principle of cycles comes too late (or was preem pted by the
problem). In these cases, the logic of restraint, acting in an oppo
site way, or tapping an opposite goal has not already been consid
ered. The application of the principle finds value in that doing the
same thing to get the opposite result has not usually been consid
ered. The paradox introduces this consideration by appealing to
an understanding (no matter how well developed) of this wis
dom. Jung wrote: "And just as the conscious mind can put the
question, 'W hy is there this frightful conflict betw een good and
evil?' so the unconscious can reply, 'Look closer! Each needs the
o th e r'" (1959, p. 153). The paradox then is one of the symbols of
this wisdom. Alan Watts stated that paradox is the truth standing
on its head to attract attention (Watts, 1953).
If, however, the paradox does not express profound truth, it
can at least be expected to capture attention and provoke thought.
Parmenides of Elea (515 B.C.) and his student, Zeno (490-85 B.C.)
were among those intrigued by the complexities of paradox.
Plato, in his Dialogue o f Parmenides, referred to Parmenides and
Zeno as the originators of the several paradoxes posed by Socra

tes, which have becom e known as the Achilles paradox, the Ar
row, and others. You may recall Zeno as the man who originated
the idea that anything in m otion is stationary. (O f course, that
idea went now here!) Socrates was at an im passe about Par
menides' reasoning that things do not change, because indeed
things do change and reasoned explanation to the contrary pre
sented an attention-capturing and thought-provoking dialogue.
A paradox used in treatm ent is less likely to convey a profound
truth than to capture attention; yet the use of paradoxical inter
ventions ought not be considered a trick. With paradoxical inter
ventions, as with all interventions, we suggest they be offered
with sincerity. We agree with the MRI group in the understanding
that a paradox is not illogical in the context of a client system . In
the case of a paradoxical symptom prescription, for instance, we
really mean for the client to continue the behavior for his or her
own benefit; we are not trying to trick the client. A paradoxical
trick is actually a sophism , or a conspicuous paradox, it often
alerts the listener to an intention to convey an elusive m eaning.
Sophism s such as Zeno's are clever m anipulations that are false
and have knowingly been com m itted. A true paradox, on the
other hand, is knowingly com m itted but true: As such it may lead
the listener to a disquieting conclusion or absurd conclusions, but
it is not fallacious. That is, a genuine paradox is a conclusion
based on valid reasoning at every step.
A paradoxical intervention then may be used to help produce
several results that can be of value in treatm ent. It symbolizes the
natural wisdom of im perm anence; it captures attention and may
provoke new thought. M ost of all, it illustrates a line of reasoning
to reach its conclusion. Therefore, it arrests attention, overloads
consciousness, and causes the listener to question axioms, postu
lates, beliefs, reasoning, facts, memory, congruity checking, and
so on. In short, it stim ulates thinking in general, and elicits re
sponses that are exactly the social and psychological results
sought for and created by paradox.
SOCIAL HYPNOSIS, PATHOLOGY,

AND PARADOX
We are all aware of the significance of a transaction for the indi
vidual, and the health or pathology that is transmitted, elicited,
shaped or reinforced in an individual. But the effect of a transac
138 CHAPTER 6
tion on the social network has an equally im portant causational

influence on mental health. A transaction is a model for all ob
servers in that it provides a map, and thereby socializes others.
W hen others operate from the map, the transaction reinforces in
that it predicts (roughly) how the person upon w hom the m ap is
used will respond. For instance, som eone makes a joke and in
some way puts another person down. W hen this transaction is
observed, the role and behavior of joking are associated and
m emory provides a partial map of conduct for the observer to use
w hen faced with a similar situation, person, or set of prom pts:
Thus the social network is sim ultaneously socialized to treat the
"o d d " or "h o stile " child as if s/he were odd or hostile. The map
predicts the results the user can get from the real world and thus
form s part of the user's self-system and ultimately part of his or
her belief system.
The role taken by the therapist w hen s/he uses paradoxical
symptom prescription with a family or family m em ber confuses
(at least temporarily) all m embers in the social netw ork regarding
their roles and behavior related to the sym ptom . Therefore, para
dox may be used to stop the deleterious social transm ission,
reinforcem ent, and shaping of the sym ptom . In m om ents w hen
the usual pattern of social interaction and transm ission of roles,
rules, affects, and ideas are stopped, the therapist gains an entry
into the broader areas of the client(s) personality(ies) and has an
unparalleled opportunity to help rearrange new com binations of
personal experiences, perceptions, and m aps of conduct. We
demonstrate in the transcript that follows several areas in which
therapeutic outcom es are created, using paradox to im pede usual
thought association processes and m etaphor to convey both the
paradoxical logic or directive, and to stimulate m ental search and
retrieve novel experience. First, however, we want to examine
the social effects of paradox on both social transactions and
personality.
PARADOX AND PERSONALITY

Although it has becom e fashionable to take only a system s ap
proach—as we have done above in analyzing transactions—we
think there is also m uch to be gained by looking at several com
m only acknowledged aspects of personality and asking what
m ight be happening with respect to feelings, imagination, and ex
pectations. In the case example presented in this chapter, the cli
ent alters his expectations and his perceptions. In fact, he in

creases and concentrates attention in a slightly different way as
ongoing behavior continues. As a result, he has an opportunity
for novel experience that is of course therapeutically directed.
This socially or transactionally created "o p e n in g " to the client
em phasizes that alterations in internal behavior are possible as a
result of com m unication containing paradoxical features.
If we assume that drives, urges, and bodily im pulses m ust in
som e incipient way be involved in movement and consequent
roles taken by the person, then we m ight ask w hat about the
change in desires, urges, im pulses, and so on that has been set in
m otion or otherw ise involved by the paradox. W hat change in im
pulses and drives m ight explain how the person com es to confide
in the therapist slightly more as a result of the changed frame of
reference that has been cocreated with the therapist? As confiding
and risk taking are seen as m easures of self-esteem (Yalom, 1970),
we might ask how drives and urges m ight have been affected
such that self-esteem and risk taking have been altered. The an
swer of course is that the client's creativity is subjectively in
creased due to the symbolic value of paradox. Further, depending
upon the subsequent com m unication, the client may experience
expanded options for previously unnoticed behaviors, feelings,
and perceptions.
W hen a person hears a paradox, s/he does not know w hat ex
pectations will be sanctioned. S/he m ust suspend the custom ary
frame of reference and therefore the predisposition to a particular
feeling state. A great deal of bodily and psychodynam ic experi
ence is temporarily altered w hen a person is engaged in response
to paradox and m uch of this alteration becom es the foundation for
therapeutic learning.
A n impulse is a wave of excitation passing through certain tis
sues. A n urge is a force that rouses activity from an otherw ise
dormant state. Urges, drives, im pulses, and the like, as bodily
experiences, are affected by the state of the body; they occur in
relation to nutrition, stim ulation, sleep, and chemistry. But ulti
mately, they are woven into the map of historical experience and
channeled into w hat the individual would call the way s/he
"fe els," the "m o o d ," or more scientifically, his or her self-system .
Usual em otions, w hether healthy or pathological, are the com
bined result of these phenom ena. Affect, then, is created as the
gestalt of bodily urges, desires, and drives interacting with the
m atch (or m ismatch) created betw een perceptual reality and antic
ipated images.
140 CHAPTER 6
SEARCH PHENOMENA
O ur synthesis of the personality consists of a redundant or re
curring blend of perceptions, expectations, and urges, all of
which combine to produce particular affects. Consciousness and
deliberate thought in imagery occur secondary to this otherwise
primary process and only in varying degrees do individuals exer
cise the ability to inhibit, direct, or modify the impact of these
affective states voluntarily and experimentally. More often, the so
cial system is created to exert control in cases where the individ
ual has not.
There is predictability to a personality even though it operates
dynamically within a living and changing person. An individual,
until s/he changes, continues to expect a finite variety of experi
ences in the world (e.g., the postm an will be anonym ous, the fish
store personnel will be friendly, Sally will be eager, our son will
like this and that, etc.). This colloquially noticed personality is of
course a result of the predictability gained by the redundant oc
currence of relatively similar urges, expectations, perceptions,
and subsequent states of affect.
In the absence of conditioned responses to paradox (a highly
unconditioned stimulus), the person has no established map to
follow from previous learning and will consequently respond with
searching behavior designed to produce an appropriate response.
The person attem pts to construct a new map to guide behavior,
affect, and expectations regarding likely consequences for both self
and others. Specific indicators of such internal searching are flat
tened cheeks, decreased movement, slowed reflexes (breathing,
blinking swallowing), pupil dilation, eye scanning, and increased
pallor. These search phenom ena are considered to be signs of light
trance in that the person's attention is at least temporarily inter
nally concentrated.
To the extent that searching for a logical map to follow is not
successful, the person can be expected to becom e increasingly re
ceptive to external direction from the therapist which stimulates
the co-construction of a map containing and detailing new op
tions. Working from an Ericksonian approach to treatm ent, we
find that deepening the naturalistic trance initiated by the paradox
and then presenting m etaphors to the client provides him or her
an opportunity to entertain novel experiences in a nonthreatening
way, examine them as possible options, evaluate them from a per
sonal perspective, and expand the map to include those m eta
phors judged personally relevant by the individual.
Though what will be personally relevant to the individual is
ultimately decided by the individual, the therapist is obligated as a

professional to diagnose and assess the client and the social net
work, make a carefully considered estimate of which therapeutic
outcomes are likely to be relevant, and design a treatm ent plan
which provides the client an opportunity to learn the affect, trans
actions, attitudes, and so on compatible with achieving such out
comes. We list in some detail six therapeutic outcom es which we
consider essential in any com plete treatm ent. Though the pre
senting problem and the greatest deficit in logical maps may
seemingly be confined to one area, we would want, in the interest
of thoroughness, to metaphorically address possible changes in
each of the other areas that m ight be related to integrating new
behaviors. In the case transcript presented in this chapter, we
metaphorically address four of these six outcom es in the first in
terview. But, prior to discussing therapeutic outcom es, we will
briefly summarize several guidelines for creating effective para
doxes to be used in Ericksonian approaches.
ELEMENTS OF EFFECTIVE
THERAPEUTIC PARADOX
Recognizing that there are many and varied ways to intervene
paradoxically, we offer the following elem ents of paradox as flexi
ble guidelines for therapists to use in initially formulating and de
livering paradoxical interventions, as well as in examining why a
particular paradoxical intervention was not effective. We do not
imply that all of these elem ents m ust be evident in any effective
paradoxical intervention, but that each elem ent is im portant to
consider and address in some way, though not necessarily in an
obvious fashion or in the order in which they are discussed here.
Empathic Activity and Establishing Rapport

Establishing rapport is a foundation goal in most therapies; it is
particularly crucial in a therapy in which clients are not encour
aged or expected to have a conscious understanding about the
manner in which therapy is proceeding. In fact the opposite is
typical, namely that clients are expected to experience a good deal
of conscious confusion and perhaps even anger at the therapist
for behaving in such an unorthodox manner. An angry reaction
may be subdued, accepted by the client, or avoided entirely if suf
ficient rapport is established prior to the therapist behaving para-
142 CHAPTER 6
doxicaUy. Occasionally, however, offering paradoxical behavior

prescriptions is itself a way of establishing rapport, at least with
that part of the client that is generating the behavior being pre
scribed. For example, in the case to be discussed later in this
chapter, the client presents him self for therapy because of his
oppositional attitudes and resulting difficulties with “ authority."
Rapport was facilitated by beginning a trance induction with the
instruction that the client was to rebel against our words to the
extent necessary for him to be certain that he followed only those
instructions that were in his best interest. It was quite likely that
h e would rebel in some way to what we said to him and by sug
gesting that he do so, with good reason, we expected to stimulate
his feeling of being understood at one level and sim ultaneously
confused (therapeutically) at another.
M ore frequently this elem ent of building rapport and helping
the client feel understood is accom plished more conventionally.
Erickson would often dem onstrate his understanding of a client's
presented difficulty by restating almost verbatim the client's com
plaint. Reflective listening training has prepared m ost therapists
to almost automatically paraphrase the client's expressed or im
plied feelings. This training allows the therapist to "autom ati
cally" respond emphathically to the client's expressed difficulty
while at another level enabling him or her to begin formulating
paradoxical interventions which will eventually help the client re
frame the difficulty and resultantly conceptualize new solutions.
Providing a Positively Framed Reason for

Paradoxical Instruction
Since clients typically present problem s that seem to them noth
ing but problem s, paradoxical interventions that suggest continu
ing the problem are more likely to receive therapeutic response if
some seemingly logical, albeit com pletely unexpected "rea so n " for
doing so is presented. Giving such a positively framed reason for
continuing and even increasing a problem behavior accom plishes
several things. At the social level, it encourages the client to go
along with the therapist in doing som ething unexpected and not
immediately understandable. At the psychological level, the client
is helped to accept the problem itself as serving som e useful pur
pose, to recognize it as an ally that may actually becom e an unex
pected m eans of achieving far more than sym ptom removal. O nce
the client accepts the notion that there may be som e benefit in
having and/or continuing the "problem ," an atm osphere is estab
lished in which the client, by shifting the previous frame of refer
ence, can begin to generate possible solutions never considered

before. These solutions may take the form of how the positive re
sult accomplished by the problem m ight instead be accom plished
in some creative, new way.
W hen offering positively framed reasons for following paradox
ical prescriptions, we do so sincerely bearing in mind the Erick
sonian principle that clients, at any given time, are always making
the best choice for them selves that they have so far learned to
make. All behavior, even problematic behavior, was learned for
some good reason and is continued in the absence of a better
choice in a particular situation. O f course the "re a so n " underly
ing the problem posed by the therapist may not always be ac
curate, but will at least stimulate the client to consider that some
positive reason exists for having and therefore continuing the
problem behavior in order to accom plish som e therapeutic aim.
Altering the Symptomatic Behavior

W hen the client has agreed to voluntarily continue engaging in
the problem behavior, the previous routine, in which the problem
was felt to be uncontrollable and com pletely negative, has already
been disrupted: The client will not be able to engage in the prob
lem in the same way once it is being viewed positively and done
volitionally. But to go several steps beyond this natural disruption
of the problem routine, it is very often desirable to alter the sym p
tomatic behavior in a therapeutic direction by either splitting the
client's experience of the problem , adding another (therapeutic)
behavior upon which the problem behavior is contingent, or m od
ifying the occurrence of the problem along such dim ensions as
timing, location, intensity, and so on.
O ne initial goal of such alteration is to provide the client with a
new sense of hope and control. Perhaps an even more im portant
goal is to create a context in which therapeutic alternatives can be
taught. The first step in teaching such alternatives often involves
a process of stimulating the search phenom ena described earlier
by suggesting som ething not readily understood by the client
such as, "G o ahead and have the ringing in your ears, but I won
der if you can tune yourself so that you d on't hear it." This direc
tive, paraphrased from one of Erickson's cases, suggested split
ting the client's experience betw een physiologically "h a v in g " and
psychologically "h ea rin g " the ringing. Though, upon hearing
such a suggestion, a client may have some vague notions about
what the therapist is implying it is more likely that intensive in
ternal searching will be initiated in which the client reviews expe
144 CHAPTER 6
riences to find some specific, personal understanding of the thera

pist's casually offered suggestion. We would consider this a
search for therapeutic alternatives; that is, for personal resources
that can be applied in new ways to solve an established problem.
Illustrating and Retrieving Needed Experiences

with Metaphor
Having confused the client's custom ary frame of reference con
cerning the problem with the paradoxical directive and having ini
tiated an internal search for specific ways of cooperating with the
suggested alteration, the therapist is in a position to stimulate the
client to retrieve needed resources and create a new map to follow
that includes therapeutic alternatives—an association of those per
sonal resources to the context of the problem. In this regard, we
find that metaphor or a series of m etaphors logically follows the
paradoxical directive and suggested alterations. M etaphorical
speaking indirectly stimulates the client's own thinking about
needed resource experiences and may model a possible map for
the client to interpret, modify, assimilate in part or whole, or re
ject if not relevant. It is not enough to simply disrupt the client's
routine with respect to the problem and consider therapy com
plete. We have only set the stage for the real work of therapy to
begin, to help the client develop a personally relevant and satisfy
ing map of conduct in which desired resources are available and
associated to stimuli in his or her social and problem network.
The case to be presented in this chapter thoroughly illustrates this
elem ent of paradox utilization.
Delivering the Paradox at the Beginning or End of

the Therapeutic Sequence
The logic of presenting a paradox at the beginning of a thera
peutic sequence has probably becom e apparent from the preced
ing discussion which em phasized that delivering a paradoxical in
tervention will cause the client to shift frames of reference and
initiate an internal search. W hen this searching does not result in
a readily understandable meaning for the client, search phenom
ena increase and the client displays signs of light trance. Receptiv
ity to additional direction from the therapist is increased at that
point; thus it is an optimal time for the therapist to present a se
ries of therapeutic m etaphors. These m etaphors may also contain
elem ents of paradox and confusion to overload the conscious
mind and allow the client to entertain novel experiences.
Delivering a paradox at the end of a therapeutic sequence is

often desirable as a way of reorienting the client and distracting
the conscious mind from excessive analysis of the preceding
work. At the same time, the paradox allows the conscious mind
something to analyze and retroactively frames the series of meta
phors in a com pelling manner. The client attem pts to understand
the metaphors based on the concluding paradox and at the same
time attempts to unravel the paradox's possibilities and implica
tions. We refer to this type of concluding paradox as one which
facilitates closure, an increased sense of m eaningfulness about
what went before based on the final paradox's com pelling nature.
Delivering in a Manner That Is Not Authoritative or

One Up
We offer this consideration about delivery of paradox based on
research by Weakland et al. (1974) which indicated that a con
fused, ignorant stance on the therapist's part (one down) seemed
to facilitate greater compliance by avoiding a high pressure ap
proach that might arouse client resistance. We tend to modify this
conclusion by recom m ending that therapists generally refrain
from being one-up or authoritative. Being indirect or m etaphori
cally paradoxical is certainly one way to accom plish this goal.
There are, however, those clients for whom binding conditions at
the outset or a challenging stance from the therapist will be the
most effective way to proceed.
SEEDS WITHOUT SUNSHINE?

THERAPEUTIC OUTCOMES
IN METAPHOR
We have m entioned that the paradox provides a stimulus that
reminds the listener of certain philosophical wisdom and possibil
ities of applying som ething previously unconsidered to the solu
tion of a problem or com pletion of a task. Simultaneously, the
paradox enters a slight confusion w hen a paradoxical situation is
posed or paradoxical logic or directives are given. The rules of ex
perience and com bining experience that once applied no longer
apply. Logic is arrested and unconscious process and search are
stimulated. Presenting paradoxes designed to stimulate uncon
scious search without directing or stimulating that search in some
manner would be similar to planting seeds in an area where they
146 CHAPTER 6
do not receive sunshine. Therefore it shouldn't be surprising to

find here a discussion of another im portant technique to be used
along with Ericksonian paradox: metaphor. Aristotle said this
about m etaphor: "T h e greatest thing by far is to be a m aster of
m etaphor and it is also a sign of genius, since a good m eta
phor implies an intuitive perception of the similarity in dissimi-
lars" (1954, p. 255). M etaphor provided a m eans by which Erick
son could gently guide the development of the growing seed-
ideas. More technically, we m ight say that m etaphor provides
controlled elaboration of ideas that are the subject of unconscious
search.
A strategic therapist begins the therapeutic process by assess
ing the individual and family system . This assessm ent is most
useful w hen it includes a decision about w hat therapeutic out
com es are desirable. The therapist m ust reach a tentative decision
on which life experience or growth process to elaborate. This deci
sion is reached by com bining what the client specifically requests
with the therapist's observations of related changes indirectly re
quested in the client's manner of presentation. This assessm ent is
not necessarily shared with the client; it is only a point of refer
ence for the therapist. It is im portant to em phasize, however, that
the client is the ultimate judge as to w hether a goal is relevant or
how any m etaphor that may depict a possibility of change will be
interpreted or acted upon.
O nce the therapist has form ulated those goals that, according
to careful clinical assessm ent, seem desirable and relevant for a
particular client-system , s/he is ready to design interventions to
facilitate those goals. These interventions will very likely include
paradox, especially as a m eans to "o p en a doorw ay" of experi
ence and temporarily confuse and suspend the client's normal
frame of reference with regard to the problem or goal. It is at this
point, as m entioned earlier, that each individual client is likely to
be m ost receptive to new possibilities or inform ation about how to
proceed. As Ericksonian therapists, we find that the m ost effec
tive and respectful way of com m unicating with such clients is in
directly, with m etaphor and anecdotes. In this way, we insure that
we do not im pose our values or goals on the client but only offer
the client a stim ulus to do his or her own thinking. Each client
who hears the same story, for example, will interpret it differently
according to his or her personal understanding. The m etaphors
and other interventions selected and designed will of course de
pend on the therapeutic outcom e goals form ulated by the thera
pist before the delivery of the paradox.
We have grouped these goals into six broad areas of therapeutic

outcom es: (1) family structure and social netw ork changes; (2)
changes in age, appropriate intimacy, and task behaviors; (3) atti
tude restructuring; (4) affect and em otional role changes; (5)
changes in self-im age; and (6) changes related to self-discipline and
enjoying living. A s we exam ine each individually, it will be useful
to bear in m ind that often the client-system may only request
change in one area and m ay do so indirectly by asking for sym ptom
removal. As therapists, however, we consider therapy to be com
plete only after we have seen evidence o f change in each area.
O ften this involves our specifically addressing each area w ith m eta
phors designed especially to facilitate change in that area.
Family Structure and Social Network Changes

M any problem s presented in therapy clearly indicate th e desir
ability o f change in this area. T h ese include m arital problem s,
fam ily difficulties, discom fort entering, dealing w ith, or appro
priately separating from a particular p h ase o f a fam ily cycle.
G oals here are usually stated in term s o f dealing m ore appro
priately w ith dem ands o f the current or next logical stage of fam
ily developm ent. D esired ch an ge in th e fam ily structure depends
o n th e unique culture, history, and goals of each fam ily bu t g en er
ally includes several featu res:
Changes in who is involved in the network.

C h an g es in w h o speaks to w hom .
Changes in what is spoken about.
Changes in the affect typically present and exchanged.
Changes in the subgroups that form and what they do.
Changes in role sharing and role support.
Changes in how needs are expressed.
Changes in how expressed needs can be responded to.
Changes in clarity of communication.
Implementation of communication aimed at relationships.
Implementation of communication aimed at goals.
Implementation of solution- and resolution-oriented communi
cation.
Changes in the use of direct and indirect communication.
Changes in the use of reinforcement and punishment among
members.
148 CHAPTER 6
O ther problems brought to therapy do not as readily indicate

that a family or individual expects change in the family or social
network. We contend, however, that even therapy with individ
uals for specific symptoms such as smoking, pain control, and so
on will be most beneficial w hen we help each individual client
and his or her family members prepare for logical changes in fam
ily structure or social network that will likely be stimulated as a
result of the symptom resolution. W hen we have only an individ
ual in treatment, we help that individual create a helpful interface
as s/he returns, changed in some way, to the environm ent of sig
nificant others who in turn will also be stimulated to change and
support changes made by the individual in therapy. Erickson was
very thoughtful about how even his individual clients would op
erate at interface points with others in their families and/or social
network system s, since the success of changes begun in any area
will depend on how well they are integrated into the client's en
vironmental circumstances. This is, then, the rationale for using
metaphors that sensitize even the individual client to the roles,
expectations, behaviors, feelings, and so on that will be called for
in the family and social network.
The Individual and Change

Regarding an individual as the client in family therapy, Erick
son did not take the psychoanalytic view (as did Akerman) as ju s
tification for seeing the individual while doing family therapy.
Erickson was both practical and ingenious. As most therapists
know, many symptoms presented in therapy which would seem
to be isolated from family structure matters are really only the su
perficial request for a much more pervasive therapy which fre
quently deals specifically with hidden problems in family struc
ture not yet disclosed. The nature of these family wide and indeed
social, networkwide changes need not be m ysterious. Frequently
theorists attempt to propose neat, systemic, hom eostatic, cy
bernetic, and rule-bound frameworks to analyze the tensions and
changes in a family system. Although we would be very pleased
to impose any neat and concise framework on families to cast an
air of scientific respectability onto our profession, Erickson fre
quently reminded us to treat each case uniquely. He went so far as
to say that he made a new theory for each individual.
He didn't consider symptom removal in one individual to al
ways result in a dysfunction in others (or the same person) in the
family. In fact, we expect that often the entire family improves
dramatically w hen the identified patient recovers from difficulty.
O ften a valiant effort at recovery in one person is an inspiration to

others to move ahead and grow. To assume in every case that
there is a sinister, systemic pathology at work in the family is to
bring to family therapy the same morbid attitude toward difficulty
that the medical model helped foster in the early theories of indi
vidual personality.
There is, however, a logic of dynamic interplay which m ust be
recognized at the level of system s theory, and that is by virtue of
nothing more (and nothing less!) than the interdependence of
each person on another. So we m ight best characterize Erickso
nian attem pts to intervene in a family as explained by the practi
cality of the unique family and how the persons in that family
necessarily depend upon one another. The social play occurring at
the interface points created betw een members is typically exam
ined by asking the following types of questions: Does the hus
band support roles needed by the wife and vice-versa? Is there a
role complementarity? Do the changes being stressed, shaped by,
or expected of one family m ember enjoy a supportive other? The
role behavior that may or may not be supported shapes the sanc
tions that will befall the family and hence shapes its destiny be
cause in the process of growth the members will m eet others and
m eet them with altered emotional flexibility, altered behavior, al
tered self-image expectations, altered attitudes, and so on. These
alterations becam e opportunities for others to support, reject, in
terfere, model, reinforce, punish, and more. Tim s in working
with the individual only, Erickson was certain to help that indi
vidual interface with im portant others and maximize those m eet
ings through the changes s/he was making or had made. In other
words, Erickson would help an individual change in order to fos
ter systemwide changes.
Family conflict is often viewed in light of the incongruities
created by role taking within the family and extending beyond the
family to interfaces with the broader social network. The way the
person forms and uses interfaces is the key to doing family therapy
with the individual in an Ericksonian style. The boundary between
separate individuals in the client's world can be shaped in many
ways. First, the individual's feelings, behaviors, expectations, and
perceptions with respect to any particular other can be changed
directly as the hypnotherapist deals with him or her. Further, the
behavior of, say, a wife can be shaped by the selective behavior of
the husband w hen only the husband is willing to come to therapy.
This latter case is of course more difficult. But the behaviors, in
terpretations, em otions, and perceptions of the wife toward the
husband are not as difficult to change via the husband's selective
150 CHAPTER 6
behavior w hen he has changed in therapy. This is true for any

com bination of persons in the network. For exam ple, w hen work
ing with an elderly woman for the reduction of psychic pain, we
simply suggested that she entice her avoiding daughter in a partic
ular m anner: By so doing, she easily created in the daughter and
son-in-law a change in their perception of her from malingering to
inspiring. The m other also found a way to shape new approach
behaviors in the daughter and son-in-law and new feelings of to
getherness were experienced by all. In this case, working to change
the family via the identified patient was probably quicker and more
successful than a family therapy and insight-confrontation ap
proach would have been. The daughter and son-in-law considered
it their idea to approach the elderly m other in a new manner. They
did not lose face, feel evaluated, find it necessary to defend the
previous avoidance, or any other such eitrogenic response, be
cause they were never even in the office!
To help clients accom plish changes in perception, em otion, and
behavior to the extent that necessary interface m aneuvers will be
learned, it is im portant for therapists to be effective in ways that
extend beyond the focused awareness of the family therapy m eta
phors. The five additional categories that follow are briefly ex
plained and will then be illustrated in the case analysis to come.
Age Appropriate Intimacy and Task Behaviors

Related to changes in family structure and social network are
changes in specific behaviors necessary for age appropriate inti
macy and age appropriate developmental tasks. G oals are fre
quently stated in term s of change in perceptions, experiences,
and transactions involved in approximating the chronological age
appropriate intimacy and task situations that could be achieved by
the client. Again, presenting problem s may or may not point to
obvious changes in this area. Com plicating this m atter is the
therapeutic task of determ ining an individual's psychological age
with regard to specific areas of functioning. As the psychological
age is assessed, logical behavioral deficits will usually be appar
ent. These deficits will often be clearly related to the presenting
problem . In the case discussion to follow, the client's presented
difficulty with "au th ority " was so pervasive that h e couldn't re
late appropriately to his boss, peers, or wife and one could easily
assum e that this difficulty would soon be evident w ith children he
m ight father. It was also clear from his adolescent m anner and his
description of competitive relations with his own father that be
haviors related to trust, openness, m aking m istakes, and expres

sions of vulnerability had not been learned naturally and were not
available to him in those situations and relationships he consid
ered problematic.
O nce psychological age and typical behavioral deficits are as
sessed, the therapeutic task is to systematically teach and model
desired behaviors, usually in metaphor. By so doing, those behav
iors (perceptions, experiences, transactions—either internal or ex
ternally observable) which were not learned in natural develop
m ent are made available to the client as s/he experiences, identi
fies with, and assim ilates them such that they can be used at
appropriate times and in appropriate situations.
In some cases, the behaviors to be learned, determ ined by
assessing developmental age and associated tasks, may not be as
clearly related to the presenting problem . Such m ight have been
the case if our client had asked for help m anaging physical pain.
Still, we assume that m aking more mature, age appropriate be
haviors available through m etaphor will expand the client's op
tions with regard to any problem , though perhaps in ways be
yond those which we can anticipate.
Attitude Restructuring
Central to any lasting therapeutic growth are changes in per
ception and in the belief system , at the very least in those percep
tions and beliefs that supported the problem . Change is often
facilitated by helping the client alter the priority of perceptions
and the weight or cognitive interpretation given to perceptions.
This includes changes in the client's assum ptions which may have
been used previously to justify or make predictions about the
problem and its continuation.
Attitude restructuring is necessary w hether the problem is pri
marily somatic or more overtly interpersonal in nature. Believing
that a skin condition is genetic and therefore incurable is an atti
tude that needs to be changed to enable the person to congru-
ently retrieve and apply the personal abilities that m ight result in
cure. And, changing that attitude enables the client to look for
and find evidence of the expected change. W ith respect to the cli
ent to be discussed in this chapter, his limiting belief was that
success or winning were only possible through superiority or re
bellion against authority. Here the attitude change indicated was
one that allowed the client to com fortably incorporate and utilize
the new behaviors and affects addressed by the other m etaphors.
152 CHAPTER 6
For this client, paradoxical intention was useful to confuse his

typical attitude and present a contrasting one w hose logic was not
immediately apparent.
Affect and Emotional Flexibility

In this category goals are usually conceptualized in terms of the
client's understanding and use of em otions necessary to achieving
the contracted therapy goals, including supporting perceptions of
internal physiological com ponents of the em otions. We begin
from the assumption that any client has, by virtue of having
lived, some experience of or at least the potential for experiencing
all the em otions. The difficulty often lies in the area of emotional
flexibility or the ability to use personal capabilities in particular sit
uations. Erickson summarized it this way w hen he said: "P sych o
logical problems exist precisely because the conscious mind does
not know how to initiate psychological experience and behavior
change to the degree that one would lik e" (Erickson & Rossi,
1979, p. 18). So, our therapeutic task is one of first retrieving and
then associating these "psychological experiences" to situations
where they are needed in order to facilitate flexibility of this "in iti
ation" process. The purpose of initial m etaphors, however, is sim
ply to retrieve, build com ponents of, or stimulate the desired ex
periences or affective states, preferably by discussing situations
far removed from those areas in which the client ultimately
wishes to use them . (The association step is usually m anaged by
other m etaphors from categories such as self-image thinking,
family structure change, or age appropriate intimacy behaviors.
A factor that influences the content of m etaphors designed spe
cifically to retrieve affect is the psychological age of the client. Tell
ing stories about experiences com m on to (or prior to) that age will
run no risk of being too psychologically sophisticated for com pre
hension and congruent identification. W ith the "au th ority" client
who follows, for example, m etaphors designed to retrieve com
fortable dependency feelings, trust, and a sense of com petence
were constructed bearing in mind the developmental period of
adolescence, since this seemed to be the stage of development in
which he was fixated. Thus we talked to him about dependency
experiences every adolescent enjoys (riding in the front of a ca
noe, playing team sports) and appealed to his sense of com pe
tence and trust from even younger periods in " a boy's life" (learn
ing to dress him self and tie knots) to enable us to proceed with
utmost confidence that we were describing experiences he could
powerfully experience.
Self-Image Thinking Enhancement

Goals here are generally stated in terms of improvement in the
client's ability to anticipate him- or herself using new options in
various situations, either privately or with significant others. This
requires the ability to imagine, form ulation of an awareness of the
self as worthwhile, and a belief that certain new situations would
be beneficial, or that it would be desirable to have certain behav
iors and qualities ("psychological experiences") newly available in
situations that have been typically problematic.
M etaphors or more direct m ethods designed to stimulate and
direct these abilities are an almost essential com ponent of any ef
fective therapy. They provide an opportunity for clients to prac
tice or mentally rehearse enacting in professional, familial, or so
cial situations those learnings that have been stimulated by the
therapy. They help the client to literally make a new map that can
be depended upon to shape future perform ance. Images guide
performance and self-image thinking provides a thoughtful alter
native to automatically relying upon residual images from the
past.
And finally of course, self-image thinking m etaphors, by stimu
lating fantasies of typical situations in the client's future paired
with new behavioral and emotional options, becom e an important
associational link that helps clients accomplish the step of "initiat
ing psychological experiences and behaviors" in situations where
this had previously been difficult or impossible. This is clearly il
lustrated in the transcript to follow w hen the client listens to a
metaphor which indirectly stim ulates him to visualize him self in a
new way, that is, with those behaviors and affective states that
w eren't typical in his interactions, and to study this new image of
him self through a variety of situations in which he had previously
behaved inappropriately. In this mental rehearsal, however, he
was led to visualize him self interacting in new ways which
"pleased and surprised him ." The metaphoric content involves
another client who is studying his reflection in the window of a
train.
Enjoyment and Discipline in Living and Changing

Clients typically come to therapy for help in solving some prob
lem and they often phrase goals in terms of what they want to not
do or behaviors they want to "g e t rid of." Very few com e to ther
apy specifically in order to enjoy life more, though in the instance
of depressed clients, this goal is often clearly indicated. But re
gardless of the presenting problem, every client can benefit from
154 CHAPTER 6
interventions designed to stimulate processes leading to greater

enjoym ent in life.
Involved in the process of enjoying life is the m atter of self-
discipline, though this may sound paradoxically illogical to some.
Erickson was an avid proponent of both enjoym ent and discipline
and frequently paired the two activities. He explained to one cli
ent ("N ick ") in no uncertain term s that life's gifts are earned and
merited: "You reach an understanding that every happiness is
earned and, if given to you, it's merited. Because there is no such
thing as a free gift; you have to earn it or you have to merit it.
And merit requires labor and effort on your p art" (Erickson in
Lustig, videotape, 1975). We observed Erickson working with a
client who was eventually stim ulated to laughter (enjoym ent) in
the trance state. At the m om ent the client was laughing hardest,
Erickson introduced the m atter of discipline, suggesting that he
"have that experience (the enjoym ent) next time you make up your
mind about what you're going to do" (Erickson, August 7, 1977). This
example illustrates that self-discipline n eed n't necessarily be
drudgery or a burdensom e responsibility, but som ething that can
be accomplished with enjoym ent.
Erickson frequently enjoyed surprising clients and students
with his own very well-developed sense of hum or and in that way
modeled enjoym ent for them . Very little needs to be m entioned
about the way in which Erickson modeled an exacting self-disci-
pline. His entire life and many accom plishm ents—despite his
physical infirmity—are testim ony enough of that. M etaphors de
signed to stimulate an appreciation of living, even during the
process of change, can also include associations to the necessary
and proper am ount of discipline that a client needs to acquire in
order to achieve defined or implied goals. The ratio of these two
elem ents to one another will vary considerably of course from cli
ent to client. W ith the client to follow, for example, though disci
pline/enjoyment w asn't specifically addressed in the one session
presented here, a good bit of self-discipline needed to be modeled
and stimulated in therapy in order to help him learn to have en
joym ent by cooperating and being responsible. O ther clients may
be neglecting enjoym ent of life by being overly responsible not
only for themselves but for others. In cases like this, the task may
be som ething akin to what Erickson told the wife who was over
protecting her handicapped husband. He indicated to her that her
most im portant task was to teach him that he didn't need her:
"And I wanted your husband to realize that there is a lot he can do
for h im s elf.. . And it isn 't right for me to tell him , 'You learn this
or you learn th at!' Let him learn whatever he wishes, in whatever or
der he wishes" (Erickson & Rossi, 1979, pp. 113, 115). W ith the
same woman, who expressed her love for her husband by worry
ing over him, Erickson shared a personal anecdote, that he (also
handicapped) encouraged his wife to show her love for him by
enjoying herself since there was no real need to w aste time being
concerned: " I'm just in a wheelchair. That's all! I want her to put
her energy toward enjoying th ings" (Erickson & Rossi, 1979, p.
109). At the end of their session, he em phasized this learning
again: "And neither of you need to be concerned about the other.
You need to enjoy knowing each other. And enjoying what you can
do as meaningful to you" (Erickson & Rossi, 1979, p. 119).
CASE EXAMPLE
We have covered various elem ents of paradox that illustrate its
significance and im portance. We now want to turn to examples of
actual interventions and their effects. We discuss their selection
and the design of a treatm ent plan in which they can be thought
fully delivered. Since much of the literature about paradox ad
dresses sym ptom prescription in the context of strategic family
therapy, we will focus on Erickson's style involving the use of
lesser noted forms of paradox in other therapeutic contexts. We
are em phasizing the use of paradox for both interview m anage
m ent and therapeutic outcom e. Specific interventions include
binds, confusion techniques, indirect suggestion, and especially
therapeutic metaphor. Ericksonian hypnotherapy is a therapeutic
modality particularly well suited to illustrate both variety and fre
quency of these interventions while carefully exam ining the cli
ent's response to them . For this reason, we have selected a case
for analysis in which hypnosis was the primary modality.
History
Frank (male, age 29, married) requested treatm ent to solve his
problem with "authority." He said that he had difficulty taking
orders and following rules, and could not accept feedback, even
constructive feedback from his wife. As a professional therapist
him self, he was aware that this situation was not appropriate. He
needed to be com fortable with criticism and he realized that his
reflexive rebellion was out of his control. He revealed that he had
a num ber of traffic and parking tickets which he had not paid,
had served a brief jail sentence in this regard, had broken mari
juana laws and been caught, and had argum ents w ith his wife,
156 CHAPTER 6
peers, and employer. He smiled and attem pted to induce laughter

throughout the interview. He in fact did not merely laugh; he at
tem pted to have us laugh with him as he "cu te ly " related his dif
ficulties. Finally, w hen asked about his father, he said that their
relationship had been strained, that his father had been openly
competitive with him, and frequently "m otiv ated " him with chal
lenge and provocation.
Frank was either one down in the challenged and ridiculed po
sition or he behaved in a manner typified by an overt mild hostil
ity and coldness. He said usual conversation between him self and
his father centered on a shared attitude of disgust about "stupid
ru les" imposed on themselves and others by "ignorant people."
In these cases, again the affect was hostile and in no case did the
relationship establish roles that would support the sharing of de
pendency needs, fears, weaknesses, scares, doubts, and reassur
ances so common in most developing children. W eaknesses were
hidden, taken to be attributes of others instead of the self, and in
m ost cases were exploited as a basis of ridicule.
Initial Treatment Plan

O ur initial planning included interventions to stimulate Frank's
com fort with his dependency needs, an alteration in his self-im-
age thinking in order to enable him to im plem ent his own use of
those feelings, and an acceptance of him self not based on his
father's provocation and encouragem ent of rebellious behaviors.
Thus new maps for his experience would provide a basis for
Frank to make changes in his relationships and better approx
imate an age appropriate intimacy. To help him accept feedback
from others, we needed to prom ote in him a sense that his needs
were actually socially acceptable. He would need a map of con
duct about expressing those needs and he would need to con
clude that compromise does not m ean loss. A new sensitivity to
acceptance of some of his dependency needs paired with a re
hearsal of how he might express those needs openly and even
enjoy doing so could be effected by stimulating his own thinking
through the use of m etaphors.
Initial planning com pleted, we were ready to induce trance and
share m etaphors about succeeding in enjoying the game of life by
breaking the rules and winning, about learning and rehearsing
self-acceptance, and about the fantasies, em otions, and self-talk
that a man could have w hen a father was close and supportive.
Each metaphor, however, would contain an elem ent of paradoxi
cal logic. We would therefore take nothing away from the client
but rather follow Erickson's directive: "O n e always tries to use

whatever the patient brings into the office. If they bring in resist
ance, be grateful for that resistance. Heap it up in whatever fash
ion they want you to—really pile it up. But never get disgusted
with the amount of resistance" (Erickson & Rossi, 1981, p. 16).
Orientating the Client to Trance

C: Have you been in trance before, by the way?
F: Well, er, ah, yeah. . .
C: You say that with some incongruence. What?
F: Well, I've never had someone who knows what they're doing.
S: So I'd suggest you lean back in your chair...
C: to find out how comfortable you can be in the cool air with your feet
flat on the floor.
S: A lot of times a person goes into a trance by staring at something.
C: There are plenty of things out there you could stare at, a few feet in
front of you. . .
S: And just find some one spot.
C: .. .or far away.
S: You don't need to move.
C: It doesn't really matter what you think about or what you notice.
S: I wonder if you're going to fail to succeed at not going into trance at
exactly your own speed or whether you won't.
C: And hypnosis, as you know, is a way for a person to learn from
himself and so I hope that you succeed in rebelling to the proper
extent against those suggestions we give so as to make certain that
you only follow those suggestions that are relevant for you to follow.
And just staring at one spot, beginning to feel the comfortable relax
ation that you know how to feel, listening to one thing we say, being
aware of the people, the place, the context. It doesn't matter whether
you go into trance with your eyes open or close your eyes. Wait until
the proper time to close your eyes, but you can begin to enjoy the
freedom of having precisely your own thoughts in response to those
stimuli that we offer.
S: And I think it would be very interesting if you try to keep your eyes
open throughout the entire trance.
C: So enjoy the freedom to rebel because you can't help but rebel and
do what is uniquely appropriate for you to do. You can wonder
about that.
The initial paradoxical instruction to Frank stated that it is im
portant for him to rebel and have his own thoughts. This was
expressed in the sentences: "H ypnosis, as you know, is a way for
158 CHAPTER 6
a person to learn from him self and so I hope that you succeed in
rebelling to the proper extent against those suggestions we give
so as to make certain that you only follow those suggestions that
are relevant for you to follow." W hat is paradoxical about the sug
gestion is o f course best illustrated by studying the options availa
ble to Frank at that point. Frank could follow the suggestion and
have his own thoughts rather than ours or he could rebel against
the suggestion that he have his own thoughts (not follow our sug
gestion) and therefore be thinking a thought of his own. This par
adoxical sym ptom prescription is a bit like the classical paradox:
"Everything I say is a lie." But better, the latter sentence is ex
tremely personal to Frank and he is bound to consider it.
How is this therapeutically useful for Frank? Perhaps it is not
useful to him therapeutically, but it is useful to the therapeutic
process as a device for interview m anagem ent. We have ad
dressed Frank at several levels. Consciously, we have increased
rapport to the extent that he recognizes the acceptance of his re
bellious behavior. We have reduced his need to be defensive and
encouraged him to continue his defensive behavior; we have al
lowed him to protect him self if he deem s it necessary to so do.
Below the level of consciousness, we have accom plished at least
two things. We have stym ied his m ost familiar m anner of resisting
feedback from others by the logical overload on his conscious as
sociation process and thus encouraged unconscious search. In a
subtle manner, we have focused his aw areness on our words and
his personal reaction. We have defined the area of experience to
w hich attention m ust be given: W hether he rebels or complies
with our instruction, he can only determ ine the difference by
noticing our words and his reaction. Thus we have helped him
accom plish several things that will deepen his trance experience:
intensified rapport, accom plished fixation of his attention, and in
tensified concentration on internal experience.
To effect the paradoxical nature of the induction then we asked
him to "fail to succeed at not going into tran ce" at his own speed,
to "su cceed in rebelling," to "e n jo y the freedom of having pre
cisely his own thoughts," and to "try to keep your eyes open
throughout the entire trance." Intim ately tied to the paradoxical
nature of the instructions was the construction o f and effects
created by indirection, confusion, and the therapeutic bind aspect
of the com m unication.
Som e other features of the induction, although not paradoxical,
are worthy of attention if for no other reason than to underscore
th e point that the use of paradox does not singularly constitute a
treatm ent. Rather, paradox is a tool that m ust be considered
against a background of treatm ent methodology. W hile this m eth

odology is explained at length elsewhere (Lankton & Lankton,
1983), we will discuss these features briefly here as a point of ref
erence for further study and as indication of the num erous other
aspects of interpersonal influence. These aspects, as we have pre
viously stated, exist to support the im pact of paradox in an Erick
sonian approach. In this discussion then, we attend to the con
veyance of paradox through the use of indirect suggestions and
therapeutic binds as well as metaphor.
Conscious/ Unconscious Dissociation

In the course of creating an induction it was im portant for
Erickson to maximize the degree of dissociation betw een the con
scious and the unconscious and establish a dissociated hypnotic
personality. "O n ly in this way can there be secured an extensive
d isso ciatio n .. .w hich will perm it a satisfactory m anipulation of
those parts of the personality under stu d y " (Erickson, 1980, p. 7).
In order to achieve such a dissociation we frequently structure our
speech in the m anner that follows.
C: Your conscious mind has a certain line of thought. You bring various
concerns, various interests and they have nothing to do with your un
conscious processes.
S: Your conscious mind is in the process of wondering, watching what
is going to happen. . .
C: But try as you might. . .
S: while your unconscious mind is in the process of taking care of a lot
of things for you that you don't appreciate.
C: .. .you have very little control over just what those things are and
how your unconscious goes about taking care of them.
S: It would be a good idea for you to take control of your responses to
the things that we say and sooner or later you develop your own
trance.
C: Your conscious mind has one idea about the depth of trance, won
ders. ..
S: Your unconscious mind has some other idea about the depth of
trance that you need or may need.
C: .. .whether or not you're responding appropriately.
S: And thoroughly understand the depth of your own experience.
C: But your unconscious mind automatically finds that depth of trance
that's relevant...
S: Now somewhere in the back of your mind are your thoughts. . .
C: . . . and your conscious mind discovers later.
160 CHAPTER 6
S: .. .you haven't known that you were going to think.

C: Of course you know that it's impossible not to have thoughts. You
have thoughts as we speak about one th ink...
S: I don't know whether you remember or not. . .
C: . . . and perhaps another thing simultaneously.
S: . . . that there are billions of brain cells and you've only had enough
time, at the speed of thought, the speed of electricity. . .
C: And listening to Stephen's voice, you can wonder...
S: .. .to notice some of those thoughts.
C: .. .just what he's getting at, what you can learn, learning things that
perhaps you already know, things you'\ e known all along but didn't
know that you know.
The reader will note that the transcript reveals an intertwining
of suggestions from both speakers. In fact, there are moments
w hen one speaker com pletes the suggestion begun by the other.
More often, each speaker continues to deliver his or her sugges
tion completely but in sectioned phrases. The aspect of dual in
duction was achieved here as one and then the other speaker al
ternated speaking and thus alternated the delivery of first one and
then the other section of the com plete suggestion. The paradoxi
cal directives at the initial stages of the induction depotentiate the
normal critical aspect of Frank's conscious mind. This conscious-
unconscious induction now unencum bered by that usual frame
work helps the client intensify his awareness for and sensitivity to
his different thought processes. His heightened appreciation for
unconscious processes thereby creates a situation of expectancy.
He has observed a difference in his ability that he may not have
previously noticed. This perceived ability becom es the seedbed
for developing therapeutic resources that are the real meat of
therapeutic change.
Ratify and Deepen Trance

The stage of induction that typically conveys to the client an
awareness that an altered state has been achieved is referred to as
ratifying trance. You want the client to select one bit of trance
phenom ena out of many and confirm it. ''W h en [the client] does
that, [the client] is also confirming or validating the others, and
that's what you want your patient to d o " (Erickson & Rossi, 1979,
p. 371).
S: Your heart is pounding and there are a lot of other things that are
beating. And while we've been talking, your cheeks have flattened
out.
C: You haven't moved.

S: And you can only notice some of those thoughts.
C: You've closed your eyes but your unconscious continues to maintain
your balance, breathing, and you can consciously alter your breath
ing, take a deeper breath. . .
S: But you should notice the enjoyment because as your thoughts come
out of the back of your mind, the more relaxation you feel, the
deeper you go into trance. And they'll come out of your mind.
C: .. .but your unconscious automatically resumes the process of regu
lating your breathing rhythmically in a way you can forget about
completely.
S: Going into a trance should be a matter of noticing that you can be
aware of a lot of things that come out of your mind. And there's no
reason to worry about taking in ideas that will make you go out of
your mind.
C: Your conscious mind might enjoy being able to monitor your prog
ress into your trance by noticing how you can go one twentieth of the
way into trance with each count. Not much difference between 20
and 19.
S: Sometimes you might want to find out that you're failing to go into
trance by increments.
C: Eighteen, 17, 16.
S: You're failing to know that one of your hands can get lighter.
C: And how can you really be sure. . .
S: Are you aware of which of your hands will fail first to get lighter?
C: . . . that you really aren't going one twentieth of the way, 1 5 ,1 4 ,1 3 ...
S: And at least one of your hands is going to fail to raise up to your face
in small increments of cataleptic movement.
C: .. .or that one of those hands really isn't becoming lighter.
S: Now you have a number of various ways you can do that.
C: And a hand can become lighter, even heavier, but the first thing you
notice is that altered sensation. And just what area will you notice
that sensation first?
S: It's lightening to know that altered the sensation and you were
pleased to find out. There are a number of ways in which you can fail
to let one of your hands. You can fail to notice one of your hands
getting lighter.
C: Will it be in the fingers? The elbow? Or maybe you'll notice it by the
change in pressure that your thigh can sense as your hand rests less
and less against it, 10.
S: Your hand can get lighter and you can fail to notice. Your hand can
fail to get lighter and you can notice. Your hand can fail to get lighter
and you can fail to notice or your hand can get lighter and you can
notice.
162 CHAPTER 6
C: Seven, 6.
S: Maybe your hand is going to feel lighter as it raises halfway to your
face or maybe a quarter of the way to your face. Maybe it's going to
get heavier and fall down to the side of the chair.
C: And that would be another interesting adaptation, another interest
ing way to follow your own rule, 4, 3.
S: And perhaps it's going to stay where it is. So we're just going to
have to find out.
In this segm ent the typical alterations of initial trance are ob
served and described to the client: heartbeat alterations, cheeks
flattening, lack of movement, deeper breathing, and relaxation.
Following this summary of obvious occurrences, we began deep
ening trance by suggesting more uncom m on trance phenom ena.
Since Frank was oppositional, we continued to use paradoxical
behavior prescriptions, challenging him to deepen his trance. This
is found in the lines "Fin d out that you're failing to go into trance
by increm ents"; "You're failing to have one of your hands get
ligh ter"; and finally, "At least one of your hands is going to fail to
raise up to your face." Again, the paradoxical elem ent lies in the
binding quality of the situation. We have appealed to the com m on
motivational relationship that his father created with him —chal
lenge. Thus we have appealed to his transference need to both
please and rebel against us. If he rebels he will have achieved a
deepening by creating the trance phenom ena. If, contrary to our
expectation of failure, he does fail, then he has followed our sug
gestion. Following our suggestion will thus lead to deepening be
cause we subsequently suggest that he doubt his ability to not
raise a hand, not go deeper, and that he dissociate further. The
sentences used for this are: "Y ou're failing to know that one of
your hands can get ligh ter"; and "H ow can you really be sure?"
"T h ere are a num ber of varieties of ways in which you can fail to
let one of your hands." "You can fail to notice one of your hands
getting lighter." And finally, a series of all possible alternatives of
the above was suggested ending with "Your hand can fail to get
lighter and you can fail to notice."
S: And you don't need to listen to the things a hypnotist says. You have
a lot of your own thoughts.
C: Your conscious mind may be interested in analyzing the structure of
what we say and w h y ...
S: And your thoughts are worth examining.
C: .. .but that has nothing to do with your true purpose in coming and
presenting that difficulty...
S: Slowly, still lighter, and thoroughly.
C: . . . and your conscious mind had no way of predicting ahead of time

just what you'll learn but you can wonder and be interested or maybe
your conscious mind won't be interested at all in what we s a y ...
S: And your thoughts are worth sharing and responding to but since
they are your thoughts, it's an opportunity to go into trance and ex
amine your own thoughts...
C: . . .because your own thoughts are so much more your own.
S: You use the hypnotist's words to stimulate your own thinking...
C: Three, 2.
S: . . . and you accept those suggestions you think would be useful, but
otherwise, the rule is to discard those suggestions and do things
your own unique way for your own betterment.
We com pleted the deepening begun in the earlier segm ent by

suggesting increm ental changes which them selves were too small
to detect. In the first sentence of this segm ent we again prescribed
the resistive behavior that we expected the client to automatically
engage: "You d o n 't need to listen to the things a hypnotist say s";
and "Your conscious mind may be interested in analyzing the
structure. . . " If Frank uses these usually defensive m aneuvers he
has done so under our direction and we therefore have control. If
instead, he rebels against them then he is listening and cooperat
ing and we again have control of the therapy. This paradoxical
bind is continued at every phase of the treatm ent to m aintain our
ability to manage the interview in the therapeutic direction. It is
continued up to the final sentence that defines in Frank's own
vocabulary "T h e rule is to discard those suggestions and do
things your own unique w ay " but true to the add-on aspect of the
Ericksonian style continues with the subtle phrase "fo r your own
betterm ent."
It is again notable that the Ericksonian form of paradoxical
sym ptom prescription is not viewed as an end but a m eans to the
therapy. We m aintain the control and focusing of attention
throughout the interview while accepting Frank's rebellious inde
pendence and frame his behavior as "fo r his betterm ent." We thus
gain and fixate rapport and becom e indispensable (as well as cred
ible) sources of inform ation. A fter all, he needs to pay attention to
us in order to discover how he has been doing the correct thing by
doing what he thought was incorrect.
Multiple Embedded Metaphors

Goals for the following set of m etaphors included helping
Frank gain an experiential understanding of the feelings, role, at
164 CHAPTER 6
titude, and self-image foundations of cooperation. These are of

course in keeping with the client's request, but achieving this per
sonality goal is contingent upon his learning to understand and
accept dependency needs, experience parental acceptance, re
hearse specifics of cooperation, create an emotional connection
with his wife, and build certain behavioral skills. The several met
aphors chosen were expected to help Frank build an experiential
map of conduct in each of these areas. Each was designed to ap
peal to his need for rebellion and each, therefore, contained para
doxical elem ents to facilitate the learnings or gains; they are sum
marized as follows:
1. Prepare him to have dependency needs (affect) and examine
them in relative com fort.
2. Learn to express such needs in intimacy encounters (age ap
propriate intimacy behavior) with significant others.
3. Create an expectation of normal parenting to help prepare
him for his role in fatherhood (family structure change).
4. Learn to conceive of new roles and feelings with pleasant an
ticipation in several areas of life (self-image thinking enhance
ment).
5. Prepare him to change his understanding (attitude restruc
turing) about how cooperation rather than com petition results
in success.
Commentary is interspersed throughout the development of
the m etaphors in this continuing transcript. C onsistent reliance
on paradoxical intention in suggestions to maintain trance is obvi
ous as are the more novel elem ents of paradoxical logic within the
metaphoric drama.
Begin Matching Metaphor

C: And so just waiting, we can talk about anything. You know only a
fool would fail in such a way as to fa il...
S: .. .but a wise man fails in such a way as to succeed.
C: Now Fred couldn't have been more clear about this fact at 40 years
old. . .
S: . . . but when he came to see us at 30, he had a good deal of bad deals
he needed to deal with. And we made a deal with him. If he would
put his cards on the table, we would see his bet and up the ante.
And the more deeply he went into trance, the more deeply we put
him into trance.
C: Fred had enjoyed those affairs a great deal. It was a different story
after his wife found out.
S: He came to therapy perplexed with the idea that he had had a num
ber of affairs for a number of years which were no difficulty to him
emotionally, so it was incomprehensible to him that when he de
cided firmly and resolutely to stop the affairs, he felt guilty.
C: He knew the game was up. He wasn't going to have the affairs but
he didn't want to have the guilt.
S: So he came to psychotherapy after he decided he had been punished
long enough and it wasn't going to go away. He came for our help. It
was a good reason to see him for therapy. We accepted the case. We
asked him to go into trance.
C: It was a pleasurable activity to go into trance even with concerns on
your mind.
S: And to develop the depth of trance you think you want or may want
is an easy matter in the therapist's office.
C: And the last client we put into trance didn't realize how fully he was
in trance until we asked him to come out of it.
This opening metaphor sets the stage for Frank to begin the
process of questioning his prevailing attitude about com petition.
It also contains six m ajor points of interest with regard to para
dox. First, "O n ly a fool would fail in such a way so as to f a i l ...
but a wise man fails in such a way as to succeed," defines the
story as paradoxical in nature. The com m ent about how a "fo o l"
fails should hook Frank's need to rebel w hen challenged. If he is
hooked, we have furthered the trance by facilitating his fixated
attention and internal concentration. We have increased the likeli
hood that Frank will becom e interested in the story in order to
determine his position on the issue of foolishness, success, fail
ure, and rule breaking. The key elem ent in such a "bull's ey e"
story is the contradiction of succeeding by failing—hence the
metaphoric paradox.
Second, "Fred couldn't be more clear about this fact at forty
years old," challenges Frank to understand or admit he is still
"yo u n g ." O r perhaps Frank will be motivated to "b e a t" Fred by
learning whatever it was that Fred learned (which is yet to be re
vealed) much faster than it was learned in the story.
Third, "T h e more deeply he went into trance, the more deeply
we put him into trance," denies our control of the hypnotic situa
tion. This is a paradoxical bind for trance m aintenance. If he
rebels he gives us control; if he complies, then he follows sugges
tions and definitions created by us and hence he goes into trance
more deeply.
Next, "H e had a good deal of bad deals" is an oxym oron—a
com bination of contradictory or incongruous words. As such, it
will subtly stimulate a new examination of what really constitutes
166 CHAPTER 6
a good or bad deal. It plants a seed for more detailed illustrations

to follow about how w hat initially seem s negative (like depen
dency) is a necessary and positive aspect of mature functioning.
Also, "W h en he decided firmly and resolutely to stop the af
fairs, he felt guilty" (and not vice versa), is another paradox. It
defines the situation as one involving rules of marital com m it
m ent and intimacy.
And finally, "T h e last client we put into trance didn't realize how
fully he was in trance until we asked him to come out of it" is again
a paradoxical bind for trance m aintenance. He can rebel and realize
now that he is in trance or if he does not realize this, he implicitly
complies with the suggestion that he not realize now but rather
later. Compliance now of course sets him up for com pliance later.
So, w hether he now believes h e is in trance or now believes he is
not, he m ust admit that he is following our suggestion.
Begin Resource Retrieval

S: Now he was a person who, going into trance, exhibited the degree of
scare that he had in a number of interesting ways that your uncon
scious would know more about than your conscious knows. I don't
even think that his conscious mind could stomach the awareness.
And we never brought it to his conscious attention. He had had a
belly full of that in his life and he didn't need any more.
In this transition segm ent we refer to the possibility of re
pressed fear that Frank experiences due to the constancy of pres
sures brought on by the situation of rebellion and rule breaking
referred to in the previous segm ent. We focused his awareness on
his stomach and presented an altered interpretation that this re
pressed fearfulness constituted (no pun) a belly full of distress.
This may be his first indication that we are " o n his sid e" in an
effort to relieve the distress he has endured. In light of this new
interpretation we now continue to shift the client's framework
even further by temporarily suspending the m atching m etaphor
and pursuing another m etaphor about another client that will al
low Frank to experience the resources needed for achieving the
goals summarized earlier.
C: And speaking of that symbolic representation of his difficulty...

S: . . . we began to speak to him about another client we had worked
with.
C: Phil had had an ulcer for five years. He had never spoken about it to
anyone.
S: He was a medical resident who was well known in his community.
C: He had a lot of responsibility professionally, and besides that, he was

the father of five children.
S: In the intake interview it became apparent that his ulcer problems
had begun when he found out that his wife was pregnant with the
fifth child.
C: It seemed as if there was absolutely no time in Phil's life when he got
to be the one taken care of. He was always taking care of some
body—from the dog or the child or that inexhaustible supply of psy
chiatric patients.
S: Building additions onto the house, reading professional journals.
C: If it wasn't this, it was that. There's never enough time, he had been
heard to say.
S: All those ideas.
C: He wanted to relax. He thought that would be a great idea. But I just
can't afford the time, he would say.
S: It was very clear that the matter of self acceptance is a result of the
way the child comes to understand himself in the process of develop
ing personality that we bring with us into adulthood.
C: We asked Phil to conserve time by focusing on his tension as a way
of going more quickly into trance.
S: There is a good deal of integrity you have when you learn to tie your
shoes. You might be interested in the double bow the first week, how
children have the ability to tie knots with manual dexterity. You think
that you can tie a square knot, you can tie a half-inch knot, you can
tie fishing line together so that it won't come apart under pressure
and uncles are interested in this. Neighborhood boys and girls are
frequently impressed with this ability. Tying a knot takes on entirely
different meaning to that girl later. You don't even need to think
about that because your unconscious learns about the appreciation
for your manual dexterity. Every child learns you have ten fingers
and it's just another way of focusing the growing child's awareness
on nimble fingers.
C: A child who becomes tired of doing things the same old way with his
ten fingers tries to figure out how he can accomplish that task even
using his toes. "Look Ma, no hands" is something every child enjoys.
The latter segm ent em phasized lessons of self-worth and self

acceptance, learned as a child learns to tie his shoes, tie a bow, tie
various knots, and increase manual dexterity. Frank had indicated
by his general preadolescent disposition that his psychological de
velopment was arrested. We would not attem pt to focus his
awareness on mature form s of self-acceptance that he may not
have mastered. We could, however, depend upon the fact that a
competitive child would have learned to com pete by attem pting
to outdo his peers; one possible area for the expression of com
petition is knot tying. We expanded upon this by m entioning the
168 CHAPTER 6
ability to impress others and even im press a parent with "lo ok ma

no hands." All these skills were framed as "Self-acceptance is a
result of the way the child com es to understand him self in the
process of developing personality that we bring with us into
adulthood."
We continued our appeal to Frank's rebellion with suggestions
containing paradoxical intention such as, "W e asked Phil to con
serve time by focusing on his tension as a way of going more
quickly into trance." Finally, the entire m etaphor was framed
around a them e calculated to appeal to Frank's rebellion toward
authority and apparent fear of growing up in a world of responsi
bility—a client who worked so hard and was so responsible that
he had an ulcer, a secret trouble. The true nature of the secret
trouble, which was not yet revealed to our client, was that the
trouble resulted from his inability to be dependent on others, to
be cared for, to recognize dependency needs, and to transact in
such a way so as to get his needs noticed and met. These are in
fact the same dynamic roots that seem to create anxiety for our
client, Frank.
S: Now you're in trance yourself. You probably wonder why your un
conscious doesn't know anything about moving your fingers up off
your lap.
C: And that's an interesting thought. You know that you know how to
lift your fingers up off your lap. You've known all along, even before
you knew that hand belonged to you. You knew how to raise it up to
your face and transport that rattle to your mouth.
S: But the fact that your hand fails to demonstrate that rapid response
doesn't mean it won't do it in its own good time as your hand is
lifting up toward your face even as I speak.
C: And it's important to do it in your own way. You knew that.
S: Your arm is resting slightly on your thigh, your wrist is demonstrat
ing that jerky movement because your unconscious moves isolated
muscle groups.
C: Minute muscle movements.
S: Elbows, and soon a finger will probably demonstrate the jerky move
ment because unconscious ideas of a finger will tend to stimulate a
movement in the muscle of that area cortex and you'll be pleased to
find out whether or not you're going to do that.
C: And on the same subject, we reminded Phil of that experience that
most midwestern children have had at summer camp or on vacation,
the idea of going on a canoe trip. You can have any fantasy you like
out on the freedom of that river.
S: You can feel the alteration in the tension of the index finger which
began to relax, and still lighter, that's right. And soon your thumb
will rest more lightly on the pants.
C: Nothing to remind you that you are in this time or this place but you
can be anywhere. Your own thoughts and imagining can be real.
And when two boys go on a canoe ride together, there is always the
question of who's going to sit in the front of the boat and who's
going to ride in the back. Now everybody knows that the person
sitting in the back has ultimate control about just which direction the
canoe will take. . .
S: But the conscious mind can notice the extreme slow speed of that
right hand raising off of your lap doesn't prevent your unconscious
mind from knowing that you learn to use your fingers in a variety of
ways as a child.
C: .. .how fast the canoe is going to slide through the water, and yet
only a fool would want to ride in the back of the canoe all the time
because the vantage point from the front of the canoe is so much
more conducive to your own imagination. The view is much better
and it's a pleasure to just relax in the front of the boat. . .
S: You may call it manual dexterity or you may call it good with your
hands but the conscious mind can be proud of the fact that you ac
cept yourself. You learn to stand in a certain posture to button your
shirts, or select which notch you use in your belt.
C: . . . and leave all of that steering and all of that effort up to someone
else. And that way when the canoe gets snagged in low branches
you can joke to the one in the back "you idiot, why did you steer us
into the branches." You can be proud knowing that you could have
done a better job but you can still enjoy allowing someone else to
enjoy making their own mistakes, giving you a free ride.
S: And how many people realize as grown men that when you fasten
your belt buckle your conscious mind feels proud of the notch in
which you are able to get your belt to buckle but your unconscious
mind has allowed you to have that feeling of your pride moving your
fingers.
C: And you have a lot of pride. Your conscious mind frets because you
notice your hand and wonder whether or not it's gotten off of your
leg or whether it ever will. Sooner or later it will.
S: Now another matter is your fingerprint.
C: And just what depth of trance have you reached and what's impor
tant to notice about that?
S: You couldn't look at your fingers and find out but you are as unique
as your fingerprint and you can't change your fingerprint.
C: Your fingerprint has a right to be here and you have a right to be
attached to it.
S: And so you should enjoy always being y ou .. . . And so Phil had no
idea about why we were speaking about those things in the context
of dealing with that stomach ulcer.
C: He knew he enjoyed it when we spoke about ball games and those
teams on which most boys have had a chance to play at some time or
another in their life.
170 CHAPTER 6
S: But he did know unconsciously that what was presupposed behind

those stories was related to that stomach ulcer because his stomach
became more relaxed when he listened to them.
C: We reminded him about playing in a game and being very proud
about the role you play in the game and we also mentioned the relief
a player can feel when he hears the coach delegate responsibility to
another player: "G o in for Phil." It's not a failure in any sense but a
pleasure to have that relief. It's all part of team spirit.
S: And his ability to relax there in the chair while the gang was out
improving their skill on the cou rt...
C: . . .being proud of their progress, proud of his progress.
S: It was because he had the kind of background that he had that nobody
needed to ask him to notice how he was relaxed in his legs. Nobody
needed to tell him your relaxation is in your arms and stomach.
C: He didn't need to learn that achievement is something that you can
have even when you're not doing anything.
S: They didn't need to say relax your neck and your back and your
eyes. He had a good coach and because he had a childhood experi
ence of that kind of relaxation being accepted you grow into adult
hood using those experiences. Now his stomach ulcer indicated that
he hadn't learned to have that response automatically in adulthood.
So we asked him to just consciously let your confusion of your dia
logue forget about the, have amnesia automatically...
C: Time passes...
S: Because it's a possibility that you'd like to use the experience of not
remembering...
C: . . . in interesting ways.
S: I didn't know whether or not he'd forget about what we said because
of the amnesia or fail to remember it because he could retrieve his
own ability at failing to not remember something that had been
forgotten...
C: Maybe he only failed to remember that he thought about what we
said.
S: .. .in such a way that his unconscious mind kept it and his conscious
mind listened to something else. Your hand is rising off your lap.
C: And that's an interesting thing to focus on. The pressure has defi
nitely lessened on your thigh and there's a little contact remaining
between the ring finger and the knee.
S: And you memorize the client's dissociation in the hand and arm ...
C: You don't really know as it clears the thigh whether it's going to float
higher or out to the side or halfway up.
S: . . . so that you could even watch a movie while you're feeling disso
ciated. Your dissociation experience sitting there watching the movie. . .
C: .. .listening to one thing...
S: . . .you can hardly care what you're seeing on the movie screen...
C: . . .having your own thoughts...
S: . . .is a common experience that he had had before, up toward your
fa ce ...
C: . . . much like the feeling that your hand experiences whether it's rais
ing up or pressing down, it's doing something separate and apart
from what you're doing.
S: It's nice to know when a conscious mind is confused...
C: . . . and up and down. . .
S: .. .you can do a lot of things that you didn't consciously know. You
had a right hand moving up to your face. Again soon, the confusion
of the conscious mind.
Begin Direct Work Metaphor While

Reinforcing Dissociation
O ur goal at this m ost intense phase of the multiple em bedded
m etaphor was to teach Frank a new behavioral role which would
allow him to deal more appropriately with his peers, boss, wife,
and eventually with his children. Specifically, he needed to un
derstand and accept his dependency needs, sense of vulnerability,
ability to trust, and the behavioral transactions involved in the
m anifestation of these needs and abilities. To solidify the role of a
m an his age who possesses and is responding appropriately to
the resource feelings ju st retrieved, we used a story that provided
a context for an imaginary encounter betw een a boy and his
father.
Frank's personal encounter was stimulated by listening to the
metaphorical one betw een protagonist and father. It was an op
portunity for us to help him build more suitable associations to
his tender, dependent feelings, by altering what he could expect
form proper parenting by a father figure. Frank can benefit from
this experience in two ways. He can "receiv e" some of the accept
ance that was never offered in real transactions with his father
and thereby be "rep aren ted " to som e extent. And he can incor
porate the behaviors of proper parenting, thereby supporting his
own changes with internal dialogue and having these responses
available w hen the tim e com es for him to literally interact with his
own children. It is im portant to rem em ber that Frank is likely to
resist, com pete with, and rebel against our rules, but his com peti
tiveness and need to take a challenge can really be considered pre
dictable therapeutic assets so long as we rely upon the use of par
adoxical intention and paradoxical logic to utilize the reversal re
sponses that can be expected from him at every turn.
172 CHAPTER 6
C: When Bill approached us in Australia at a treatment marathon he

told us that he was a sheep farmer.
S: You could think about your hand. . .
C: He struck us as a very naive therapy client. He didn't know the first
thing about really having his feelings.
S: . . . amazed moving up to your face, why your unconscious. . .
C: He didn't even know that a man has feelings.
S: . . . mind takes that feeling of dissociation you have in your arm. . .
C: He didn't know how he'd have an arm raise up toward his face. He
didn't even know it was possible to go into trance.
S: He complained that there were two wc.lds, the work world and the
world at home and his wife didn't like his inability to separate them
and he asked for therapy about that.
C: He wanted to make his wife happy and he wanted to be happy at
work.
S: He had missed out on his entire childhood. His father hadn't spent
any time with him at all.
Once again resistance was utilized by paradoxically stating the
opposite of what was intended. We stated that the Australian cli
ent was naive and we expected that Frank would condescendingly
assume him self superior to the Australian. Actually, it was possi
ble for Frank to be naive and still realize he could gain from the
therapy, or to rebel and consider him self smarter (and thereby be
expected to gain even more). Further, if Frank disidentified with
the type of person in the story, he would create a situation in
which he could be less defensive w hen the painful relationship
that character had with his father was m entioned. So again, we
have seemed to fail to be correct in order to be even more effective
with Frank's therapy.
Next, dissociation is suggested. The following segm ent helps
Frank develop the ability to view him self at a dissociated distance.
We expected that the intensity of the dependency and tenderness
feelings we would be suggesting to him might be overwhelmingly
threatening, unless we "d ilu ted " them with dissociation and
splitting of affect. We considered examination of the feelings,
even with dissociated splitting, to be better than his characteristic
repression and suppression.
S: You get the feeling of dissociation developed some place in the body,
convince the client at the conscious level it doesn't matter if you're
paying attention to that or not.
C: But in the same way you have that dissociation in the body, make a
transition, to psychological dissociation.
S: And we asked him to realize that he could see a movie with an itch
and he wouldn't feel an itch in his body.
C: And on the other hand, if you're not interested in that movie, you
don't even notice what's going on there because you're having your
own thoughts and they are so much more interesting.
S: Somebody could be crying in the movie and you could be laughing. I
know it's the context that makes a difference. When I saw Night of the
Living Dead at the drive-in, I left twice. It was too scary for me to see
then, but when I saw it in graduate school we all laughed at the
monsters and I could deal with that.
C: Meanwhile, the people on the screen were still very frightened. It
was no laughing matter to them.
S: Likewise, you could watch somebody having a love scene on televi
sion and you're not interested in that.
C: So you can have your own reactions no matter what the characters
on the screen have or know.
Describing laughter at a movie known to be scary would be still

another illogical conclusion reasoned paradoxically were it not for
the manner in which the change of context was revealed. By al
lowing a client to understand the leap to the next logical level, a
possible paradoxical statem ent is made logical. Viewing the movie
in a room full of sarcastic college students changed the impact of
the images. So too, being in a room filled with professionals
viewing him in trance will alter the im portance of his feelings; in
this situation, his feelings will likely becom e more intense. We
shared with Frank the secret of frame change in the film, but ob
fuscated the logical level of frame change in the actual relation
ship with us. The metaphor allowed a natural context for the par
adoxical bind to enhance therapeutic success.
S: You don't realize your own self sitting here in the trance listening to
our stories with your eyes closed. You don't know that a lot of peo
ple have their hands up in the air. But if you had your eyes open and
saw, you'd still have your hand down on your thigh, because you
don't need to have a reaction to things you see.
C: The intensity of your response is enough for you to know that you
can have that dissociation. Your conscious mind can be interested in
how your unconscious mind creates it.
S: You have reactions to your own unconscious processes that develop a
dissociation so you could watch yourself winning the award for the
best-stuffed fish.
C: Watch yourself hitting that golfball the farthest.
S: Watch yourself getting a paper in school.
174 CHAPTER 6
In the segm ent above, the stuffed fish was m entioned as a ludi
crous image to appeal to Frank's sense of "m aking fun o f" or be
ing "o n e u p " on others. How easy to get his conscious mind
engaged by jerking him to an image as laughable as a stuffed fish!
In the segm ent to follow, we continued to appeal to this need to
be "o n e u p " by expanding that even a sheep farm er could picture
himself. We could have further insured that he would picture
him self by using paradoxical intention as well, by saying som e
thing like: "T h ere is an interesting way in which you will fail to
make a clear image of yourself." However, we expected we could
satisfactorily elicit his visualization at this point by relying on his
need to be superior.
S: You could imagine yourself sleeping, riding a bicycle, speaking with
your parents, and all the while your unconscious allows you to have
your o w n ... Sitting here in the chair, you're relaxed, shoulders
down, hand on your thigh for n ow ...
C: .. .face muscles smoothed out, your eyes are closed. There's no need
to do anything. But we asked Bill to do something very complex,
despite being naive and despite not having a father that he could
respect.
S: See himself with his father but even though he had had no father
who had developed good feelings in him as a child, he could picture
himself interacting with what he would imagine to be a father.
C: He was quite able, even a sheep farmer could picture himself and
picture that father that we described.
S: We reminded him that as a grown man we didn't need to remind him
what it would be like had his father said "I'm proud of you." We
didn't have to tell him what it was like to have a father say, "You've
worked very hard. You deserve a rest."
C: And he didn't know how, but he responded in a way that we didn't
have to explain when we told him that a father could say, "That's
alright son, that's an interesting mistake. What can you learn from
that mistake?"
In this segm ent, the interaction with a father constitutes a
learning for Frank but in a context in which we denied that we
were in any way educating him (or the protagonist). This was ac
com plished with statem ents that denied an aspect of responsibil
ity of the com m unication (Haley, 1963, p. 31), such as: "W e didn't
need to remind him ," "W e didn't have to tell him ," and "W e
didn't have to explain." In this manner, Frank could take credit
for knowing the parenting information (which in som e way, he
m ost certainly did) and for applying it to him self, without rebel
ling against our authoritative knowledge. Finally, we made the
decisive point that framed m istakes as opportunities for learning
and improvement rather than proof of inadequacy—an existential

polarity to the point repeatedly taught in Frank's competitive rela
tionship with his own father. Likewise, the possibility that a m is
take is a chance to learn and change supports (but doesn 't fully
explain) the logic of the original paradoxical m etaphor: You can
lose in such a way as to win. The learning will bridge the gap
betw een the emotional experience in this present stage of trance
and the eventual conclusion that will be brought to that original—
now suspended and postponed—metaphor.
S: Your conscious mind is doing quite well thinking about your own
responses, no reason to pay attention to our words consciously
when your unconscious can pay attention and learn something. We
didn't need to tell Bill you can see an embrace. We did have to re
mind him, because he wouldn't have thought of it on his own, to
picture his father crying.
C: He seemed to go into a dream in order to picture that.
S: But it was a new thought for his unconscious mind to realize th a t...
C: So we directed the dream, suggesting that he watch the softness on
your father's face; see that tear roll out of his left eye or was it his
right eye first. Notice that he doesn't seem to be ashamed of that at
all.
S: We were tricking him. We told Bill to first imagine seeing his father
place his hand on his son's hand and say "You know how hard it is
for me to tell you I love you."
W ith the preceding segm ent, we continued to use denials to
educate Frank. We com m ented about the father, who is of course
sought after by both the protagonist and Frank in real life. The
father was described by us as dem onstrating softness, crying, ten
derness, and so on to help Frank experience association to a new
male role. We hoped Frank would accept those parts of him self,
and accept them as part of the role he would com e to play in the
not too distant future with his own children. Perhaps the m ost
im portant aspect of this segm ent is the continual dissociation of
conscious and unconscious functioning to support am nesia, and
the conscious confusion about this reassociation of experience.
W ith the assistance of such trance phenom ena, Frank would be
less likely to interfere with his changes by criticism, analysis, and
his usual condem nation. (It was, by the way, revealed upon re
orientation that he did have am nesia for this portion of his ther
apy experience.)
S: And then suggest that he experience what it'd be like for an Austra
lian sheep farmer to caress his son and say "I do love you."
C: Now, then, so far back in the past. [Client coughs.] And it wasn't
176 CHAPTER 6
difficult at all the way that Bill was able to imagine it, though it had
never happened he had the experience [Client coughs again.] and it
choked him up. [Client's stomach muscles visibly shake.] He visibly
shook. A tear came out of his eye.
S: And his conscious mind didn't even know what we were talking
about, wasn't even interested, because Bill's ability to develop a dis
sociation let his unconscious learn a great deal or [Client coughs again
and raises hand to cover mouth—noticeably becoming increasingly uncom
fortable.] maybe changed his mind. And that's not the only way you
know you can raise your hand to your face.
At this point, we had obtained the parasym pathetic clues that in
dicated success in getting Frank's response for tenderness, risk-
taking feelings, and dependency.
To briefly restate our goal regarding these feelings, we saw
Frank's authority and rebellion problem being rooted in an inabil
ity to accept his dependency needs. Since he did not feel com fort
able with his dependency, Frank had not learned to trust. Instead
of having learned to ask for help, he had learned to com pete. He
had not learned to take advice about his feelings of weakness, but
rather to project those feelings and condem n others. Therefore,
we regarded Frank's coming to terms with his dependency needs
as the foundation for change in the area he presented to us and
contracted with us to change.
Link Resources to Immediate Social Situation

In the next segm ents, the client's attention is abruptly directed
back to the previous metaphor storyline, without associative links
to the story ju st com pleted. Switching storylines rapidly and
without reference can be expected to produce am nesia or similar
trance phenom ena in most subjects. Always, it is necessary to ap
peal to the client's unique personality and in this case Frank's re
bellious, contrary nature was addressed in such a way as to in
clude an outlet for his pseudoanger, with phrases such as "dam n
b u s" and "d am n Lanktons" in the linking segm ents that follow.
The general goal at this stage of the multiple m etaphors is to
link the resources that were retrieved and organized to family
roles and relevant situations in the social network. This linking is
accomplished with imagery that we expect will facilitate Frank's
having several thoughts about how to apply the new resources in
his immediate social situation after therapy. Here, Frank's self-
image thinking is fostered through visual fantasy rehearsal in the
context of com pleting the suspended metaphor about the ulcer
client.
S: When Phil left our office he was confused.

C: Besides that he was late. He was late for the bus he intended to catch
for home.
S: He didn't know that he was catching the wrong bus.
C: He ran for it, ran and ran.
S: He later reported that he ran two blocks.
C: He saw the bus about to pull off as the doors began to close.
S: He banged his fist on the bus, "damn bus." The bus driver stopped.
C: The bus had already rolled a foot but did stop.
S: He hoped the bus driver hadn't heard him say "damn bus." He got
on the bus and fell onto the leather seat.
C: You can imagine how relaxed he really became as he sank into that
seat with the relief—knowing that he had caught the bus and could
leave the driving to someone else. He was exhausted from running
and his heart was pounding.
S: And in the hours where the heart is weak and memory is strong and
it just seems that time stands still. . .
C: And when the bus pulls off jiggling along. . .
S: .. .his thoughts went to various things, inluding our posthypnotic
suggestion he'd never intended to follow.
C: And what an excellent context to have your own thoughts, jiggling
along on a bus.
S: We had been certain he would catch the bus that would go through
the tunnel.
C: At first you look out the window, things are happening out there that
have no special interest to you and you just watch life passing by
effortlessly.
S: When the lighted bus goes into a tunnel, you see reflections in the
window. And we knew that he was going to be staring blankly into
the window in trance because that was the posthypnotic suggestion.
In the preceding segm ent, Frank's rebellion toward authority

was addressed with the protagonist's condem ning and remorseful
reaction to the bus driver. It was concluded with an ambiguous
m ention that "h is heart was pounding," which was true in both
the protagonist's experience and for Frank, who had becom e anx
ious in response to the previous metaphor. In the context of m en
tioning anxiety, associations to feeling relieved and relaxed were
suggested. W hen Frank appeared to be relaxing again, we re
turned to the paradoxical twist of the m etaphor: "H is thoughts
went to various things including our posthypnotic suggestions
he'd never intended to follow." We could be certain that Frank would
recall the discussion of posthypnotic suggestions to follow. He
would also expect him self to be clever enough not to follow them
178 CHAPTER 6
since he typically censors, m onitors, and com petes with sugges

tions from others. Thus w hen he sooner or later com es to think
about these suggestions, he will have followed them , since we
stated that his thoughts went to the posthypnotic suggestions "h e
never intended to follow." The use of past perfect tense allows the
possibility that he folllowed or did not follow the posthypnotic
acts; either way they corectly highlight his intention. Therefore,
he must come to admit that we predicted his thinking and his
intention. Thus we gain more credibility and increase the likeli
hood that Frank will expect to follow the posthypnotic acts.
C: And so when the bus did, in fact, enter the tunnel, his own reflec
tion was so striking that he couldn't help but notice it.
S: In the trance we had explained that you're going to see your hair is
combed nicely, you have relaxed lips, your cheek muscles are flat. . .
C: .. .you're breathing quite comfortably...
S: Your face is red, your posture is erect, shoulders are square and your
head is somewhat tilted.
C: You have every opportunity to be relaxed on that bus, no responsibil
ity really. For Phil that was quite unusual not to have responsibility.
S: He tried to move and he realized that he was fixated on that picture
of himself breathing relaxed.
In the above segm ent, we began describing Frank's actual facial
appearance. The switch in pronouns to your created a convenient
ambiguity. It was likely that Frank would begin to wonder what
the posthypnotic thoughts m ight be: O nly by knowing our sug
gestions could he be sure to avoid them .
C: And he thought, those damn Lanktons, they've got me looking at
myself in a way that I don't know how to look at myself.
S: They wanted me to pay attention to my face but that was only half of
the story he consciously understood.
C: He thought he had the whole thing figured out.
S: But the unconscious mind was learning something entirely different
he didn't even understand.
C: But despite that thought his conscious mind had over and over his
unconscious, as if running a projector, flashed on the screen of that
window a scene from his life.
S: Every time he passed one of those little booths that separate the tun
nel, the scene would change.
C: And in those scenes the background would change to include those
situations that he'd been in before and he knew he would encounter
again, such as those with his wife—a most difficult situation and yet,
he was interacting with her in a way he really could be pleased
about. He could see her appreciation of those changes, her apprecia
tion of his softness, her appreciation of his vulnerability.
S: Breathing relaxed, looking her in the eyes. . .

C: And he watched that reflection and was just fascinated by it, fixated
on it.
S: And he had no intention to change that picture.
In the segm ent above, various scenarios were painted in which

Frank's self-image interacted in new ways, using the resources
just retrieved in the previous m etaphors. Now, since Frank was
given som ething of a challenge, he probably thought consciously
that he knew what was m eant w hen we said such things as "th a t
was only half of the story," and " h e didn't even understand." In
fact, this was another purposeful way for us to seem ingly fail at
distracting him (from the posthypnotic aspects of the suggestion)
while actually succeeding, at another logical level (as Bateson
would say), to distract him from the association of the previously
revitalized feelings of risk taking, dependency, tenderness, and
pride. It is easy to see how this paradoxical approach distracts the
client's conscious mind in this example.
In the following segm ent, the same approach is expanded but
the visual scenarios include other interactions. Each time care is
taken, by m onitoring his physical dem eanor and nonverbal be
havior, to associate his fantasies to his novel feelings. He thus
learned to think of him self with the addition of new feelings and
new, available, and organized behavioral options.
C: But then another bump in the road and he'd shake his head and try
to shake the image thinking again, those damn Lanktons, what do
they have me doing and why?
S: They think I'm going to stare at that image of myself in the window
and see myself.
C: I know what they're up to but it's not going to work.
S: And he was only half right because his conscious mind was ques
tioning the suggestions but his unconscious was still learning some
thing else.
C: And despite that conscious thought once again another background
emerged and there he was with his boss. . .
S: Chest was relaxed, stomach was relaxed, breathing was quite
relaxed ...
C: . . . the chief psychiatrist and he was being criticized constructively
for failure to succeed completely with a patient.
S: And he was smiling incongruently because he was smiling at the
thought that the Lanktons have got me picturing myself relaxed here
on this bus and I can't move a muscle but the face was quite relaxed.
C: And then he found to his surprise that in the image of himself in
teracting with the boss, he was smiling there while receiving criti-
180 CHAPTER 6
cism and interacting in a very appropriate way that he could be

pleased about, curious about.
S: He knew that the sounds on the bus had very little to do with the
voice he was hearing in the back of his head that was saying "I un
derstand." Or was it him hearing his own voice? And his conscious
mind didn't understand at all.
C: And another jiggle of the bus, another shake of his head, another
image.
S: He found out that in adulthood you use those childhood learnings
that have stayed with you. . .
C: It didn't matter how irritated he became thinking about his doing
something that was actually instigated by us because he was still
learning something even though he only knew the half of what it
was that he was learning.
S: . . . and your conscious mind needn't pay any attention to them but
you can appreciate your own ability to use that and there he was
seeing himself appreciating himself while he was watching himself
interact with that fifth baby that had created that tension all along.
C: And so when the bus came out of the tunnel it was a shocking
change. He realized at that moment the bus he was supposed to be
on didn't even go through a tunnel. It was at that moment he real
ized he had been on the wrong bus.
S: Suddenly his attention was disrupted and “those damn Lanktons"
was the thought that didn't even get finished because he had amne
sia for what he thought he had not remembered about before and
was certain of it at this time.
C: And everyone knows that certain surge of panic that comes when
you've realized you've made a terrible mistake that's going to cause a
lot of inconvenience. But just as that familiar tension began to rise
and register in his stomach he thought to himself “I don't give a
damn if I did catch the wrong bus. I enjoyed what happened here
even if those Lanktons were behind it.
S: And that was the end of the growing tension that had created the
ulcer.
C: And he knew he had learned something even though he didn't know
what he had learned, he still had that relaxation.
End Matching Metaphor

In the following segm ent, we finally return to the original meta
phor and bring closure to that long suspended paradox about fail
ing in such a way as to succeed.
S: And that's the only thing that we really had to share with him. So
when the first baby was bom it finally dawned on him and he didn't
let us know until he was 40 years old the meaning of that phrase he
had heard in therapy, "Only a damn fool would fail in such a way
that he fails. It takes a wise person to fail in such a way that he
succeeds." We had explained in great detail that it's an absolute rule of
social contract, that you define your behavior and when you operate
outside of that. . .
C: .. .guilt is an automatic result...
S: . . . definition of yourself you feel guilt.
C: .. .and you can't help but have guilt. It's a rule.
S: So we explained in great detail that you operate within your self
system and have the feelings you think you're entitled to and the
expression of guilt is merely the understanding that when you oper
ate out of your defined and accepted roles you feel an unusual feel
ing called guilt and it's simply the rule. We explained very carefully
and it was the only thing he remembered when he left the office until
that child was bom and then it dawned on h im ...
C: He'd had his own way of doing things. He had managed to break the
rules by following the rules without having any guilt. It wasn't nec
essary in his case.
S: He wasn't having any affairs, but he had broken the rules and he
was free of the guilt and that was a very wise way to prove that any
fool can fail by failing but it takes a wise man to succeed by failing.
C: And it was his own way of failing in such a way as to succeed.
Since Frank's proclivity for breaking rules was well known, we

reminded him that the "absolute rule of social contract" is to feel
guilty for having affairs. We purposely shifted em phasis away
from the actual (unspoken) rule that one does not have affairs,
and instead stressed that part of the rule that concerns guilt. This
provided Frank an opportunity to break a rule and have neither
guilt nor affairs.
Once again, we want to state that Frank was not know n to have
had affairs, but the value of m etaphor is that it is a symbolic re
flection into which the client projects actual and literal m eaning. If
Frank thinks the m etaphor refers to real affairs and that we have
correctly guessed, that is fine. If, on the contrary, he thinks "In
my case this is like rebelling against authority," that too is fine. In
the latter situation, the meaning becom es som ething like " I'm
supposed to feel foolish that I have broken rules and rebelled so I
w on 't." The presupposition is that he will break the rule and
avoid unpleasant feelings, but he will not continue to rebel. How
ever, this is probably m uch more apparent to the reader than it
was to Frank, who remained in trance listening through the filter
of his rebellious personality.
182 CHAPTER 6
In the final segm ent, the client was also referred to his next
logical stage of family or social development. In this case, we
m entioned children, stimulating Frank to think about m aking this
likely change in his family structure. (He had indicated to us in
the interview that he and his wife did hope to have children at
som e point in the future.) We did so, keeping in mind the impor
tance for Frank of paradoxical intention. By suggesting that the
protagonist didn't com e to realize the point until the birth of his
first child, we encourage Frank to take exception and conclude
som ething like: " I can figure it out now. I d o n 't have to wait until
m y first child."
In other words, Frank will attem pt to com pete with the protag
onist in term s of achieving the understanding more quickly and in
this story more quickly is m easured against the time necessary to
have a child. Thus Frank will be less inclined to challenge the act
of having a child and will begin to consider the possibility o f fa
thering children and com fortably expanding the roles that he can
imagine him self filling. These few references to fathering are only
a beginning. But this beginning is a very tidy addition to the ther
apy. It is the generative change, the future change, that will sup
port the larger changes Frank has learned to make.
Reorienting the Client

S: Now you don't need to wait as long as he did in order to express a
learning. Thursday night is a very good n ig h t...
C: . . .as is Tuesday but you might prefer to do it on Wednesday.
S: We had a client we really wanted to change on Wednesday and he
thought we wanted him to change on Wednesday so he'd show
u s . ..
C: . . .he did it on Monday.
S: And while we've been talking you can realize that your hand raised
up off of your lap to let you know that your unconscious really wants
to demonstrate your ability to y o u . ..
C: Change is constantly occurring in your body, your thoughts, your
conscious thoughts going one way and your unconscious thoughts
going another way.
S: . . .and your hand is rising up off your lap again now.
C: You don't even know how your unconscious responded.
S: I don't know if it will get off of your lap by the time you come out of
trance.
C: You may know about the images your unconscious mind allowed
your conscious mind to notice but you don't know what rate you're
going to come out of trance. . .
S: But I do know that you can come out of trance suddenly or gradually
as I count by increments, 1, 2, 3 . . .
C: . . .or whether you'll be completely out just because you managed to
open your eyes or maybe you'll be out of trance even before you
open your eyes.
S: Four , a little more, 7, 8.
C: Ten, 12.
S: You might want to have yourself remove from trance slowly enough
to let your arm come up again and demonstrate your ability to use an
unconscious learning outside your own awareness even if it's not
spoken about again.
C: Fourteen, 16, and just roll those ideas around and allow them to set
tle comfortably in a fabric of your own design, your own choosing.
S: Seventeen, 18.
C: Nineteen and 20.
In this standard reorientation by counting, we have again incor

porated the paradox and counterintention. The first example oc
curs in the debate over which day to change. The fact that change
will happen is again presupposed. Given the structure of the
story that has preceded this reorientation, Frank can either be like
the protagonist and succeed in life but break one rule and not feel
rem orse for his past m istakes, or rebel and not be like the protag
onist. In the latter case, he will feel remorse and be following the
rules.
We assumed that, in keeping with his personality style, Frank
would decide to com pete and outdo all possible rivals. In that case
h e would be changing. O ur only question was how to be in charge
o f the change. The answer was sim ple: Give him a time and expect
him to rebel against it. We m apped this out for him in the discus
sion of change happening on Tuesday or Wednesday. In the m eta
phor, the client “ showed u s" and changed on Monday. So we pro
vided what seem ed a w ay for the client to beat us. But paradox
ically the client only apparently beat us, since from a larger frame of
reference (a higher logical level) we have still defined that the client
change. In order to distract Frank's conscious mind from examin
ing this higher level and possibly reconciling the paradoxical bind,
a final paradoxical intention was offered with the challenge, "B u t
you d on't know w hat rate you're going to com e out of trance."
O nce again the client can seem ingly defeat us or rebel against our
suggestion at the level of apparent content; in so doing he m ust
follow our suggestions at the higher logical level.
Finally, Frank was reoriented with the closing remark at the con
tent level that h e allow the thoughts to settle into " a fabric of your
184 CHAPTER 6
own design, your own choosing." Again he was given a paradoxical

symptom prescription. Since he was bound to modify the thoughts
from his personal need to com pete, improve, find exception, con
tradict, or criticize, we simply made it explicit that he was to do just
that. Frank must frame the entire session in the paradoxical bind
that he either accept our suggestions and modify the thoughts that
were stimulated or reject our suggestion that he modify his
thoughts, in which case he would accept the thoughts that were
stimulated, and those of course were alco our suggestions!
After reorienting from trance, Frank reported that fifteen m in
utes had passed; in actuality an hour had elapsed. This severe
time distortion is a rough indicator of his depth of trance. He also
stated that he had in the past taken many psychedelic drugs such
as LSD and STP. He stated that, to his astonishm ent, during
(what he assumed m ust have been) the m idportion of the trance,
he had felt more disoriented than he had ever felt on the drugs.
He took this as evidence that som ething powerful had happened
to him. We considered the middle segm ent to be the portion in
which he was experiencing his tender feelings and imagining a
close relationship with a father. His degree of fear, which m ani
fested itself in the trance as coughing and severe jerking of the
abdominal muscles, indicated the strength of the treatm ent ap
proach. It had served as a vehicle in which he effectivly con
fronted some of his dependency needs, tender feelings, and
vulnerability. He had even been able, upon em erging from the
trance, to explain to us that he had experienced a unique feeling
state and that he was scared by it—but he had lived through it to
brag about it. We took this to mean that the trance had become
significant to him in the way his drug usage had been important.
Identifying with it was a vehicle for discovering his own auton
omy. He would use it to change.
Finally, Frank's am nesia about the events of the trance and his
loss of content, evidenced by the severe time distortion, stood as
a testimony to the value of using paradoxical Ericksonian inter
ventions—including symptom prescription, binds, and paradoxi
cal m etaphors—to distract, overload, confuse, bypass, and utilize
Frank's conscious mind and rebellious personality traits.
Although we do not intend to imply that effective treatment
can always be accomplished with a single session, one session
was all that was necessary with this client. In a follow-up five
m onths after this session, Frank reported continued am nesia
about the actual content of his trance. All he could surmise was
that it involved " a series of things that related around a boy and
his father." He also reported that shortly after the session he grew
a beard, som ething he had hesitated to do for some time. The
result of this symbolic action was that Frank changed his self-im
age in a way apparent to him each time he looked into a mirror.
He stated that he "feels older and much more sexually m ature"
which seemed to us to be indication of changes in Frank's self-
image and age appropriate intimacy. He summarized that others
now seemed to perceive and respond to him as a man rather than
a boy. Though he tended to attribute this change to the presence
of his beard, we concluded that he actually evoked such differ
ences in response by the use of different behaviors that were
shaped with metaphoric detailing.
The act of growing a beard probably would not by itself have
resulted in the feelings Frank reported; these were more likely the
result of the metaphoric guidance he received in areas of affect.
The beard did, as we m entioned, illustrate Frank's increased at
tention to the social characteristics of his self-image in a way that
is consistent with the self-image thinking metaphor. And as a
symbolic representation of trance learnings, his beard seem ed to
provide a daily stimulus for Frank to associate to all of those
learnings and act in ways consistent and appropriate for someone
who is "older," "m ore sexually m ature" and no longer " a b o y "
even though he did so without conscious insight or awareness of
this unconscious aspect of association.
We attributed Frank's am nesia and lack of resistance to change
to the use of sym ptom prescription and the paradoxically confus
ing structure of his trance. More importantly, we observed evi
dence in his self-report of the impact created by the coordinated,
sim ultaneous, and concurrent use of thoughtfuly planned meta
phors which indirectly stim ulated a controlled elaboration of his
experiential resources.
Since our goal was to illustrate the planning and therapy with
lesser used form s of paradoxical intervention, we used a tran
script from a hypnosis session. We expect that readers will be able
to recognize how many of these interventions can be applied in
their own work even if that work is not done in hypnotic trance.
We have offered an assessm ent and treatm ent planning format as
a framework that will translate directly to the work of marital,
family, group, or gestalt therapists who do not use hypnotherapy
per se in treatm ent. In so doing, we hope to stimulate an in
creased understanding of how a variety of Ericksonian styles of
paradoxical intervention can be beneficially em ployed—even in
non-Ericksonian treatm ent approaches.
186 CHAPTER 6
REFERENCES
Aristotle. Rhetoric and poetics of Aristotle. New York: Modern Library,
1954. (Originally published, 1459.)
Bateson, G. Steps to ecology of mind. New York: Random House, 1972.
Erickson, M. H. Personal communication. August 7, 1977.
Erickson, M. H. The application of hypnosis to psychiatry. In E. Rossi
(Ed.), The collected papers of Milton H. Erickson on hypnosis (Vol. 4).
New York: Irvington, 1980.
Erickson, M. H., & Rossi, E. L. Hypnotherapy: An exploratory casebook.
New York: Irvington, 1979.
Erickson, M. H., & Rossi, E. L. Experiencing hypnosis: Therapeutic
approaches to altered states. New York: Irvington, 1981.
Fisch, R. Personal communication. December 2, 1982.
Jung, C. The relations between the ego and the unconscious. In V.
DeLaszlo (Ed.), The basic writings of C. G. Jung, New York: Modem
Library, 1959.
Lankton, S., & Lankton, C. The answer within: A clinical framework of
Ericksonian hypnotherapy. New York: Brunner/Mazel, 1983.
Lankton, S., & Lankton, C. Multiple embedded metaphor and generative
change (Audio tape # M323-37C). Phoenix: Milton H. Erickson
Foundation, 1982.
Lao Tsu. The tao te ching; The writings of Chuang-Tzu; The thai-shang.
(J. Legge, Ed. and trans.). Taipei: Ch'Eng-Wen, 1976.
Lustig, H. S. The artistry of Milton H. Erickson, M.D. (a videotape).
Haverford, PA: Herbert S. Lustig, M.D., Ltd., 1975.
Peris, F., Hefferline, R., & Goodman, P. Gestalt therapy: Excitement and
growth in the human personality. New York: Dell, 1951.
Weakland, J., Fisch, R., Watzlawick, P., & Bodin, A. Brief therapy:
Webster, M. Webster's third new international dictionary of the English
language unabridged. Chicago: Encyclopedia Britannica, 1976.
Yalom, I. The theory and practice of group psychotherapy. New York: Basic
Books, 1970.
7
Paradox in Context*
by Howard Tennen, Ph.D.
Joseph B. Eron, Psy.D.
Michael Rohrbaugh, Ph.D.
Paradoxical interventions such as prescribing the sym ptom and

encouraging resistance are in seem ing opposition to the thera
peutic goals they are designed to achieve. These intriguing tech
niques are receiving increased attention in the psychotherapy lit
erature, as evidenced by the following examples:
1. A patient com plaining of intense anxiety in public places is
asked to have a panic attack deliberately as the first step in
bringing the sym ptom under control (Weakland et al., 1974,
pp. 141-168).
2. After reviewing a voluminous history of unsuccessful medical
and psychological therapies, a psychiatrist tells a headache
patient that her condition is probably irreversible and that
therapy should concentrate on helping her live with the prob
lem . Despite persistent pessim ism by the doctor, the head
aches improve (Watzlawick et al., 1967, pp. 246-247).
3. A depressed stroke victim, who would attend only an initial
family interview, improves dramatically over six m eetings in
*The authors wish to thank Laurie Pearlman, who reviewed earlier versions of
the manuscript.
187
188 CHAPTER 7
which a therapy team coaches his spouse and grown children

to be ineffectual and helpless in his presence (Watzlawick &
Coyne, 1980, pp. 13-18).
4. A therapist asks a symptomatic husband to pretend to be irre
sponsible and inadequate three times before the next session
and instructs his wife to try to find out w hether he is really
feeling that way (M adanes, 1980, pp. 73-85).
5. A therapist asks an overinvolved grandmother to take full re
sponsibility for a misbehaving child. Grandm other backs off
to a supporting role, allowing m other to take charge, and the
child begins to behave more appropriately (Haley, 1976, p.
132).
6. A psychiatrist tells a well-educated, depressed, young hom e
maker in the presence of her preoccupied, career-oriented
husband that her true nature is to find happiness in serving
others. She therefore should relieve her husband of all re
sponsibilities at hom e so he can work undisturbed in his
study. The husband becom es more involved in the household
and the wife in activities outside the hom e, saying she is no
longer depressed (Hoffman, 1981, p. 307).
7. A team of therapists gives a family a letter to read aloud at
prescribed times before the next session. The letter praises the
identified patient for acting crazy to protect his father, ex
plaining that by occupying m other's time with fights and tan
trums, he allows father more time for work and relaxation
(Palazzoli, Boscolo, Cecchin, & Prata, 1978, pp. 127-128).
8. A team com pliments a severely obsessional young woman
and her parents for protecting each other from the sadness
associated with the death of a family m ember several years
earlier. The team prescribes that the family m eet each night to
discuss their loss, and instructs the patient to behave sym pto
matically whenever her parents appear distraught (Hoffman
et al., 1981).
Presented in this way, paradoxical interventions appear to be

simply new tricks or techniques which can be added to a clini
cian's armamentarium. We will argue, on the contrary, that these
m ethods are most effectively (and responsibly) used, not as iso
lated techniques, but within the framework of a coherent theory
of problem maintenance and problem resolution. The most useful
framework is one in which paradoxical m ethods have been used
m ost extensively: It com bines a systemic or cybernetic under
standing of clinical problems with a strategic orientation to
change. Though paradoxical interventions can also be used and
PARADOX IN CONTEXT 189
understood in other (e.g., psychoanalytic, behavioral, existential)

frameworks, we will review them here in the context of a strategic
approach (Hoffman, 1981; Rohrbaugh & Eron, 1982). As back
ground, we will examine briefly other perspectives on paradox.
FROM PARADOXICAL INTENTION

TO INTENTIONAL PARADOX
The use of paradox in resolving hum an problem s is not new.
References to it appear as early as the 18th century (Foucault,
1973). More recently, Dunlap (1930) applied the technique of
"negative practice" to problems such as stammering and enure
sis. Frankl (1960) used "paradoxical intention" to alter the m ean
ing of symptoms for his patients, and Rosen's (1953) "direct psy
choanalysis" stressed the benefits of encouraging psychotic pa
tients to engage in symptomatic behaviors, particularly w hen
relapse is an issue.
The most significant contributions to the literature on paradox
come from Gregory Bateson's 1952-62 research project on com
munication (Bateson et al., 1956; Haley, 1976). Bateson, Jackson,
Haley, Weakland, and others explored the role of "dou ble bind"
com m unications in resolving as well as creating problem s. Watzla
wick, Beavin, and Jackson (1967) summarize the ingredients of a
therapeutic double bind: W ithin the context of an intense relation
ship, the psychotherapeutic situation provides that
an injunction is given which is so structured that it (a) reinforces
the behavior the patient expects to be changed, (b) implies that this
reinforcement is the vehicle of change, and (c) thereby creates a
paradox because the patient is told to change by remaining un
changed. (p. 241)
Haley, Weakland, Watzlawick, and colleagues, com bining ideas
from com m unications and system s theory with clinical methods
inspired by m aster-hypnotist M ilton Erickson (Haley, 1973) went
on in the late 1960s and early 1970s to develop therapy models
with paradox as a central feature.
Historically, applications of therapeutic paradox have been as
sociated with particular theoretical frameworks or paradigms.
Negative practice was understood in terms of learning theory,
Frankl's paradoxical intention in terms of existentialism, direct
psychoanalysis in term s of psychoanalytic psychotherapy, and
therapeutic double binds in terms of cybernetics and systems the
ory. In recent applications, some of these m ethods have been re
190 CHAPTER 7
moved from their original contexts. For exam ple, Frankl's para
doxical intention technique, used originally to help patients gain
perspective on their existential pain, was adopted by pragmatic
behavior therapists who discarded the existential framework. Be-
haviorists have dem onstrated that such techniques can be useful
with specific sym ptom s such as insom nia (Ascher & Turner,
1979), urinary retention (Ascher, 1979), blushing (Lamontagne,
1978), obsessional thought processes (Solyom et al., 1972), and
school attendance (Kolko & Milan, 1983). But downplaying theory
in a framework of technical eclecticism offers a limited and limit
ing vision, not only of paradox, but of possibilities for interven
tion generally. A similar trend can be seen in the family therapy
literature, where paradoxical techniques are being described and I
or explained apart from the models in w hich they were originally
developed (Fisher et al., 1981).
Taking paradoxical intention as a point of departure, we can
begin to recontextualize paradox in the framework of a systemic
theory of behavior and a strategic orientation to intervention and
technique. Explanations of paradoxical intention in the behavior
therapy literature suggest that the intervention works because it
interrupts an "exacerbation cy cle" through which a sym ptom is
maintained and exaggerated. In cybernetic term s, an exacerbation
cycle is a simple positive feedback loop cycling at the level of the
individual (e.g., the more the person tries to go to sleep the more
s/he stays awake). The circularity implicit in the idea of exacerba
tion cycle em bodies the system ic view that the way a problem is
maintained is more relevant to therapy than is the way it started
(Weakland et al., 1974; Sluzki, 1981).
Still, localizing problem m aintenance at the level of the individ
ual overlooks the possibility that similar feedback loops may be
operating at the level of the patient's interaction with other peo
p le -fa m ily m embers, friends, or even therapists. This highlights
a second im portant system s idea—that the social context of a
problem is highly relevant to understanding and changing it.
Thus in all but the first vignette at the beginning of this chapter,
intervention was targeted at relationships as well as sym ptom s.
W hen the conceptual problem unit is larger than an individual
therapy m ust address people other than, or in addition to, the
identified patient.
With respect to strategy or technique, paradoxical intention
raises the question of w hose intention is (or should be) paradoxi
cal—patient's or therapist's? As Frankl used paradoxical intention
(and as m ost behaviorists use it), the "in ten tio n " referred to is
PARADOX IN CO NTEXT 191
clearly the patient's. That is, if the patient can adopt the paradoxi
cal attitude of trying to bring on a sym ptom deliberately, s/he may
lose it by attem pting to keep it. The therapist's intention is not
paradoxical: S/he w ants the patient to do (or at least attem pt to
do) what s/he says. But in other paradoxical approaches (Exam
ples, 2, 5-8), the therapist expects the patient (or family) to do the
opposite of w hat is proposed, and in this sense it is the therapist's
intention that is paradoxical.
The distinction betw een what the patient expects and what the
therapist expects—betw een patient strategy and therapist strat
egy—is fundam ental to the strategic orientation. The strategic
therapist usually does not explain his/her strategy to the patient
and, following Erickson's principle of accepting w hat the patient
brings, attem pts to use the patient's strategy (attem pted solu
tions) as a fulcrum for therapeutic leverage. For greater impact,
suggestions and directives may be framed in a m anner consistent
(or deliberately inconsistent) with a patient's own idiosyncratic
"lan g u age" (Weakland et al., 1974; Watzlawick et al., 1974).
Strategic intervention (of which paradox is a subclass) is de
signed to provoke change irrespective of insight, awareness, or
emotional release. The strategic therapist does not assum e that
change with awareness is more efficient or enduring. In fact, as
M adanes (1980) states, " I f a problem can be solved without the
family knowing how or why, that is satisfactory" (p. 75). The stra
tegic orientation introduces new possibilities for therapeutic inter
vention as well as the occasion for its adherents to view them
selves as m anipulators with potential ethical im plications. As
Watzlawick (1978) notes, the therapist becom es
more a chameleon than a firm rock in a sea of trouble. And it is at
this point that many therapists dig in behind the retort, "Anything
but that," while for others the necessity of ever new adaptations to
the world images of their clients is a fascinating task (p. 141).
The systemic and strategic them es converge in three interre

lated models which together best define a context for paradoxical
intervention. These are: (1) the brief, problem -focused therapy
developed by Watzlawick, Weakland, Fisch, and others at Palo Al
to's M ental Research Institute (MRI); (2) the strategic/structural
approach of Haley and M adanes (Haley, 1976, 1980; M adanes,
1981) and Stanton (1981 [b]); and (3) the systemic family therapy
pioneered by Selvini-Palazzoli et al. in Milan, Italy (Palazzoli et
al., 1978, 1980). Each assumes that problem s are m aintained in
ongoing cycles of interaction and are inextricably interwoven with
192 CHAPTER 7
the social context. In each, the therapist (or team) intervenes de
liberately, on the basis of a specific plan, to resolve the presenting
problem as quickly and efficiently as possible.
THE STRATEGIC SYSTEMS THERAPIES

In the Palo Alto (MRI) brief therapy model, a fundamental
premise is that problems would be self-limiting were it not for the
problem-solving attempts of the people involved. For example,
insomnia may be maintained by trying to go to sleep deliberately,
or depression by well-intentioned reassurance and prodding.
From this perspective, the attem pted solution is the problem, and
interdicting these problem-solving efforts paves the way for
change (Watzlawick et al., 1974; Watzlawick & Coyne, 1980). In
the structural approaches, on the other hand, circular sequences
involve at least three people. These sequences are the basis for
inferring organizational anomalies such as confused family hierar
chy (Haley, 1980) or relationships in which people are too in
volved with one another or not involved enough (M adanes, 1980).
The goal of intervention is to establish a workable relationship
structure.
System descriptions in the Milan approach (Palazzoli et al.,
1980) are more abstract and formulated as hypotheses about the
"rules of the family gam e" in which hom eostatic processes are
responsive to " a family system in danger of ch an ge" (Palazzoli et
al., 1980, p. 7). Here, the clinician's hypotheses specify both the
nature of that danger (e.g., the threat of separation) and ways in
which interlocking behaviors of family m embers protect against
it. Strategic interventions attempt to change dysfunctional family
rules, but not by pointing them out to family m embers. Rather, as
in Examples 7 and 8 at the beginning of the chapter, the rules are
challenged indirectly by positively connoting and prescribing the
very sequences of interaction that define these rules.
Haley (1980) believes that the chief value of system s theory is
that it teaches therapists to recognize repeating sequences and so
make predictions. If there is a fundamental concept in the strate
gic systems therapies and their reliance on paradoxical interven
tions, it is the idea of sequence or problem cycle. Strategic inter
ventions—paradoxical or straightforward—are designed to bring
about change by interrupting recursive (cyclical) patterns of be
havior. The breadth of context considered relevant to intervening
in the cycle defines one's unique approach: For the MRI brief
therapy approach, analysis is usually limited to one or two peo-
pie, while the structuralists generally adopt a triadic view and the
systemic (Milan-style) therapist looks for broad historically based
patterns of interaction.
DIMENSIONS OF PROBLEM
MAINTENANCE
W hatever the unit of interaction, problem cycles may be seen
as governed by prem ises, beliefs, labels (epistemologies), and ex
pectations (axiologies). W hether a premise is functional or dys
functional depends on the context in which it is used. W hen
premises are applied rigidly, they are more likely to becom e the
impetus for problem-maintaining patterns of behavior. A few in
terrelated epistemological or axiological assum ptions appear to be
tied repeatedly to problem m aintenance in one way or another.
We will describe four such dim ensions which govern problem cy
cles at both the individual and interactional level.
Perhaps the most pervasive premise involved in problem m ain
tenance is the dim ension of perceived control. Indeed, some thera
pies maintain strict adherence to certain prem ises concerning peo
ple's control and responsibility over their lives (e.g., Peris et al.,
1951). These therapies also assert a correlated expectation about
the way people should control their lives. Sim ply stated, their
epistemological-axiological position can be summarized by the
statem ents: “People are m asters of their fate and they are respon
sible for their behavior. Furtherm ore, people should take responsi
bility for their own behavior." W hile there are som e contents
where premises of m astery and control are adaptive, there are
many in which such prem ises are dysfunctional and problem
maintaining.
O ne class of problems maintained by a premise of m astery and
control is that in which a person tries to be spontaneous. Here a
person tries to produce a state of affairs which, by its very nature,
requires not trying. For example, an erection is som ething that,
for most men, cannot be willed. It just seem s to happen w hen the
time is right (and, embarrassingly, sometimes w hen the time is
wrong). For the man who is having difficulty maintaining an erec
tion, it may seem reasonable to try to create an erection by willing
one. The problem-maintaining premise of course is that there is a
correlation between effort expenditure and intensity of erection.
The harder he tries, the softer he becom es, which is interpreted as
evidence that he is not trying hard enough, and so on.
194 CHAPTER 7
Another example of a problem cycle m aintained by faying to

assert control in a situation requiring spontaneity is the student
who cannot concentrate on his studies because he is distracted by
more pleasant thoughts. He usually tries to force him self to con
centrate, but without success. It has been found that improve
m ent can be quickly achieved in these cases if the therapist can
get the student to set him self a reasonable time limit for his stud
ies, after which he can do anything except study. Here, leisure
tim e is redefined as punishm ent and consequently loses its lure
(Watzlawick, 1978). As in the case of trying to have an erection,
trying to not be distracted seem s to rest on the prem ise that
pleasant distractions are within the realm of one's personal con
trol. W hen the therapist redefines pleasant distractions as punish
m ent, s/he interrupts the problem cycle of trying, distraction,
more trying, and so on. W hether one is trying to m ake som ething
happen (an erection) or trying to m ake som ething not happen
(pleasant thoughts) it is the trying that represents what Watzla
wick refers to as "m ore of the sam e."
A second set of prem ises often implicated in problem m ainte
nance has been labeled utopian (Watzlawick et al., 1974; Watzla
wick, 1977) and usually takes two form s, utopian expectations
and utopian assertions. Utopian expectations are based on the no
tion that "All life events should be perfect, delightful, and ju st the
way I want them ; nothing less is acceptable." A large num ber of
adolescents who come for consultation (more often than not, at
the request of a parent) m aintain utopian expectations.
Utopian assertions, on the other hand, imply that all is well,
perfect, and delightful in situations where in fact others would
contend that some action is needed. The father of a delinquent
adolescent boy referred by the courts, may talk about his son as
if, rather than having been picked up by the police, the boy had
just been picked as class valedictorian. Utopian assertions are
usually tied to underaction as a m eans of problem m aintenance.
The therapist might define his or her task as reframing the m ean
ing of the boy's behavior so as to energize father into appropriate
action. Just how to do this depends, am ong other things, on the
father's "language," which we will discuss below as a parameter
of technique.
A third class of problem-maintaining beliefs which is particu
larly important in governing com plex patterns of family interac
tion concerns w hether behavior is labeled mad or bad. A label of
mad or bad may be clearly articulated or, like the prem ises already
discussed, may remain covert. In either case, people will act in
accordance with the way they label others' behavior. Those who
are labeled mad are not treated in the sam e way as those labeled
bad, the best example being our legal system 's distinction be
tween "g u ilty " and "n o t guilty by reason of insanity."
In the world of hum an relations, labels have distinct effects on
those labeled (Berger & Luckm ann, 1966). W hether mad-bad la
bels are clearly articulated, for exam ple, "y o u are m entally ill" or
merely implied, w hen these labels are part of a problem se
quence, the therapist's task is to shift the m eaning of that behav
ior so that new patterns of interaction em erge.
A n elegant example of reframing mad-to-bad in the sevice of
therapeutic change is provided in Lynn Hoffm an's (1976) descrip
tion of Jay Haley's training tape "Leaving H om e." In this tape,
Haley from the start reframes his work with a hospitalized 24-
year-old as helping him leave hom e. For this man, who had been
in and out of hospitals for eight years, the problem-maintaining
cycle was a familiar one:
After he came back home, he would become threatening and abu
sive; he would then be moved out to an apartment; after that he
would get on drugs, and go out and cause trouble in the commu
nity. The police would find him, the parents would hospitalize
him, and the whole cycle would start again. (Hoffman, p. 515)
The therapist's first intervention was to have the parents sign a

paper in their son's presence, stating that the next time the young
m an got into trouble, they would have the police put him in jail.
This intervention would prevent the parents from visiting their
son as often as they would if he were in the hospital. More impor
tant is how, without confrontation, this intervention reframed the
son's actions from mad (for which one goes to the mental hospi
tal) to bad (for which one goes to jail).
Later in therapy, the identified patient threatened his m other
with a knife and his sister with a bat. At the next family session,
the therapist arrived with a bat and a knife. He pushed the young
man so hard that he threatened the therapist with the bat. Finally,
the father took the bat from him , and the parents and therapist
agreed that using weapons was unacceptable (bad) and would not
be tolerated. These interventions helped pave the way for other
redefinitions of the family situation which culm inated in the
young man being on his own and out of the hospital.
In this example, undesirable behaviors were labeled as mad by
family m embers, contributing to a long-term problem cycle. Just
as often, however, the label bad governs a problem cycle. For ex
ample, for m any alcoholics, the label badness or moral depravity
is precisely what m aintains the pattern of drinking — immorality
196 CHAPTER 7
— "sw earing o ff" — drinking. In fact, Bateson (1972) attributes

much of the success of Alcoholics Anonym ous to their implicit
redefinition of alcoholism as a sign of m adness, evidenced by
their dictum: "W e admitted we were powerless over our lives—
that our lives had become unm anageable."
The fourth and final premise to be discussed is the definition of
who's up and who's down in a relationship—who takes are of
whom. These hierarchical definitions, like mad and bad labels, se
cure roles for the principals of the scenario. Clinical symptoms
may be seen as a way of defining relationships hierarchically. The
command aspect (Bateson, 1972) of a sym ptom includes the in
junction "care for m e," which can play an im portant role in prob
lem cycles. For example, a young man makes an appointment
with a therapist for him self and his "frig id " wife. This young man
is quite invested in his masculinity, and his obviously angry wife
simply "ca n n o t" enjoy sex with him. The therapist turns to the
wife, and in the husband's presence, gently supports her efforts to
make her husband look good by not enjoying sex. The systemic
reframes of M. Palazzoli and the Milan Group (1978) provide ele
gant examples of changing relational structures by redefining up
as down and down as up.
“LANGUAGE” AND PATIENT

COMPLIANCE
The effectiveness of most intervention in therapy seem s to de
pend in one way or another on the patient's accepting the thera
pist's definition of reality or complying with his or her sugges
tions or directions. Unfortunately, patients often don't listen,
don't follow directives, and may even defy our attem pts to rede
fine their reality.
An important contribution of the Palo Alto group is their em
phasis on maximizing therapeutic influence by framing interven
tions in the client's own "lan g u age" or construct system. From a
cognitive-behavioral perspective, M ahoney (1980) has discussed
the importance of considering a patient's "paradigm " while
Watzlawick (1978) chooses to refer to the person's "w orld view."
Consistent with Erickson's principle of accepting and using what
a patient offers (in a manner analogous to psychological judo), the
assumption is that people are more likely to accept new defini
tions of reality if they represent extensions or variations of their
own views. Consider the following examples:
To a client associating his problem with low self-esteem, we may

consider that he is evidently in need of self-punishment and that
this is an excellent way of fulfilling this need. To somebody in
volved in Eastern thought we may recall the seeming absurdity of
Zen koans And to types like ourselves, we may even lecture in
terms of Group Therapy the Theory of Logical Types, first-order
change and second-order change. (Watzlawick et al., 1974, p. 126)
In practice, a patient's "lan g u ag e" is inferred. The therapist
might ask him- or herself, "H ow does this person want to be
seen ?" Clinically useful answers m ight be: " a s unique and crea
tive," "a s always in control of m yself," "a s a sacrificing spouse and
parent," or even " a s discovering things on my own, not because
som eone else tells m e" (Rohrbaugh et al., 1981). We m ight then
conceive of language as a series of bipolar constructs (Kelly, 1955),
for which each patient perceives a desirable and an undesirable
pole. Examples are "carefree-dutiful," "tough-delicate," "in con
trol-loose." This dialectical nature of language sets the stage for
two distinct form s of intervention, which we will now discuss.
COMPLIANCE AND DEFIANCE

In 1981 (Rohrbaugh et al.) we differentiated two types of para
doxical intervention. In one, change follows from attem pted com
pliance with a therapist's directive; in the other, change follows
from defiance. The com pliance-defiance distinction has been used
by Papp (1980), M adanes (1981), and others (Rohrbaugh et al.,
1982) in discussions of paradoxical and strategic intervention.
Compliance-based interventions, such as prescribing the sym p
tom, work because complying with a therapeutic directive inter
rupts or short-circuits the process that maintains the symptom.
This type of paradoxical sym ptom prescription can be effective
with obsessions, anxiety, insom nia, and other problems that are
maintained in part by attem pts to stave them off. W hen the pa
tient attem pts to bring on such a sym ptom deliberately, s/he can
not continue in usual ways of trying to prevent it, and under
these conditions, the sym ptom often dissolves.
Defiance-based interventions work because people change by re
belling. Haley (1976) cites the example of disengaging an overpro-
tective m other from a symptomatic child. A therapist might ask
the m other to hover even more, setting aside a full hour each
night to warn the child about all of life's dangers. " I f this ap
proach is done well, the m other will react by rebelling against the
therapist and hovering over the child less" (p. 71).
198 CHAPTER 7
We suggested two assessm ent param eters, derived from J. W.

Brehm's (1972) theory of psychological reactance (S. Brehm,
1976), which guide the use of com pliance- and defiance-based
paradox with individuals. O ne concerned the relational stance
vis-k-vis the therapist of the person or persons to be influenced—
specifically, w hether the potential for reactance (rebellion) is high
or low. The second parameter was w hether the person to be in
fluenced (not necessarily the identified patient) perceived the
target behavior as "free," or under voluntary control.
We hypothesized that defiance-based intervention is indicated
w hen reactance potential is high and the target behavior is per
ceived as free, whereas com pliance-based paradox is appropriate
w hen the target behavior is unfree (e.g ., a symptom) and the
probability of reactance is low. Since the overprotective m other in
Haley's example would probably define her hovering as free, she
may do it less if told to do it more. If the child's sym ptom were
stuttering, it would probably be unfree and not a good target for
defiance-based intervention. If the child were cooperative, how
ever, s/he might be asked to practice stuttering deliberately, which
might interrupt problem -m aintaining attem pts to avoid dis-
fluency.
Brehm and Brehm (1981) assume that a good deal of patient
behavior is "u n free," and according to them this seem s to "vitiate
the utility of the free versus unfree distinction. M oreover,. . .th e
'freeness' of the sym ptom may not be as relevant to therapy as
the 'freeness' of trying to cope with the sym ptom " (p. 318).
Brehm and Brehm eliminate the free versus unfree distinction
from our model and work only with the characteristics of the
symptoms and the reactance potential of the patient.
W hile reactance theory has implications for how to intervene,
we used the Palo Alto brief therapy model for guidelines about
where to intervene. The basic premise of brief therapy is that irre
spective of how problems originate, they are m aintained by the
well-intentioned but persistently misapplied solutions of the
problem bearer and those with w hom s/he interacts. The task of
therapy, therefore, is to identify these "problem -m aintaining solu
tion s" (e.g ., a parent's hovering, a spouse's reassurance, a pa
tient's attem pts to conceal his or her own nervousness), and inter
dict them as parsimoniously as possible. In this framework, com
pliance- and defiance-based interventions are used to help people
do less of the same.
Although the com pliance-defiance distinction is helpful in
thinking about the general direction of intervention, the model
itself has som e disadvantages. In practice, concepts such as "reac
tance potential" and "freedom of target behavior" draw attention

away from patterns of interaction to traits and behaviors of indi
viduals. "Psychological reactance" is a hypothetical motive state
aroused w hen a person's perceived freedom is threatened. The
concept is therefore linear (caused by threats to freedom ), individ
ualistic (localized in one person), and not easily reconciled with
the circular, cybernetic, epistem ology of a system s approach. Al
though reactance theory and brief therapy each offer im portant
perspectives on paradox, attem pting to integrate them may only
confuse further what we had hoped to clarify.
MODES OF STRATEGIC INTERVENTION

O ne way to describe the how of strategic therapy and its para
doxical com ponents is in term s of three interrelated m odes of in
tervention: prescribing, reframing, and restraining. Prescribing m eans
telling people what to do (giving tasks, making suggestions, etc.)
either directly or indirectly. The com pliance-defiance m odel pre
dicts that change may com e about regardless of w hether the pa
tient follows the therapist's directive. Reframing involves redefin
ing the m eaning of behavior in a way that makes change possible.
As w e've already suggested, such redefinitions are designed to
alter prem ises or labels which govern problem -m aintaining pat
terns of interaction. Although reframing may resemble interpreta
tion, its goal is to provoke change rather than provide insight and
reality testing—and the accuracy of the redefinition is less impor
tant than its impact.
M any psychotherapies implicitly assum e that the role of the
therapist is to replace a patient's faulty world view with the cor
rect (true, logical, rational) view. Therapy ends w hen the patient's
world view parallels that of the therapist. The strategic therapist,
in his/her use of paradox, contends that m eaning depends on
context. Adaptive functioning does not require being in touch
with reality and having an accurate view of the world. Rather, any
world view that interdicts problem sequences is considered in
planning interventions. This position is nested not only in "fu n c
tional theory," but in a growing empirical literature pointing to the
healing effects of illusion (Lazarus, 1983; Taylor, 1983). Adaptive
functioning has been associated with denial (Lazarus, 1983), the
illusion of control (Alloy & Abram son, 1979; Tiger, 1979) and the
misperception of one's success and its causes (Greenwald, 1980;
M uller & Ross, 1975). High self-esteem (Tennen & Herzberger,
subm itted for publication; Tennen et al., subm itted for publica
200 CHAPTER 7
tion), lack of depression (Lew insohn et al., 1980; Abram son & Al
loy, 1981) and better adjustm ent to illness and injury (Bulman &
Wortman, 1977; Taylor, 1983; Tennen et al., in press) have been
associated with nonveridical and illogical perceptions and beliefs.
The shift from psyche to system and from reality testing to useful
illusions may represent the m ost significant contributions of the
strategic therapist in his/her use of paradox.
Restraining strategies discourage or caution against change in
some way. The therapist's m essage is "y o u should not change as
m uch as you want to," or "y o u shouldn't change in the way that
you want to," or even, "perhap s you shouldn't change—at all."
This intervention alters the therapist's actual or potential role in
side a problem-maintaining system . For example, w hen a patient
has sought help repeatedly but failed to benefit, or therapy is
stuck after straightforward approaches have failed, the therapist
m ight reverse the cycle by being pessim istic or even advising
against change.
In practice, the prescribing, reframing, and restraining modes
are interwoven: Each may be implicit in any intervention, and
each is an im portant elem ent in therapeutic paradox. Before con
sidering some examples of the use of these interventions in com
bination, we will discuss the use of therapeutic restraint in greater
detail.
PROMOTING CHANGE BY
DISCOURAGING IT: RESTRAINT
AS A STRATEGIC
(PARADOXICAL) INTERVENTION
Paradoxical restraining maneuvers are powerful interventions.
They are also easily misused. Their potential for therapeutic influ
ence is high—but their potential for therapist-induced deteriora
tion effects is equally high. The various restraining interventions
can be arranged along a com pliance-defiance continuum (see Fig
ure 1) based on w hether the therapist desires attem pted com
pliance or outright defiance and based on the therapist's creative
use of "langu age."
Implicit restraints are com pliance-based interventions. They
work because the patient accepts what the therapist tells him/her.
For problems maintained by trying too hard, implicit restraints
can effectively interrupt problem cycles. Implicit restraints allow
the patient to do less of the same. From the very start of therapy,
FIGURE 1
Varieties of Therapeutic Restraints

Com pliance Defiance
"go slowly" 'the dangers of improvement' “you c a n 't change'

"set limited goals" ‘you shouldn't change" "you won't benefit
"have your symptom" from treatment"
Im plicit restraint Soft restraint Hard restraint
setting minimum, concrete goals counteracts the problem -m ain

taining, utopian premise that change is an all-or-none affair. Sim
ply convincing a patient that the best way to get going is to get a
slow start can interrupt a problem cycle. A nother variation of im
plicit restraint is the "n o step is too sm all" principle—a notion
well known to behavior therapists. Finally, having a patient vol
untarily express his/her sym ptom can interrupt overaction in an
attempt to ameliorate the problem. In cases of obsessions, certain
sexual dysfunctions, and insom nia—to take a few exam ples—
symptom prescriptions can be considered variants of implicit re
straint because they interrupt trying too hard problem cycles.
Soft Restraint
In the class of interventions we call soft restraint, the therapist
either suggests directly or implies that the patient shouldn't change
the very behavior that s/he or others find troublesom e. O ften the
therapist will worry about the dangers of improvement and ex
press the possible unfavorable consequences of change.
The Milan Group (Palazzoli et al., 1978) describes the use of
what appears to be soft restraint in their work with families. In
one case, the identified patient is a six-year-old boy diagnosed as
severely autistic. After assessm ent sessions, and after taking into
consideration the material gathered from those sessions:
We began by praising Lionel (the identified patient) for his great
sensitivity. He had thought that his grandmother, generous as she
was, needed to love only those who weren't loved. Since Uncle
Nicola (grandmother's son) had gotten married six years ago and
was therefore loved by his wife, and no longer needed his mother's
202 CHAPTER 7
love, Granny was left with no one unloved to love. Thus, ever since
he had been small, he had done everything he could to make him
self unlovable, (pp. 63-64)
The implication of course is that perhaps the identified patient, in

his ultimate wisdom, exhibits his "sy m p tom s" because he is pro
tecting his grandmother and that perhaps he shouldn't do other
wise. W ith regard to soft restraint, note that the therapist in the
case just described implied that change m ight not be an alto
gether good idea and framed that com m ent in a way that en
hanced defiance. Specifically, the family's world view was that
Lionel, the identified patient, was weak and helpless. The thera
pist redefined his sym ptom s as helping his needy grandmother.
To prove the therapist wrong, som ething would have to change.
Jay Haley (1976) offers another example of soft restraint with a
problem child who will not go to school. The therapist, suggests
Haley, can talk to the family about how perhaps the boy should
not go to school. The therapist "ca n suggest that it m ight be bet
ter if the boy ju st stayed hom e and can offer various reasons for
this, depending on the particular family. He m ight say that per
haps the family would get upset if the boy went to school like
normal children and therefore it would be better if he stayed
h om e" (pp. 68-69).
As Haley points out, soft restraint m ust be used carefully. The
therapist is saying things to the family that can be interpreted as
insulting. It is important that the therapist show benevolent con
cern (Haley, 1976) and clearly frame the problem behavior with a
positive connotation (Palazzoli et al., 1978). The greatest danger of
soft restraint is that the individual or family perceive the therapist
as insulting, uncaring, or malevolent. Given w hat w e've said
about utilizing a patient's (family's) "langu age," the therapist's
difficult task begins to crystallize. S/he m ust use the negative pole
of the patient's language without appearing sarcastic or uncaring.
Worrying about w hether change should occur or about the pos
sible dangers of improvement early in therapy is a general strat
egy with m any applications. First, it can be used as a strategy by
itself. Som etim es an initial soft restraint can initiate changes and
problem resolution. Consider the treatm ent of a five-year-old boy
who had never been toilet trained and who several times a day
had a bowel movement in his pants (Haley, 1976). The therapist
decides that he would treat the family by doing only one thing—
restraining the family from improving by being benevolently con
cerned about what would happen to the family if the child be
came normal. He raised the question of w hether the m other could

tolerate being a m other who successfully solved her child's prob
lem . He ever so subtly implied that he thought she could tolerate
success, but he wanted to be sure. The therapist also questioned
the couple's capacity to deal with a free evening together if they
did not have this problem to discuss. A fter three interviews, the
symptom began to remit. O ne way of looking at this soft restraint
is that the therapist makes an intolerable suggestion in a tolerable
fashion.
Soft restraint can also facilitate other (com pliance-based) inter
ventions which the therapist—perhaps reluctantly—offers. For ex
ample, there may be a task which, if carried out, will initiate
change. W ithin the context of a soft restraint, the patient can
prove the therapist wrong by carrying out the task that the thera
pist thought s/he shouldn't jum p into. The task itself need not be
very fancy, it m ight be a very simple assignm ent or a graded task,
like those often suggested by therapists. The purpose of the soft
restraint is to increase the likelihood that the patient will carry out
the task—usually to prove the therapist wrong.
An interesting use of soft restraint is to em bed a paradoxical
symptom prescription within the restraint. W hen carried out sen
sitively, this approach provides a potent "double-barreled" para
doxical strategy, especially w hen a patient seem s unlikely to com
ply. For example, w hen a patient suffering from insom nia has
failed to com ply with several tasks suggested early in the treat
m ent, such as charting his or her sleep schedule, the therapist
might suggest several dangers of improvement, and then imply
that staying awake all night for several nights could offer a better
understanding of the problem (assum ing, of course, "u n d er
standing" were a part of the patient's language), but again it
might not be the best idea because of the potential dangers of
losing the sym ptom . If the patient defies the therapist once
more—by trying to stay up all night for several consecutive
nights—his problem is likely to begin to resolve.
The creativity and subtlety necessary in framing these thera
peutic restraints cannot be overem phasized. Very subtly, the ther
apist defines the implications of rem aining the sam e in a m anner
that is inconsistent with the patient's view of him self and his
world. The therapist should never be sarcastic or degrading. In
stead, ! h : implicit m essage should be "y o u shouldn't change for
the following reasons, all of which are unacceptable to you."
Haley (1976) speaks of restraining as challenging people to be
"norm al." He suggests that these m aneuvers work best with mid
204 CHAPTER 7
dle-class families. A challenge to be normal probably strikes a sen

sitive note with middle-class families because for m any middle-
class families, an im portant dim ension of their world view is
appropriate I norm al—inappropriate I abnorm al. Nonm iddle -class
individuals and families, however, can be induced to change if the
implications of remaining the same are defined in a m anner subtly
inconsistent with their world view.
Hard Restraint
Hard restraint is the most extreme of the restraining strategies.
It requires, perhaps even more than implicit and soft restraints, a
careful assessm ent of the patient's language and a sensitivity to
his or her plight. Here, the therapist suggests that the patient
probably can 't change. The m essage about the improbability of
change can be strongly assertive or mildly pessim istic; in either
case, it is offered empathically.
Perhaps one of the m ost com m on m istakes of novice therapists
applying paradoxical techniques is to confuse hard restraint with
sarcasm or one upsm anship. They forget that if therapy is to help
a person, that person has to remain in treatm ent long enough to
be helped. Not many people will endure a sarcastic therapist.
In a milder form, hard restraint can focus on the patient's ca
pacity to carry out a therapeutic task. Here the therapist predicts
that the patient w on't be able to do what is required to initiate
change. O f course the therapist m entions exactly what the " i t " is
that the patient can't do, and supports his or her pessimistic out
look with the appropriate use of language—in this case the subtle
utilization of the negative pole of the patient's language. O f
course if the patient ju st happens to engage in the task (which
may be a symptom prescription or behavioral task), s/he is likely
to start the change process.
TOWARD A RECONTEXTUALIZATION
OF PARADOX
We can not place the opening vignettes in context. The first is a
simple sym ptom prescription aimed at an intraindividual exacer
bation cycle. The second is a hard restraining maneuver that
breaks a "help-rejecting-com plainer" cycle in which therapists
have repeatedly failed to cure the headaches. Examples 3 and 4
target cycles of family interaction. The frustrated relatives of the

stroke victim had been vacillating betw een encouraging the pa
tient and coercively attem pting to motivate him —but he only be
came more depressed. The paradoxical prescription reversed this
pattern even though the patient him self did not attend the ses
sions. The prescription in Example 4 interrupts another marital
cycle by subtly reframing the sym ptom : Since the husband may
only be "pretending," it is now difficult for the wife to continue in
her usual ways of responding to him.
In Examples 5 and 6, paradox is used to change relationship
structures. W hen asked to take all of the responsibility for her
grandchild, the grandmother backs off, establishing clearer gener
ational boundaries. W hen the psychiatrist exaggerates the prob
lem-maintaining complementarity of the couple in Example 6,
they reorganize in a more symmetrical way. Both of these inter
ventions could be called defiance-based, but it would be a mistake
to say they work because the people involved are "d efian t." The
systemic explanation is that paradox introduces an imbalance in
the family (or in the balance of internal and external forces for
change) such that the system recoils to a more workable pattern of
organization (Hoffman, 1981).
Examples 7 and 8 illustrate Milan-style systemic paradox in
which the entire pattern of problem-maintaining family interac
tion is positively reframed and prescribed (Palazzoli et al., 1978).
These interventions sim ultaneously com bine several modes of
strategic intervention. The therapist reframes the identified pa
tient's behavior as protective of his parents, thereby altering the
hierarchical m eaning of the sym ptom . Consistent with this re
framing, the therapist restrains the family from precipitous change
(an example of soft restraining). The implicit prescription is that no
one change just yet since a shift might destabilize a delicately bal
anced family structure. The strategy is defiance-based in that
change comes about by the family reorganizing against the thera
pist's cautionary directives. In this approach, the paradoxical
stance is often m aintained throughout the therapy and imple
mented by a team of therapists using a one-way mirror. The con
sultation team plans the strategy and provides outside support for
the therapists who work with the family directly. The team is also
a powerful source of therapeutic leverage in the form of high-
impact, hard-to-disqualify m essages from unseen experts. Papp
(1980), in likening the consultation team to a Greek chorus, shows
how some (outside) team members can take a position against
change, so that others (inside) can encourage it more directly.
206 CHAPTER 7
RESEARCH STRATEGIES
The strategic approach to treatm ent described in this chapter
has now developed to the point where empirical research is both
possible and necessary. There appear to be two m ajor research
tasks (Rohrbaugh & Eron, 1982). O ne is the docum entation and
clarification of the role of interactional-contextual factors in the
development and m aintenance of clinical problem s. The work of
M inuchin, Rosm an, and Baker (1978), W ynne (1978), Coyne
(1976), and M adanes, Dukes, and Harbin (1980) represent signifi
cant first attem pts in this area.
The second research task is to dem onstrate that the strategic
therapies compare favorably to other well-established treatm ents
(see Stanton, 1981 [b]; Gurm an & K niskem , 1981). Brehm and
Brehm (1981) note that m any clinical problem s described in our
previous work as appropriate for com pliance-based strategies
would also seem to be appropriate for treatm ent by systemic de
sensitization (Wolpe, 1959). In view of the strong evidence for the
effectiv en ess o f system ic d esen sitizatio n (e. g. , K azdin &
W ilcoxon, 1976), Brehm and Brehm (1981) believe that it is the
treatm ent of choice for anxiety reduction.
Defiance-based strategies have received few empirical tests;
m ost supportive evidence derives from case studies. Studies by
Ayllon, Allison, and Kandel (unpublished m anuscript) and Kolko
and Milan (1983) used multiple baseline designs. Unfortunately,
these studies involved very small samples of children.
O ne reason for the dearth of empirical research in this field is
that the notion of patterns, which underlies strategic therapies,
presents serious methodological issues (Dell, 1980; Abeles, 1976).
Therefore, research on clinical outcom e may have to advance in
the absence of "b a sic " research. The "e ffe cts" of the strategic
therapies m ust be held up to the same scrutiny as com peting
therapeutic approaches. In a series of studies on procrastination
and depression, Strong and associates (Beck et al., 1982; Feldman
et al., 1982; Lopez & Wambach, 1982; Strong & Clayborn, 1982;
W right & Strong, 1982) test the effectiveness of positive connota
tion, sym ptom prescription, and defiance-based strategies.
These studies are com m endable because they test directly the
efficacy of certain paradoxical interventions. Unfortunately, each
has some serious limitations. Each study uses a no-treatm ent con
trol group rather than a placebo control. In each study, the thera
pists were graduate students and the dependent m easure was
self-report. Subjects received the sam e intervention without con
sideration of language in the service of com pliance regulation. Fi
nally, these studies used student subjects who received course

credit for their participation. These factors limit the ecological va
lidity of the investigations. N onetheless, they represent the first
systematic investigations of paradox in psychotherapy.
ETHICAL ISSUES
The question arises: To have a system s orientation is it neces
sary to be strategic and manipulative? Perhaps not, yet system s
by definition are controlled. To a large extent, cybernetics is the
study of reciprocal influence and control. In therapy, therefore,
one cannot not influence, just as one cannot not com m unicate
(Watzlawick et al., 1967). The question is not w hether to influ
ence, but how to do it m ost constructively. N eedless to say, strate
gic m ethods raise im portant ethical questions, such as what is a
reasonable degree of inform ed consent, and should or can it be
obtained from all w hose lives may be touched by an intervention?
Significantly, therapists such as Haley and Palazzoli who pio
neered the use of paradoxical m ethods are now giving them less
em phasis. Indeed, as paradoxical techniques becom e more pop
ular, there is reason for concern about ways in which they can be
m isused. Encouraging a sym ptom or restraining people from
changing can be disastrous if done sarcastically or from a sense of
frustration ("T h ere's the window—go ahead and ju m p !"). Nor
should paradox be used for shock value or to give the therapist a
sense of power (Weeks & L'Abate, 1979). Since strategic work
presents the special risk of making patients objects for dehum an
ized treatm ent by therapists, it is m ost im portant that paradoxical
m ethods be used in a systemic framework that views the therapist
or team as part of the environm ent being modified.
The strategic-paradoxical approach to psychotherapy appears
to raise ethical concerns am ong its practitioners as well as its
critics. We have been w arned against giving prepackaged direc
tives and against m aking strategic interventions without linking
them to the family system (Stanton, 1981 [b]). We have also been
reminded that paradoxical interventions may be manipulative
with respect to concealing treatm ent goals, use of a controlling
m ethod, and absence of inform ed consent (Weeks & L'Abate,
1982).
Perhaps the m ost extensive discussion of the ethics of paradoxi
cal interventions is found in the work of Haley (1976), who reminds
us that all therapy is "m anipulative." The psychoanalyst, for exam
ple, decides upon the proper tim ing and depth of an interpreta
208 CHAPTER 7
tion. Haley correctly points out that by considering issues of timing

and depth, the analyst conceals many of his/her observations from
the patient, until the patient is ready to make optimum use of the
insights offered. Behavior therapists similarly reinforce their cli
ent's adaptive behaviors without always announcing their intent.
Haley goes so far as to claim that w hen a therapist refrains from
offering insight, s/he is simply being courteous.
A thornier and perhaps the most crucial ethical issue facing
practitioners of strategic-paradoxical interventions is w hether a
therapist can ethically give a patient false information in an at
tempt to initiate change. Haley (1976) believes that it is generally
unwise for a therapist to lie to a patient, but he cautiously en
dorses more "com plicated lies" such as a therapist telling a pho
bic patient that the therapist wants the patient to be afraid in a
particular situation.
W hen confronted with this ethical dilemma, we may be
tempted to retreat to epistemological sanctuaries. For example, we
might claim that it is naive to be concerned with truthtelling in
therapy, since there is no such thing as objective reality. There is
now considerable empirical support for the saying "th e truth
hurts." Depressed people appear to be more realistic about their
ability to control their environment (Alloy & Abramson, 1979) and
their effects on others (Lew insohn et al., 1980) than are nonde
pressed people. W hen the depression lifts, these people join their
nondepressed counterparts in maintaining an "illusory glow." Is a
therapist bound to a code of ethics which forces him/her to reveal
the depressing truth to a formerly depressed patient?
The epistemologically oriented therapist might also note that
denial has many benefits (Lazarus, 1983; Taylor et al., unpub
lished manuscript). Acknowledging certain "tru th s" has been as
sociated with poorer coping and mood disturbance (Affleck et al.,
1982). Does a therapist have an obligation to point out these
harmful truths, or is Haley (1976) correct in labeling this behavior
"discourteous?"
Finally, one might note that false beliefs about blam eworthiness
have been associated with better coping and less personal distress
in young diabetic patients (Tennen et al., in press), breast cancer
patients (Timko & Janoff-Bulm an, submitted for publication), and
accident victims (Bulman & Wortman, 1977). O nce again, should
a therapist be required to tell "th e whole truth," or is the
withholding of certain truths acceptable?
These epistemological arguments can lead a therapist to ignore
truthfulness as an ethical issue. In the treatm ent situation, how
ever, epistemology cannot claim priority over ethics, and the
moral domain of intended deception m ust be considered along

with issues of truth and falsity (Bok, 1978).
The clinician who uses paradoxical interventions runs up
against these and other complicated ethical issues, which might
be minimized by the following four guidelines for not misusing
paradoxical interventions:
1. Define behavior positively. Avoid attributing unseem ly motives
to people (like needing to "control," "resist," or "d e fe a t" one an
other). W hen reframing, ascribe noble intentions not only to the
sym ptom but to what other people are doing to support it (Hoff
man et al., 1981; Papp, 1980; Stanton, 1981 [b]). Be m ost careful
with challenging or provocative interventions such as in Example
2 at the beginning of the chapter: it is better to suggest that
change is not advisable (soft restraining) than to predict it will not
be possible (hard restraining).
2. Consider context. Evaluate interactional as well as individual
levels of problem m aintenance. W hen both are present, give prec
edence to the former.
3. Work with colleagues. In paradoxical work with difficult cases,
it is most helpful to have a consultation team and one-way mirror,
or at the very least to discuss the case with a supervisor or col
leagues. It is easy to lose one's m oorings, and outside consultants
provide valuable grounding and support.
4. Have a theory. The m ost im portant guideline for paradoxical
(or any other) intervention is to have a coherent rationale for us
ing it. This requires a clear form ulation of how a sym ptom is be
ing maintained and how a particular intervention will change the
pattern of problem m aintenance. If we are saying anything here,
it is that theory and technique are (or should be) inseparable. Be
fore telling patients not to change, a strategic system s therapist
should give as much attention to problem cycles, confused hierar
chies, and family rules as a psychoanalyst would give to intensity
of transference, level of resistance, and the availability of a reason
able ego before offering an interpretation.
REFERENCES
Abeles, G. Researching the unresearchable: Experimentation on the
double bind. In C. E. Sluzki & D. C. Ransom (Eds.), Double bind.
New York: Grune & Stratton, 1976.
Abramson, L. Y., & Alloy, L. B. Depression, nondepression, and
cognitive illusions: Reply to Schwartz. Journal of Experimental
Psychology: General, 1981, 110, 436-447.
210 CHAPTER 7
Affleck, G., Allen, D., McGrade, B. J., & McQueeney, M. Maternal

causal attributions at hospital discharge of high risk infants.
American Journal of Mental Deficiency, 1982, 86, 575-580.
Alloy, L. B., & Abramson, L. Y. Judgment of contingency in depressed
and nondepressed students: Sadder but wiser? Journal of
Experimental Psychology: General, 1979, 108, 441-485.
Ascher, L. M. Paradoxical intention in the treatment of urinary
retention. Behaviour Research and Therapy, 1979, 17, 267-270.
Ascher, L. M., & Turner, R. M. Paradoxical intention and insomnia: An
experimental investigation. Behaviour Research and Therapy, 1979, 17,
408-411.
Ayllon, T., Allison, M. G., & Kandel, H. G. Changing behavior through
systematic verbal persuasion. Unpublished manuscript, Georgia State
University, 1980.
Bateson, G. Steps to an ecology of mind. New York: Random House,
1972.
Bateson, G., Jackson, D. D., Haley, J., & Weakland, J. H. Toward a
theory of schizophrenia. Behavioral Science, 1956, 1, 251-264.
Beck, J., & Strong, S. Stimulating therapeutic change with
interpretations. A comparison of positive and negative
connotations. Journal of Counseling Psychology, 1982, 29, 551-559.
Berger, P. L., & Luckmann, T. The social construction of reality.
Doubleday, 1966.
Bok, S. Lying: Moral choice in public and private life. New York: Random
House, 1978.
Brehm, J. Responses to loss of freedom: A theory of psychological reactance.
Morristown, N.J.: General Learning Press, 1972.
Brehm, S. The application of social psychology to clinical practice.
Washington, D.C.: Hemisphere, 1976.
Brehm, S. S., & Brehm, J. W. Psychological reactance: A theory of freedom
and control. New York: Academic Press, 1981.
Bulman, R. J., & Wortman, C. B. Attributions of blame and coping in
the "real world": Severe accident victims react to their lot. Journal
of Personality and Social Psychology, 1977, 35, 351-363.
Coyne, J. C. Toward an interactional description of depression.
Psychiatry, 1976, 39, 28-40.
Dell, P. F. Researching the family theories of schizophrenia: An exercise
in epistemological confusion. Family Process, 1980, 19, 321-326.
Dunlap, K. Repetition in the breaking of habits. Science Monthly, 1930,
30, 66-70.
Feldman, D. A., Strong, S. R., & Danser, D. B. A comparison of

paradoxical and nonparadoxical interpretations and directives.
Journal of Counseling Psychology, 1982, 29, 572-579.
Fisher, L., Anderson, A., & Jones, J. E. Types of paradoxical
intervention and indications/contraindications for use in clinical
practice. Family Process, 1981, 20, 25-36.
Foucalt, M. Madness and civilization: A history of insanity in the age of
reason. New York: Random House, 1973.
Frankl, V. Paradoxical intention: A logotherapeutic technique. American
Journal of Psychotherapy. 1960, 24, 520-535.
Greenwald, A. G. The totalitarian ego: Fabrication and revision of
personal history. American Psychologist, 1980, 35, 603-618.
Gurman, A. S., & Kniskern, D. P. (Eds.). Handbook of family therapy.
New York: Brunner/Mazel, 1981.
Erickson, M.D. New York: W. W. Norton, 1973.
Haley, J. Development of a theory of a research project. In C. Sluzki,
D.C. Ransom (Eds.), Double bind. New York: Grune & Stratton,
1976.
Haley, J. Problem-solving therapy. San Francisco, Calif.: Jossey-Bass,
1976.
Haley, J. Leaving home: The therapy of disturbed young people. New York:
McGraw-Hill, 1980.
Hoffman, L. Breaking the homeostatic cycle. In P. J. Guerin (Ed.),
Family therapy: Theory and practice. New York: Gardner, 1976.
Hoffman, L. Foundations of family therapy. New York: Basic, 1981.

Hoffman, L. Therapeutic paradoxes: A model for working with difficult
families. Conference presented by faculty of Ackerman Institute for
Family Therapy, New York, 1981.
Kazdin, A. E., & Wilcoxon, L. A. Systematic desensitization and
nonspecific treatment effects: A methodological evaluation.
Psychological Bulletin, 1976, 83, 729-758.
Kelly, G. A. The psychology of personal constructs (Vols. I and II). New
York: Norton, 1955.
Kolko, D. J., & Milan, M. A. Reframing and paradoxical instruction to
overcome "resistance" in the treatment of delinquent youths:
A multiple baseline analysis. Journal of Consulting and Clinical
Psychology, 1983, 52, 655-660.
Lamontagne, Y. Treatment of erythrophobia by paradoxical intention.
Journal of Nervous and Mental Disease, 1978, 266, 304-307.
212 CHAPTER 7
Lazarus, R. S. The costs and benefits of denial. In S. Breznitz (Ed.),

Denial of stress. New York: International Universities Press, 1983.
Lewinsohn, P.M ., Mischel, W., Chaplin, W., & Barton, R. Social
competence and depression: The role of illusory self-perceptions.
Journal of Abnormal Psychology, 1980, 89, 203-212.
Lopez, F. G., & Wambach, C. A. Effects of paradoxical and self-control
directives in counseling. Journal of Counseling Psychology, 1982, 29,
115-124.
Madanes, C. Protection, paradox, and pretending. Family Process, 1980,
19, 73-85.
Madanes, C., Dukes, J., & Harbin, H. Family ties of heroin addicts.
Archives of General Psychiatry, 1980, 37, 889-894.
Mahoney, M. J. Psychotherapy and the structure of personal
revolutions. In M. J. Mahoney (Ed.), Psychotherapy Process. New
York: Plenum, 1980.
Miller, D. T., & Ross, M. Self-serving biases in the attribution of
causality: Fact or fiction? Psychological Bulletin, 1975, 82, 213-225.
Minuchin, S., Rosman, B., & Baker, L. Psychosomatic families.
Cambridge: Harvard University Press, 1978.
Palazzoli, M., Boscolo, L., Cecchin, G. F., & Prata, G. Paradox and
. Palazzoli, M., Boscolo, L., Cecchin, G. F., & Prata, G. Hypothesizing,
circularity, neutrality: Three guidelines for the conductor of the
session. Family Process, 1980, 19.
Peris, F., Hefferline, R. F., & Goodman, P. Gestalt therapy. New York:
Dell, 1951.
Rohrbaugh, M., & Eron, J. The strategic systems therapies. In L. E.
Abt, R. I. Stuart (Eds.), The newer therapies: A workbook. New York:
Van Nostrand Reinhold, 1982.
Rohrbaugh, M., Tennen, H., & Eron, J. Paradoxical interventions. In
J. H. Masserman (Ed.), Current Psychiatric Therapies, 21. New York:
Grune & Stratton, 1982.
Rohrbaugh, M., Tennen, H., Press, S., & White, L. Compliance,
defiance and therapeutic paradox: Guidelines for strategic use of
paradoxical interventions. American Journal of Orthopsychiatry, 1981,
51, 454-467.
Rosen, J. Direct psychoanalysis. New York: Grune & Stratton, 1953.
Sluzki, C. Process of symptom production and patterns of symptom

maintenance. Journal of Marriage and Family Therapy, 1981, 7,
273-280.
Solyom, L., Garza-Perez, J., Ledwidge, B. L., & Solyom, C. Paradoxical
intention in the treatment of obsessive thoughts: A pilot study.
Comprehensive Psychiatry, 1972, 13, 291-297.
Stanton, M. D. An integrated structural/strategic approach to family
therapy. Journal of Marriage and Family Therapy, 1981, 7, 427-440. (a)
Stanton, M. D. Strategic approaches to family therapy. In A. S.
Gurman, D. P. Kniskern (Eds.), Handbook of family therapy. New
York: Brunner/Mazel, 1981. (b)
Strong, S. R., & Clayborn, C. D. Change through interaction: Social
psychological processes of counseling and psychotherapy. New York: John
Wiley, 1982.
Taylor, S. E. Adjustment to threatening events. American Psychologist,
1983, 38, 1161-1173.
Taylor, S. E., Lichtman, R. R., & Wood, J. V. Adjustment to breast cancer:
Physical, socio-demographic, and psychological predictors. Manuscript
submitted for publication.
Tennen, H., Affleck, G., Allen, D. A., McGrade, B. J., & Ratzan, S.
Causal attributions and coping with insulin-dependent diabetes.
Basic and Applied Social Psychology, in press.
Tennen, H., & Herzberger, S. Depression, self-esteem, and the absence of
self-protective attributional biases. Manuscript submitted for
publication.
Tennen, H., Nussbaum, N., Sharp, J. P., & Baldya, B. Depressive
attributional style: The role of self esteem. Manuscript submitted for
publication.
Tiger, L. Optimism: The biology of hope. New York: Simon & Schuster,
1979.
Timko, C., & Janoff-Bulman, R. An attributional model of coping with
breast cancer. Presented at the meetings of the American
Psychological Association, August 1982.
Watzlawick, P. How real is real? Confusion, disinformation, communication.
New York: Random House, 1976.
Watzlawick, P. The language of change. New York: Basic, 1978.
Watzlawick, P., Beavin, J., Jackson, D. D. Pragmatics of human
Watzlawick, P., & Coyne, J. C. Depression following stroke: Brief
problem-focused treatment. Family Process, 1980, 19, 13-18.
214 CHAPTER 7
Watzlawick, P., Weakland, J. H., & Fisch, R. Change: Principles of

problem formation and problem resolution. New York: W. W. Norton,
1974.
Weakland, J. H., Fisch, R., Watzlawick, P., & Bodin, A. Brief therapy:
Weeks, G. R., & L'Abate, L. A complication of paradoxical methods.
Mazel, 1982.
Wright, R. M., & Strong, S. R. Stimulating therapeutic change with
directives: An exploratory study. Journal of Counseling Psychology,
1982, 29, 199-202.
Wynne, L. C. Knotted relationships, communication deviances, and
metabinding. In M. E. Berger (Ed.), Beyond the double bind. New
York: Brunner/Mazel, 1978.
Part Three
Research on Paradoxical
Techniques
8
An Introduction to Research on the
Clinical Efficacy of Paradoxical
Intention
by Ray S. Kim, MA
James Poling, MA
L. Michael Ascher, Ph.D.
In the first edition of this book, Ascher, Bowers, and Schotte (1985)
reviewed the data of m any of the studies incorporating control— or
some sem blance of control— procedures that were available at that
time. U sing their review as a point of departure, the present
authors aim to discuss research that largely has been undertaken
subsequently. O f course, science moves in a slow and deliberate
fashion, and an updated review on this topic would ordinarily not
have been warranted after the relatively brief interval since the
appearance of Ascher, Bowers, and Schotte's earlier work (1985);
the am ount of additional data do not, on their own m erit, require
a progress report at this tim e. However, as opportunities do not
always arise under optim al circum stances, the present authors
agreed to revise the chapter of Ascher, Bowers, and Schotte (1985)
as sufficient new m aterial was available to perm it at least one ju st
noticeable difference betw een the former and the present chapters.
A ttesting to their popularity, various forms of therapeutic para
216
THE CLINICAL EFFICACY OF PARADOXICAL INTENTION 217
doxical procedures appear in alm ost every approach to psycho

therapy (e.g., Seltzer, 1986). It was therefore inevitable that behav
ior therapists would becom e interested in them as well. A s each
group of therapists applies these techniques in a m anner that is
roughly consistent with their own orientation, behavior therapists
have approached therapeutic paradox in a characteristic fashion.
In applying the procedures, behavior therapists have used the tra
ditional m anner of supporting conventional behavioral program s
by enhancing cooperation (Ascher, 1989a). More unique has been
the use of paradoxical intention as a central procedure in programs
directed toward the amelioration of anxiety disorders and anxiety-
based difficulties (Ascher, 1989b). Here, the attem pt has been to
dem onstrate that paradoxical intention can serve as a central con
cept in organizing conventional behavioral program s. In close
association w ith the clinical practice is a basic tenet of behavior
therapy: the em pirical validation of the efficacy of procedures
employed. The em phasis in this updated ch apter is on controlled
studies in which subjects exhibit clinically significant levels of the
target behavioral disorder.
RESEARCH IN PARADOXICAL TECHNIQUES
Insomnia
Insom nia, the chronic inability to fall asleep w ithin a satisfac
tory period of time and/or to m aintain a satisfactory level of sleep,
has been estim ated to affect 10 to 15 percent of the population in
its m ilder form s and 10 to 15 percent in its more severe form s
(Kales, Bixler, Lee, Healy, & Slye, 1974; W ebb, 1975). Research on
the use of paradoxical intention in the treatment of insom nia dem
onstrates the increasing experim ental sophistication in the inves
tigation of paradoxical techniques.
Early reports were based on uncontrolled case studies (e.g.,
Ascher, 1975), and later progressed to single case, experim ental
designs of varying degrees of sophistication (e.g., A scher & Efran,
1978; Relinger & Bom stein, 1979; Relinger, B om stein, & M ungas,
1978). Recent investigations have com pared paradoxical intention
to placebo and no-treatment control groups (e.g., Ascher & Turner,
1979a), as well as to other more established behavioral treatm ents
(e.g., Lacks, Bertelson, G ans, & K unkel, 1983; Espie, Lindsay,
Brooks, Hood, & Turvey, 1989; Turner & Ascher, 1979, 1982).
The first, em pirically based claim for the effectiveness o f para
doxical intention as a treatm ent for sleep-onset insom nia was
218 CHAPTER 8
m ade by A scher (1975). At the tim e, A scher reported success in

the treatm ent of a sm all num ber of clients w ho had sleep-onset
insom nia that was refractory to a standard behavioral program .
However, the om ission of experim ental controls w ith som e clients
precludes m aking causal inferences for a variety of reasons
(A scher & Schotte, 1983; H ersen & Barlow, 1976).
Building on this early success w ith the technique, A scher and
Efran (1978) applied paradoxical intention in the treatm ent of 5
individuals com plaining of clinically significant levels of sleep-
o nset insom nia w ho had not benefitted from a standard behav
ioral treatm ent program . Client self-report data on a num ber of
variables (e.g., m ood w hen retiring, latency of sleep onset, tim e
of retiring, num ber of awakenings, restfulness of sleep) were col
lected pre- and posttreatm ent. Paradoxical intention instructions
were presented using either an assessm ent rationale or a veridical
explanation, with the intention of having the client attem pt to stay
awake at bedtim e. The assessm ent rationale su bjects were asked
to stay awake so that their therapists could gain detailed descrip
tions of their thoughts prior to falling asleep. The su bjects receiv
ing the veridical explanation were asked to lengthen the period
of bedtim e relaxation, even if this m eant resisting the urge to fall
asleep.
In response to the paradoxical intention instructions, clients
often reported that they had been unable to comply with the direc
tives because they had fallen asleep too quickly. Self-reported,
sleep-onset latency decreased from a pretreatm ent m ean of 48.6
m inutes to less than 9 .8 m inutes during the second week of para
doxical therapy. A further test of this treatm ent was m ade by ask
ing one su bject to return to the initial behavioral treatm ent
program for an additional 3 w eeks. D uring this period, the self-
reported, sleep-onset latency increased from 6 .0 to 28.3 m inutes.
Subsequent readm inistration of paradoxical instructions once
again resulted in a reduction in sleep-onset latency for this
subject.
Bornstein and his associates conducted several investigations in
an effort to replicate and extend the findings of A scher and Efran
(1978). In the m ost interesting of these, Relinger and Bornstein
(1979) employed paradoxical intention to treat 5 clients com plain
ing of severe, chronic sleep-onset insom nia. The design incorpo
rated a multiple baseline across su bjects. Random ly selected
su bjects were treated sequentially, each receiving one more week
of baseline assessm ent than the previously assessed individual.
Self-report data on multiple variables were collected following pre-
and posttreatm ent intervals and at follow-up.
Paradoxical instructions were provided in association w ith an

assessm ent rationale. T his was supplem ented by counterdem and
instructions (Relinger, Bornstein, & M ungas, 1978) to control for
the effects o f su bjects' expectancies and dem and factors. Thus,
su bjects were inform ed not to anticipate any im provem ent until
the end of treatm ent. Presumably, any effect occu rring despite
these instructions could be attributed more to the treatm ent than
to experim ental dem ands. W hile not as powerful as designs
involving a return to the baseline phase, this approach lends more
credence to claim s of efficacy than do the AB and A BC designs
used in previous studies.
Over the course of five treatm ent sessions, sleep-onset latency
decreased by over 50 percent from a baseline average of 110 m in
utes to a posttreatm ent, sleep-onset latency of 47 m inutes. By the
end of the three-m onth follow-up period, clients reported an aver
age sleep-onset latency of 20 m inutes. Statistically significant
improvements were also found for degree of restfulness, difficulty
falling asleep, and num ber of awakenings, both at the end of treat
m ent and at follow-up. In addition, results supported the efficacy
of the counterdem and control procedure.
W hile these case studies support the treatm ent of sleep-onset
insom nia w ith paradoxical intention, a series of case studies
reported by Espie and Lindsay (1985) highlights the variability of
client response to paradoxical intention. Six su bjects reporting
chronic sleep-onset insom nia received therapy utilizing paradox
ical intention over an eight-w eek period. Although three su bjects
obtained a rapid decrease in sleep-onset latency, three evidenced
significantly exacerbated insom nia. Eventually, 1 subject from the
latter group did exhibit im provem ent following several w eeks of
additional sessions; the rem aining two clients were unable to con
tinue with paradoxical intention but experienced success w ith pro
gressive relaxation.
For a variety of reasons, including factors inherent in the single
case designs of studies described to this point, it is im possible to
dism iss, w ith confidence, the influence of nonspecific factors in
the positive results obtained. Therefore, experim ents incorpora
ting appropriate controls and com parisons w ith other effective
behavioral treatm ents are necessary to validate the qualifications
of innovative behavioral techniques.
The first of these investigations was conducted by Turner and
Ascher (1979). They com pared paradoxical intention to behavioral
techniques that were established as treatments of choice for sleep-
o nset insom nia. Individuals com plaining o f clinically significant
levels of sleep-onset insom nia were random ly assigned to 1 of 5
220 CHAPTER 8
groups: paradoxical intention, stim ulus control, relaxation train

ing, attention placebo, and no-treatment control. In addition to the
self-report outcome m easures typically employed in such studies,
ratings of sleep behavior from spouses or room m ates were
obtained. Paradoxical instructions were presented in a straightfor
ward treatm ent rationale. Statistical analysis dem onstrated the
superiority of the active behavioral treatments over the two control
conditions. In contrast, paradoxical intention did not differ from
the treatm ents of choice.
A m ajor criticism of this study concerns the use of only 1
experienced therapist who was not blind to the hypothesis
being tested . The largely subjective nature of the dependent
m easures was another problem . A lthough the results of a vari
ety of control and evaluation procedures (e.g., counterdem and
instructions, subject ratings of treatment credibility and relevant
therapist characteristics) served to m itigate the criticism , the
concerns remain valid and limit the extent to which the results
are generalizable.
In a partial replication of their original study (Turner & Ascher,
1979), A scher and Turner (1979a) random ly assigned 25 individ
uals (10 males, 15 fem ales) w ith clinically significant levels of
sleep-onset insom nia to paradoxical intention, credible placebo (a
quasi-desensitization procedure), or no-treatment conditions. The
paradoxical intention su bjects reduced their sleep-onset latency
by over 50 percent and showed statistically significant improve
m ents on m ost self-report m easures, as com pared to su bjects in
the placebo and no-treatm ent conditions. No difference in cred
ibility was reported by su bjects regarding the placebo and para
doxical treatments. Spouse-roomm ate checks again supported the
reliability of the sleep-onset data. This study and Turner and
Ascher's earlier study provide evidence for the effectiveness of
paradoxical intention as a treatm ent for sleep-onset insom nia.
However, the criticism s associated w ith Turner and A scher (1979)
apply to this study as well.
To study the role of experience in the adm inistration and out
come of paradoxical intention in Turner and Ascher (1979), Turner
and A scher (1982) com pared the data obtained in the 1979 study
w ith data collected using less experienced therapists. The results
were interesting, but inconclusive. For each of the treatm ent
groups, posttreatment, sleep-onset latencies were twice the length
of the sleep-onset latencies obtained by the experienced therapist.
This result is surprising since the literature generally indicates that
experim enter differences were not obtained for progressive relax
ation or stimulus control. W ith regard to paradoxical intention, the
su bjects seen by the inexperienced therapists failed to dem on

strate significant posttreatm ent reductions in sleep-onset latency,
as compared to the significant improvements obtained by the sub
jects of the experienced therapist.
Although a variety of factors m ight explain these differences,
an im portant distinction concerns the su bject pools. The baseline
sleep-onset latency of the paradoxical intention group in the repli
cation study was significantly shorter than that of any other group
treated by either experienced or inexperienced therapists. This
finding suggests that initial sym ptom severity may affect response
to treatm ent, indicating the possible need to control for sym ptom
severity before direct com parisons betw een studies can be m ade
(cf., Lacks et al., 1983). Certainly, from a data analysis perspective
the paradoxical intention groups in the two studies had disparate
potentials for improvement.
Results sim ilar to Turner and A scher (1979), and A scher and
Turner (1979a) were obtained by Ladouceur and Gros-Louis
(1986). In this study, 27 su bjects were random ly assigned to one
of four experim ental conditions: paradoxical intention, stim ulus
control, an educational control condition, and a no-treatment con
dition. Procedural details were generally similar to those of Turner
and Ascher's study (1979). A lthough both paradoxical intention
and stim ulus control produced results superior to control condi
tions, the two behavioral treatm ents were equally effective. Since
the m ean sleep-onset latency in this study was approxim ately 60
m inutes, as it was in the Turner and Ascher (1979) and Ascher and
Turner (1979a) studies, direct com parisons am ong the studies
seem appropriate.
O tt, Levine, and A scher (1983) com pared the efficacy of
paradoxical intention using both self-report inform ation and,
possibly, more objective inform ation obtained from a sleep-
monitoring unit. Fifty-five subjects with an average sleep-onset
latency of 60 m inutes were randomly assigned to one of the fol
lowing conditions: paradoxical intention, feedback from a sleep-
m onitoring u n it, feedback plus paradoxical in tention , or
no-treatm ent. Pretreatm ent analyses revealed no differences
am ong groups regarding age, sex, problem duration, and
M M PI scores. Posttreatm ent analyses found no group differ
ences regarding adherence to instruction, correctness of sleep
log, and constancy of m edication intake. Counterdem and
instructions and an assessm ent rationale were used in an
attem pt to offset dem and characteristics.
D ecreases in sleep-onset latency were found in su bjects receiv
ing paradoxical intention alone or feedback alone, w hile su bjects
222 CHAPTER 8
receiving both treatments actually experienced increases in sleep-

onset latency. In an attem pt to explain this latter ostensibly anom
alous finding, the authors suggest that the use of the recording
device may have facilitated the su bjects' literal adherence to the
therapist's request to "rem ain awake as long as p o ssible."
Am ong this study's shortcom ings is the use of univariate an al
yses w here a multivariate analysis was indicated. Due to the
greater num ber of analyses bein g perform ed, the chances of
obtaining spurious results are increased. In addition, only 1 ther
apist conducted all of the therapy sessions, w hich, of course,
introduces the possibility of experimenter bias. A further problem
is that the "o b jectiv e" m easure could be m anipulated by the sub
ject. Finally, the experim ental design does not seem com pletely
suitable for the stated objectives.
Lacks, Bertelson, G ans, and Kunkel (1983) studied the interac
tion betw een the level of severity of sleep-onset latency and var
ious behavioral treatments associated with success in the treatment
of insom nia. Sixty-four su bjects were m atched for self-reported,
sleep-onset latency and were random ly assigned to one of the fol
lowing conditions: paradoxical intention, progressive relaxation,
stimulus control, or credible placebo. Counterdemand instructions
were employed to com pensate for dem and characteristics.
The results revealed no interaction betw een the severity of sleep-
onset latency and the type of treatment. However, unlike previous
researchers, Lacks and associates (1983) found stimulus control to
be significantly more effective than any other treatment procedure
regardless of the level of severity. Paradoxical intention produced
results that were not significantly different from the placebo con
dition. A s was the case w ith Espie and Lindsay (1985), Lacks and
colleagues (1983) noted that paradoxical intention som etim es
increased sleep-onset latency.
In a replication of the studies of Turner and A scher (1979) and
Lacks, Bertelson, G ans, and K unkel (1983), Espie, Lindsay,
Brooks, H ood, and Turvey (1989) studied a sam ple of 84
physician-referred, chronic insom niacs (mean sleep-onset latency
of 83 m inutes). Subjects were assigned random ly to one of five
conditions: progressive relaxation, stim ulus control, paradoxical
intention, placebo, or no-treatm ent. W hile one therapist con
ducted all of the sessions across treatm ents, ratings of nine ran
dom ly audiotaped sessions revealed no significant differences
across patients and treatments. A s an additional control, counter
dem and instructions were used in the first four w eeks of treat
m ent, followed by four w eeks under positive dem and.
Subjects were evaluated on a variety of self-report variables
including sleep-onset latency (m ean and SD), total sleep duration

(m ean and SD), restedness, and sleep enjoym ent. A n im portant
additional m easure assessed night to n ight variability in sleep, a
factor potentially as im portant in sleep distress as the m ean sleep-
onset latency. The validity of the self-report measures was assessed
using the Sleep A ssessm ent Device (SAD), w hich tape records
responses to a fixed interval cue tone (Kelly & L ichstein , 1980).
Twenty su bjects were thus evaluated for a com bined total of 110
nights. C om parisons of the self-report and objective m easures
produced highly significant correlations in m easures of sleep-
onset latency and total sleep time (r = .85 and r = .87, respectively).
W hile su bjects did overestim ate sleep-onset latency by a 17 per
cent m argin, the high correlations indicated that they did so
consistently.
The results showed that only active treatm ents were associated
with significant improvement, and that the nature of the improve
m ents varied. Stim ulus control very quickly reduced sleep-onset
latency, w hereas som e su bjects in the paradoxical intention con
dition experienced an exacerbation of sleep-onset latency. Stimulus
control su bjects w ho com pleted therapy experienced a reduced
sleep-onset latency of 62 percent (around 32 m inutes), as well as
a more predictable pattern of sleep-onset latency. However, per
ceived quality of sleep did not increase during stim ulus control
treatm ent.
By the end o f treatm ent, the paradoxical intention group
achieved a final sleep-onset latency at levels similar to the stimulus
control group, but the treatment progress was less straightforward.
For exam ple, five of the 15 su bjects exposed to paradoxical in ten
tion instructions experienced increases in sleep-onset latency dur
ing the first week of treatment. O n the other hand, night-to-night
variability in sleep-onset latency was significantly reduced in the
paradoxical intention group prior to reductions in sleep-onset
latency. In other words, the paradoxical intention su bjects expe
rienced increased stability in sleep-onset latency prior to their aver
age sleep-onset latency decreases. C orresponding to this finding
are the subjects' reports of greater restedness after sleep following
com pletion of the treatm ent program .
In contrast to stim ulus control and paradoxical intention, relax
ation achieved its prim ary results w ith significant im provem ents
in the subjects' perception of sleep quality and daytime m easures
of concentration and general w ell-being. These im provem ents
occurred in spite of relatively lim ited effects upon sleep-onset
latency. Total sleep duration increased by about 40 m inutes, as it
did w ith the other treatm ents.
224 CHAPTER 8
The authors concluded that stimulus control appears to be effec

tive in habit restructuring, although patients will not necessarily
be more contented w ith their sleep, and relaxation may be used
when qualitative im provem ents are more im portant than qu anti
tative. Paradoxical intention seem s to reduce internight variability
prior to decreasing latency. Thus, paradoxical intention may pro
duce im provem ents in both sleep-onset latency and quality of
sleep, although the authors caution that som e patients seem to
show tem porary exacerbation of sleep problem s.
In summary, the three previously discussed studies exam ining
the m ost popular behavioral treatm ents for sleep-onset insom nia
(paradoxical intention, progressive relaxation, and stim ulus con
trol) produced different results. Turner and A scher (1979) found
that these three treatm ents were equally effective and superior to
no treatm ent or placebo treatm ent conditions. Lacks, Bertelson,
G ans, and K unkel (1983) dem onstrated superior results for stim
ulus control regardless of severity, with paradoxical intention pro
viding an outcom e that was not significantly different from a
placebo condition. Espie, Lindsay, Brooks, H ood, and Turvey
(1989) found that stim ulus control very quickly and effectively
reduced sleep-onset latency, while paradoxical intention achieved
comparable reductions in sleep-onset latency over a longer period
of time.
A vast m ajority of experim ental studies on the effectiveness of
paradoxical intention has employed a counterdem and procedure
to control for the effects of experim ental dem ands (e.g., Espie &
Lindsay, 1985; Espie et al., 1989; Lacks et al., 1983; Ladouceur
& G ros-Louis, 1986; O tt, Levine, & Ascher, 1983; Relinger &
Bornstein, 1979). However, the counterdem and instruction— not
to expect any results in a specified tim e period— can be viewed
as a paradoxical treatm ent in its own right. Viewed in this way,
the treatm ent procedures employed by Lacks et al. (1983) and by
Espie et al. (1989) may be more correctly termed: stim ulus control
plus paradoxical intention, progressive relaxation plus paradoxical
intention, and paradoxical intention alone. This, in turn, may
account for the relatively quick decrease in sleep-onset latency
established in su bjects assigned to the stim ulus control groups,
in contrast to the slower response of individuals in the paradoxical
intention groups of the Lacks et al. (1983) and Espie et al. (1989)
studies. It is, therefore, not surprising that stim ulus control pro
cedures used along with paradoxical instructions should yield bet
ter results than paradoxical intention instructions alone. The m ost
obvious reason for this is that as sleep disturbances are m ultiply
determ ined and m orphologically varied, the greater is the diver
sity of treatm ent com ponents and the higher is the probability of
a successful outcome. This suggests that counterdem and instruc
tions confound treatm ents in paradoxical intention studies. The
continued use of this m ethod of control in such studies should be
accom panied by design m odifications that would serve to rectify
the confusion.
D espite the discrepancies found am ong these studies, the effi
cacy of paradoxical intention as a treatm ent m ethod for prim ary
sleep-onset insom nia appears to be reasonably well supported in
both the single-case (A scher & Efran, 1978; Relinger et al., 1978;
Relinger & Bornstein, 1979) and group experim ental design
(Ascher & Turner, 1979a; Espie et al., 1989; Turner & Ascher, 1979,
1982). The studies further suggest that clinically significant
improvements in sleep complaints can be produced with relatively
m inim al expenditures of therapists' time. The results of these in i
tial reports (although flawed to som e extent) are com parable to
those obtained w ith other established behavioral techniques and
indicate that paradoxical intention, progressive relaxation, and
stim ulus control are all useful procedures in the am elioration of
sleep d isturbances— all producing results superior to those of
credible placebo and no treatm ent control conditions.
Noting that different paradoxical intention instructions have
been utilized under a variety o f circum stances, several investiga
tors have evaluated diverse m ethods for presenting paradoxical
instructions. In the first, "assessm ent" or traditional explanation,
su bjects are instructed to rem ain awake for as long as possible in
order to becom e aware of cognitions to be used in an upcom ing
desensitization procedure. In this approach, individuals are kept
blind to the true nature of the intervention to which they are being
exposed. In the second approach, termed the "veridical" explana
tion, su bjects are inform ed of the relationship betw een perform
ance anxiety and insom nia and are provided w ith relatively
straightforw ard paradoxical instructions aim ed at neutralizing
perform ance anxiety. In the "auth oritarian " explanation (Fisher,
Nietzel, & Lowery, 1985), the im portance of following instructions
in order to obtain the desired results is em phasized, but su bjects
are provided w ith no specific rationale.
Fisher, Nietzel, and Lowery (1985), in an analogue study, inves
tigated the credibility o f these explanations on the effectiveness
of paradoxical intention w ith insom niacs. College students were
presented with veridical, bogus, or authoritarian instructions and
asked to rate their credibility. The authors also included a progres
sive relaxation rationale for insom nia treatm ent. Respondents
found the explanations for paradoxical therapy to be equally cred
226 CHAPTER 8
ible. Overall, however, the students preferred the relaxation ratio

nale (70 percent) to both the paradoxical explanations (30 percent)
and the authoritative directions (0 percent); the latter finding is
not surprising since the authoritative condition provides no ratio
nale. The preference of the progressive relaxation rationale over
that of paradoxical intention is in contrast to the previous inves
tigations of Turner and A scher (1979) and A scher and Turner
(1979a) in w hich rationales tended to be equally acceptable. Fisher,
N ietzel, and Lowery (1985) attribute this difference to the com -
m onsensical nature of progressive relaxation as a m eans of treat
ing insom nia, in contrast to the paradoxical procedures w hich are
counterintuitive in nature. O f course, the fact that the su bjects of
Turner and A scher (1979) and A scher and Turner (1979a) were
treated in a clinical setting for severely disturbed sleep, in contrast
to the use of college students in the analogue situation employed
by Fisher, Nietzel, and Lowery (1985), m ight also account for some
differences.
The authors also investigated the role of im pulsivity as a per
sonality trait w hen assigning preference to these rationales. The
authoritarian rationale received its highest acceptance from sub
jects rated high on impulsivity, w hom the authors speculated
may have been w illing to suspend ju dgm ent or may have been
unconcerned about the lack of a logical explanation. Subjects high
on control, however, preferred the assessm ent rationale over the
veridical explanation. Fisher, N ietzel, and Lowery (1985) argued
that to highly controlled subjects, a veridical instruction to relin
quish control m ust seem alien. Such people are likely to assum e
that m aintaining tight control is the best way to achieve their
goals.
In another investigation of this issue, Ascher and Turner (1979b)
com pared the effectiveness of veridical explanations versus an
"assessm ent" rationale. Forty subjects were randomly assigned to
one of four groups: paradoxical intention w ith either veridical or
"assessm ent" rationale, attention control, or a waiting list control.
A variety of self-report outcom e m easures were employed.
In the veridical condition, subjects were asked to rem ain awake
for as long as possible once in bed, and were inform ed of the
hypothesized relationship betw een perform ance anxiety and
sleeping difficulties (as in A scher and Turner, 1979a). In the
"assessm en t" condition, it was suggested that clients rem ain
awake in order to becom e aware of anxiety-producing thoughts.
Placebo subjects received quasi-desensitization by pairing 18 hier
archical bedtim e scenes with neutral images. Subsequent to treat
m ent, su b jects com pleted a questionnaire designed to tap the
credibility of the procedures and the quality of the therapeutic

relationship.
Su bjects exposed to the veridical procedure exhibited signifi
cantly greater improvement than those in the "assessm en t" group
or in either of the control groups. Confounding factors such as dif
ferential treatment credibility, therapist differences, and statistical
probability pyram iding were ruled out. However, other possible
confounds do im pact on these findings. Since the therapists were
also the authors, a variety of sources o f bias may have been incor
porated. In addition, the "assessm ent" rationale instructions indi
cated that desensitization would begin after the disturbing
thoughts were collected. Consequently, there is the possibility that
these su bjects assum ed that "th erap y " had not beg u n , thereby
serving to dilute the potential effects of the "assessm en t" proce
dure. O f course, the m ost ethical m ethod of adm inistration is full
disclosure of the nature and pu rpose of paradoxical procedures,
w hich apparently also produces better outcom es.
Agoraphobia
It is possible that as m uch as 5 .8 percent of the population of
the United States could be classified as agoraphobic (Barlow, 1988).
A s a result of this prevalence, great effort has been expended in
developing and testing treatm ent m ethods for this clinical popu
lation. Among these is paradoxical intention, the efficacy of w hich
is supported by data from a variety of sources ranging from
uncontrolled case studies (e.g., Frankl, 1955, 1975, 1985; G erz,
1966) to controlled experim ents.
A scher (1981) evaluated the efficacy of paradoxical intention in
alleviating the travel restriction experienced by agoraphobic indi
viduals. Ten agoraphobic clients (nine fem ales, one m ale) were
randomly assigned to two treatment conditions. In the first, clients
were given graded in vivo exposure for six sessions follow ing a
baseline phase, after w hich they received paradoxical intention
instructions. In the second, clients were exposed to six sessions
of paradoxical intention following the baseline phase. Statistical
analysis indicated that individuals reduced travel restriction to a
significantly greater extent in the context of paradoxical intention
as com pared w ith graded exposure.
In a subsequent study, A scher (1983) hypothesized that agora
phobic clients m ight experience more rapid improvement for travel
restrictions if paradoxical instructions were employed earlier in the
course of therapy and if the procedure was to be com plem ented
by supportive, ancillary techniques. Three procedures were com
228 CHAPTER 8
pared: paradoxical intention, "en h a n ced " paradoxical intention,

and the "en h an cem en t" com ponent alone (consisting of various
behavioral and cognitive-behavioral techniques). Fifteen agora
phobic clients were randomly assigned to these three groups. The
results revealed that both paradoxical intention groups signifi
cantly reduced travel restrictions more rapidly than the en h an ce
m ent group alone. A lso, individuals in the en han ced paradoxical
group seem ed to use the procedure earlier in the course of ther
apy. However, there were no other differences betw een these two
groups.
These data are com prom ised by various aspects of the design,
procedure, and m ethod of su bject selection. However, they pro
vide the first tentative support for the efficacy of paradoxical inten
tion in reducing travel restrictions in agoraphobic individuals and
suggest a m ethod by w hich this treatm ent procedure m ight be
employed by clients earlier in the course of therapy.
M a v issa k a lia n , M ich elso n , G reen w ald , K o rn b lith , and
Greenwald (1983) com pared paradoxical intention to self
statem ent training in treating agoraphobics. Twenty-six clients
were random ly assigned to one of the two procedures, each of
w hich was conducted in groups of four to five over the course of
12 ninety-m inute sessions. Su bjects in the paradoxical intention
group improved significantly more than did th ose w ho received
self-statement training. However, at a six-month follow-up assess
m ent, the self-statem ent training group exhibited a level of
improvement equivalent to that of the paradoxical intention group.
In an interesting series of experiments, M ichelson and his asso
ciates systematically investigated several popular strategies for the
general treatment of agoraphobia. In the first of these, Michelson,
M avissakalian, and M archione (1985) com pared the relative effi
cacy of paradoxical intention (PI), graduated exposure (GE), and
progressive relaxation training (RT). Thirty-nine severe and
chronic agoraphobics with panic attacks were random ly assigned
to one o f these groups. Twelve weekly tw o-hour group sessions
were conducted by experienced therapists. A nalyses revealed that
su bjects in all conditions showed significant im provem ents on a
variety of m easures. However, GE and RT were associated w ith
more rapid effects than PI. In fact, su bjects w ho received PI actu
ally experienced increases in physiological reactivity during treat
m ent. Not until a three-m onth follow-up period were they
equivalent in their physiological functioning. T his finding su g
gests that PI may require greater tim e for su bjects to effectively
employ this strategy.
Using the sam e data set, Michelson, M avissakalian, Marchione,
Dancu, and Greenwald (1986) investigated the role of self-directed

in vivo exposure (SDE) practice in the treatments mentioned in the
previous study. SDE involves a conscious decision by the client to
engage in phobic encounters in order to cause therapeutic change.
Overall, RT was the m ost effective strategy in en han cin g practice
time, while GE followed closely. PI was least effective in increasing
SDE practice. These results suggest the im portance of developing
treatm ents that significantly affect SDE.
In a second reanalysis of these data, M ichelson (1986) focused
on the interaction betw een client and treatm ent variables. It was
determ ined that those clients w hose treatm ents were consonant
w ith their sym ptom profiles showed higher rates o f im provem ent
than those w hose treatm ents were dissonant. Thus, clients w ith
a sym ptom profile that was dom inated by cognitive factors did
best w ith PI. T h ose w ith a sym ptom profile largely reflecting
behavior difficulties exhibited greatest im provem ent w ith GE.
Finally, clients w hose profiles indicated that physiological sym p
toms were m ost salient benefitted m ost from RT. This study, then,
underscored the im portance of m atching a subject's response pro
file w ith a consonant treatm ent modality.
In a final project, M ichelson, M avissakalian, and M archione
(1988) provided a more com prehensive outcom e investigation on
a large group of clients. In this study, 88 agoraphobics w ith panic
attacks were random ly assigned to either PI, GE, or RT. A nalyses
showed significant improvement across all dom ains and treat
m ents w ith few betw een-group differences. The three interven
tions were found to be equally effective at posttreatm ent and at
a three-month follow-up. These authors concluded that in the treat
ment of agoraphobic clients, individual procedures m ight be more
likely to produce incomplete results, whereas the appropriate com
bination of techniques would serve to provide a more satisfactory
outcom e.
Obsessive Disorders
Solyom, G arza-Perez, Ledw idge, and Solyom (1972) employed
a quasi-experim ental design in the treatm ent of 10 individuals
w ith obsessional concerns. A fter a pretreatm ent assessm ent con
sisting of several questionnaires and a psychiatric interview, two
obsessions were selected for each su bject. O ne served as the con
trol obsession, w hile the other served as the target obsession.
Subjects were instructed in paradoxical intention and told to asso
ciate this procedure w ith the target obsession. The results of this
study were mixed and, of course, w ithout the n ecessary control
230 CHAPTER 8
procedures were difficult to assess. However, the fact that five of

the 10 individuals reported some reduction in the target obsession
supported the m yriad of previous case studies reporting such a
relationship.
M ilan and Kolko (1982) evaluated the effect o f paradoxical inten
tion on the obsessional ruminations of a 33-year-old female client.
The client sought treatm ent to help deal w ith anxiety regarding
her perceived high incidence of m alodorous flatulence. Previous
m edical evaluation had ruled out physical problem s or dysfunc
tions, and she had had prior unsuccessfu l courses of phar
macotherapy, dietary m anagem ent, and psychotherapy.
The authors used an A BC design and collected follow-up data.
Throughout the baseline and treatm ent phases, the client m oni
tored the frequency and intensity of her perceived flatulence on
an hourly basis. The first phase of treatm ent involved the correc
tion of m isconceptions through the use of scientific evidence. The
second phase involved the use of paradoxical intention. Here, the
therapist "acknowledged the imprecision of m odem science" and
instructed the client to enhance flatulence during each episode in
order to expel her flatus completely. The latter procedure produced
an im m ediate reduction in the client's perceived frequency and
intensity of flatulence episodes by approximately 75 percent. Data
collected during two weeks of self-monitoring conducted one year
after treatment revealed the client's further improvement, suggest
ing an overall reduction of nearly 95 percent.
Unfortunately, the design employed in this study incorporated
several m ethodological shortcom ings, including serial treatm ent
effects and several threats to internal validity (H ersen & Barlow,
1976). In the absence of more powerful single-case experim ental
designs (e.g., those involving a return to baseline phase) or group
research, causal inferences are speculative at best. Thus, the evi
dence for the efficacy of paradoxical intention in the treatm ent of
obsessions is weak. The studies cited here, however, do suggest
the possibility of the efficacy of paradoxical intention w ith this
behavioral difficulty; further investigation, using more sop h isti
cated designs and larger groups of subjects, seem s w arranted.
Disorders of Elimination
Both urination and defecation require a complex interaction
betw een social control and autonom ic nervous system activity.
O ne reason for a disruption of this interplay may be one's contin
uous, zealous attem pts to control. This effortful behavior and the
associated performance anxiety m ight serve to inhibit the natural
occurrence of the appropriate elim inative reflexes. That is, trying

too hard can paradoxically hinder successful elimination behavior
(Jacob & M oore, 1984).
Functional U rinary Frequency. Tim m s (1985) treated a 13-year-old

boy w ho com plained of u rinary frequency associated w ith school
attendance. The discom fort occurred in the m orning, ju st before
leaving home for school. This problem becam e so profound that
eventually the client did not bother to m ake an attem pt to go to
school. Paradoxical intention was the m ajor procedure employed in
the treatment program . Pretreatment recordings showed that the uri
nary frequency occurred on each school day. Posttreatment data indi
cated that the paradoxical instruction was carried out and complaints
of urinary frequency rapidly extinguished. Follow-up at 6 m onths
showed no recurrence of the frequency problem.
Functional U rinary Retention. Functional u rinary retention is

the condition in w hich individuals cannot urinate in certain bath
rooms but function normally in others. People with urinary reten
tion have been treated behaviorally using a wide variety of
behavioral techniques, yet because of the difficulty of the problem
results are inconsistent. For this reason, additional strategies are
of interest to clinicians who seek to increase probability of success
by approaching such clients w ith expanded behavioral reper
toires. Recently, paradoxical intention has been shown to be effec
tive w ith som e cases of psychogenic urinary retention.
Ascher (1979) conducted a study in w hich 5 clients were treated
in the context of an ABC design consisting of a two-week baseline
phase followed by an eight-w eek conventional behavior therapy
program . Following this experience, these individuals were dis
satisfied w ith their level of progress, and a series of sessions was
instituted, during w hich paradoxical intention was employed.
Clients were instructed to enter a discom forting bathroom and to
engage in all the norm al activities associated w ith urinating, but
to actively inhibit the passage of urine. Then they were to con
clude their visit and leave the bathroom. The rationale for this tech
nique, involving the deleterious effects of perform ance anxiety,
was presented to the clients. Subjective ratings of discom fort func
tioned as the dependent variable. Results indicated that greater
improvement occurred following the institution of paradoxical
intention as com pared w ith the trial of the more conventional
procedure.
Functional Encopresis. Encopresis is characterized by inappro

232 CHAPTER 8
priate fecal soiling. A num ber o f studies have been conducted to

test the efficacy of paradoxical intention in treating this disorder
(e.g., M argolies & G ilstein, 1 983-1984; Propp, 1985). However,
only that of B o m stein , Sturm , Retzlaff, Kirby, and C hong (1981)
incorporated an acceptably controlled (ABAB) design.
In their investigation, B o m stein , Sturm , Retzlaff, Kirby, and
C hong (1981) used paradoxical procedures in the treatm ent of a
nine-year-old boy who suffered from infrequent bowel move
m ents, daily soiling, and considerable anxiety related to toileting
behavior. Parental records of appropriate bowel movements and
soiling episodes functioned as the dependent measure. Following
a three-w eek baseline period, the boy was required to enter the
bathroom each hour and sit on the toilet for five m inutes. He was
to act as though he was going to have a bowel movement, although
he was instructed not to have a bowel movement. T his procedure
was intended to decrease anxiety associated w ith toileting.
Following 3 weeks of treatment, a return to baseline was instituted
for two w eeks. Subsequently, paradoxical procedures were rein
stated and eventually faded.
The results supported the efficacy of paradoxical intention in
the treatm ent of encopresis. D uring the initial baseline phase, the
child had an average of 6 .7 soiling incidents and 0 .7 appropriate
bowel movements per week. Paradoxical intention resulted in a
decrease to 0.0 soiling incidents and an increase of 4.3 appropriate
bowel movements per week. W hen treatment was withdrawn, the
dependent m easures returned to near-baseline levels (5.0 soiling
incidents and 0.5 appropriate bowel m ovements per week). W ith
the reinstatem ent of paradoxical procedures, soilings decreased
to 0.0 and appropriate bowel m ovements increased to 4 .8 per
week. Treatment effects were m aintained through a one-year
follow-up.
Miscellaneous Clinical Complaints

In addition to the previously discussed behavioral com plaints,
paradoxical intention has been associated w ith a vast array of dis
orders and adjustm ent difficulties. These include, but are not
restricted to, anorexia nervosa, erythrophobia, panic attacks, gen
eralized anxiety, psychogenic nausea, social phobia, depression,
and procrastination. Although the research in each of these areas
is extremely limited, typically involving uncontrolled single cases,
they do illustrate the clinical potential of paradoxical intention. In
this section, studies of particular interest will be highlighted.
H su and Lieberm an (1982) used a short-term (six sessions)
paradoxical intervention with eight chronic anorexic clients. These

individuals had previously been unsuccessfu l in m aintaining
treatm ent-related w eight gains. W hen rehospitalized, they were
placed on a program that incorporated paradoxical intention. It
was suggested that m aintaining anorexia nervosa was, in ways
specific to each client, better than attem pting to defeat it, esp e
cially since previous attem pts had resulted in only tem porary
rem ission. At two to four years after treatm ent, h alf of the clients
reported normal w eight (w ithin 15 percent of average) and only
one exhibited very low w eight (75 percent o f average); one
rem ained in treatm ent at the time.
Lamontagne (1978) reported the results of paradoxical intention
in the treatm ent of erythrophobia, the fear of blushing. The client
was a 25-year-old male who had a 12-year history of the com
plaint. His social, educational, and occupational functioning were
severely hindered by this concern. Baseline data, collected in the
month prior to treatment, showed 28 severe episodes of blushing,
each w ith an average duration of 13 m inutes. D uring the m onth
long treatment program, it was suggested to the client that he stop
fighting his sym ptom s and avoiding places w here they tended to
occur. Further, he was to attem pt to blush as much as possible for
three, 10-m inute periods every day at home and in situations
where he felt anxious. Results indicated that during treatm ent
there were only three episodes of blushing. These treatm ent
effects were m aintained at the end of an eight-m onth follow-up
interval in w hich the client typically reported fewer than two
severe episodes of blushing per m onth.
In a related study, Boeringa (1983), em ploying a paradigm sim
ilar to that of Lam ontagne (1978), also described the treatm ent of
a single case of erythrophobia. The patient was a 27-year-old male
for whom blushing had becom e a significant problem . The diffi
culty had started when he accepted a promotion to a position that
involved interaction w ith m any people each day. Blushing fre
quency and intensity increased to the extent that it began to dis
turb the client and becam e significantly noticeable to co-w orkers.
Boeringa's paradoxical intervention resulted in sign ifican t
improvement after seven sessions.
The use of paradoxical intention in the treatm ent of panic
attacks was illustrated by Dattilio (1987). He provided a brief
case example and discussion of the nature and use of the tech
nique. Ingrid, a 32-year-old m arried w om an, had experienced
over 200 panic attacks w ithin 18 m onths before being referred
for psychological treatment. The first session involved a detailed
behavioral analysis to gather inform ation for treatm ent. A
234 CHAPTER 8
detailed hierarchy was developed in the next two sessions to

indicate the anxiety-producing situations that were associated
with her panic attacks. Ingrid was then instructed to expose
herself to the least feared situations w hile allow ing herself to
becom e anxious w ithout trying to interfere w ith it. W hen she
tried to exaggerate her attacks, she was unable to do so. W ith
this m ethod, eventually her anxiety was greatly reduced
through the other hierarchical levels.
The effects of paradoxical intention on generalized anxiety were
investigated by Last, Barlow, and O 'Brien (1983). This study u ti
lized an alternating treatm ent design in w hich a single client was
exposed to different treatment procedures on successive days. The
two treatm ent procedures that were com pared were paradoxical
intention and M eichenbaum 's (1977) self-statem ent training. The
client was exposed to four sessions of each technique w hile par
ticipating in in vivo exposure. In addition, he was asked to employ
a particular day's technique throughout the week. Although treat
m ent was conducted in a group format w ith agoraphobic group
m em bers, the results focused only on the one client w ith gener
alized anxiety disorder. Results showed a m arked decline in fear
hierarchy m easures, w hich was m aintained throughout the one-
year, follow-up period. Although there appeared to be no signif
icant difference in efficacy betw een the two treatment procedures,
the client reported that paradoxical intention caused her greater
discom fort than the alternate strategy.
Analogue Research
Clinical analogue research presents a model representative of
significant aspects of the actual clinical situation, in order to pro
vide an em pirical framework for testing process and outcom e
hypotheses. A nalogue studies are popular because they require
less cost, are generally more quickly and easily executed, and typ
ically involve su bjects w ho are more accessible and plentiful,
w hen com pared w ith actual clinical trials. The m ajor disadvan
tage is that because they are only representative of real clinical
conditions, the data of analogue studies are not entirely gener-
alizable to the clinical setting and therefore often lack clinical util
ity. For exam ple, m any clinical analogue experim ents use
volunteer college students as subjects, who report som e degree
of concern regarding the advertised target sym ptom . The focal
problem has a high probability of occurrence in a college setting
and is less likely to have complicating clinical correlates (e.g., pro
crastination, test anxiety). O ften these su bjects receive rem uner
ation in term s of course credit or a sm all fee. If this situation is

com pared to one in w hich an individual goes to a clinical p sy
chologist, in a clinical setting, and pays a substantial fee for assist
ance w ith som e sign ifican t behavioral problem , one can
im m ediately im agine the difficulty of developing functional rela
tionships based on the former analogue research that would be
useful for the latter clinical situation.
N otwithstanding this note of caution, clinical analogue research
can provide helpful direction in this prelim inary stage in the inves
tigation of the efficacy of therapeutic paradox. The following sec
tion is com posed of a survey of the more notable contributions to
this category of inquiry.
Several studies have evaluated the efficacy of paradoxical inten
tion in the treatm ent of college students com plaining of procras
tination. For exam ple, Wright and Strong (1982) contrasted
paradoxical and choice directives. Thirty students were randomly
assigned to one of three experim ental conditions: a paradoxical
condition in w hich the su bjects were instructed to continue pro
crastinating exactly as they had been, a choice procedure in which
they were instructed to choose to continue som e of their procras
tination behaviors, or a no-intervention control group. Both treat
m ent grou p s ach iev ed eq u a l and m ark ed red u ctio n s in
procrastination com pared to the control group, thus show ing that
directives to stay the sam e were as effective as directives to
change. However, since both active treatm ents caused equivalent
am ounts of ch an ge the possibility rem ains that, in the analogue
situation em ployed, any treatm ent procedure generates sim ilar
results. The effects of an attention placebo would serve as a
response to this criticism. It is interesting that although both con
ditions could be characterized as paradoxical to some extent, sub
je cts in the paradoxical condition attributed their change to
spontaneous, nonvolitional causes, whereas subjects in the choice
condition attributed their im provem ent to volitional causes.
Lopez and W ambach (1982) employed 32 students w ho com
plained of procrastination and randomly assigned them to a para
doxical, a self-control, or a no-interview control condition. Subjects
in the paradoxical condition were provided w ith a "b o g u s"
rationale— to deliberately bring about their procrastination for the
purposes of observation and control. Su bjects in the self-control
condition were told that procrastination is a learned habit, allevi
ated by the development of new behaviors incompatible w ith pro
crastination. They were then provided w ith strategies com patible
w ith this hypothesis. Both treatm ent groups were seen for two,
30-m inute sessions spaced one w eek apart. A s was the case with
236 CHAPTER 8
Wright and Strong (1982), the results suggest that both active-
treatm ent groups exhibited greater im provem ents on self-report
m easures of procrastination frequency and controllability w hen
com pared w ith the no-treatm ent control. However, the nature of
these im provem ents varied greatly betw een treatm ents. In the
paradoxical control condition, five o f the 10 su b jects experienced
an exacerbation of their procrastination behaviors before show ing
sharp decreases in procrastination. Progress in the self-control
condition was of a more even, steady nature, spread out over the
four-week experim ental period. W ith regards to controllability,
su bjects in the self-control situation reported view ing their pro
crastination as controllable by direct effort. S u bjects in the para
doxical condition, despite reporting im provem ents in their
procrastination, did not view their problem as significantly more
controllable. The findings of Wright and Strong (1982) corroborate
this relationship to som e extent. A gain, treatm ent gains due to
therapists' attention cannot be ruled out due to the lack of an
attention control.
Shoham -Salom on, Avner, and N eem an (1989) studied the role
of "reactan ce" in change induced by paradoxical intention and
self-control interventions. The authors defined reactance as "th e
state of mind aroused by a threat to one's perceived legitimate free
dom, m otivating the individual to restore the thw arted freedom "
(p. 590). Their experim ental design m irrors that of Lopez and
Wambach (1982), whereby 58 undergraduates com plaining of pro
crastination were random ly assigned to a paradoxical condition,
a self-control condition, or a no-treatm ent control condition.
Reactance was experimentally m anipulated by having each sub
je ct read two generic descriptions of treatm ents: an attractive
description and a less attractive, but not appalling description.
Both descriptions were appropriate for either treatment condition.
Subjects were then requested to pick the treatment they would like
based on these descriptions. Since everyone picked the more
attractive description, high-reactance subjects were created by tell
ing one-third of the su bjects that they were being assigned to the
less desired treatm ent, even though they were initially told that
they would be able to choose their desired therapy. A no-reactance
condition was also created by assigning clients to a treatm ent
w ithout indicating that a choice was possible.
Therapy consisted of two, 30-m inute sessions spaced one week
apart, w ith an advanced graduate student therapist. Data con
sisted of study logs com pleted at home and subm itted at the
begin ning of each session and at one- and four-week follow-up
intervals. During the follow-up session, subjects also filled out the
Perceived Self-Efficacy Scale (PSE)— a scale m easuring how much

clients think they are able to do.
The m ain results of this study were consistent w ith those of
W right and Strong (1982) and Lopez and W am bach (1982), show
ing that, on average, both treatment conditions increased students'
self-rated, effective study tim e by the sam e am ount. However,
w ithin each treatment condition, the nature of the results differed
for high- and low-reactance subjects. In the paradoxical condition,
high-reactance subjects showed significant improvement in effec
tive study tim e, while low-reactance su bjects showed little
improvement. O n the other hand, for the self-control condition,
high- and low-reactance su bjects showed similar, m oderate
improvement.
Regarding perceived self-efficacy, subjects in the high-reactance
paradoxical condition showed m inim al increases, w hereas low-
reactance su bjects in the sam e condition showed m uch larger
increases. T hose in the self-control, high-reactance condition
showed som ew hat larger increases in perceived self-efficacy than
did low-reactance su bjects. The authors concluded that the para
doxical condition incorporated an"either/or" pattern not found in
the self-control condition. That is, w ith the paradoxical interven
tion, individuals experienced more reactance and cam e to study
more effectively, or experienced less reactance and showed
increases in perceived self-efficacy in controlling their procrasti
nation. Thus, in keeping w ith Watzlawick, Beavin, and Jackson's
(1967) contention that "you are changed if you do and changed
if you d o n 't."
A s w ith the previously m entioned procrastination studies, the
investigation of Shoham-Salom on, Avner, and N eem an (1989) also
fails to include, am ong other things, an attention placebo w hich
limits interpretability. However, the study is im portant because of
its focus on a significant, m ediating factor im plicated in the suc
cess of paradoxical intention. A gain, the data are useful only for
the formulation of hypotheses since the authors' m anipulation of
reactance occurs in the absence of any reliable, objective m easure
of this construct. Thus, there is no way to determ ine w hether they
were successful in generating and m anipulating the reactance
variable.
D epression am ong college students represents another area in
which analogue research into paradoxical interventions is concen
trated. For exam ple, Beck and Strong (1982) com pared two types
of interpretations and a no-interview condition w ith m oderately
depressed, volunteer college students. In the first condition, stu
dents' depressive sym ptom s were interpreted as connoting neg
238 CHAPTER 8
ative and undesirable behaviors such as irrational thinking and

attem pts to m anipulate others. In the second condition, the para
doxical approach, depressive sym ptom s were associated with pos
itive and desirable personal characteristics, such as sensitivity or
self-sacrifice. Although both groups had improved at posttesting,
su bjects who had received the negative interpretations relapsed
quickly, whereas those in the paradoxical condition m aintained
their gains.
Feldm an, Strong, and D anser (1982) also studied m oderately
depressed college students by adm inistering consistent or incon
sistent paradoxical or nonparadoxical interpretations and direc
tives using a 2 x 2 design. There was also a no-interview control
condition. Through the course of two interview s, six interpreta
tions and two directives were given. Students in the interview con
ditions reduced their depression scores more than those w ho did
not participate in the interview s. T h ose w ho received paradoxical
interpretations or directives, regardless of their consistency,
improved more than those who received nonparadoxical interpre
tations and directives. Consistency, however, resulted in more
favorable im pressions of the counselor.
Kraft, Claibom , and Dowd (1985) employed 47 college students
who reported "negative em otions." Using a 2 x 2 design, subjects
random ly received (1) either paradoxical or nonparadoxical direc
tives and (2) either positive refram ing statements or no refram ing
statem ents. Each su bject participated in two, 30-m inute cou nsel
ing sessions, focusing on the individual's negative em otions.
C hange was m easured by pre/post differences on the Beck
Depression Inventory and on a self-report m ood scale. The study
found that positive refram ing produced greater reduction in neg
ative em otions than no refram ing, although negative em otions
were reduced in all conditions. Neither of the two directive con
ditions (paradoxical and nonparadoxical) evidenced greater
im provem ents than the other. These results suggest that while
directives contribute to client improvement, it does not seem that
paradoxical directives have any advantage over other directives.
The findings are mitigated, however, by the authors' stated uncer
tainty as to the success of the paradoxical m anipulation. Subjects
in this condition were asked to schedule time each day to feel neg
ative em otions "so that you will understand them more clearly."
It is possible that, in som e cases, this turned out sim ply to be a
self-m onitoring task. Thus, while this study would seem to indi
cate that positive refram ing statem ents are an effective m eans of
reducing "negative em otion s," it is not possible to support con
clusions regarding the paradoxical directives.
In another attem pt to account for the m ech anism s of success in

the use of paradoxical therapy, Z od un, G ruszkos, and Strong
(1985) evaluated the use of paradoxical therapy w ith m oderately
depressed college students. Specifically, they assessed the role of
internal, nonvolitional attributions of change (e.g., developmental
m aturation, personal grow th) and therapeutic double bind (e.g.,
diffusing the impact of the double bind in paradox through explicit
identification of the paradoxical nature and intent of the treat
ment). The authors argue that if paradoxical techniques achieve
their effects by instilling nonvolitional, internal change in the cli
ent, then students receiving this type of explanation should evi
dence greater treatm ent gains than those receiving a differing
explanation. Conversely, if paradoxical treatm ents produce their
gains through therapeutic double bind, as in preparation asserted
by Haley (1976), then a clear explanation of the nature and intent
of paradox should attenuate improvement (a veridical explanation).
The depressed students were random ly assigned to one of five
treatm ent conditions: paradoxical interpretations and directives
only, paradox plus developmental rationale, paradox plus veridical
rationale, nonparadoxical interview, or a no-interview control. The
results did not support either hypothesis. Students in the devel
opmental rationale group improved very little. Similarly, provision
of a veridical rationale did not seem to enhance treatm ent effects
as those in this group did as well as those in the paradox-only
group.
A n analogue study of the effects of various rationales on sym p
tom prescription was conducted by Boettcher and Dowd (1988).
Fifty college undergraduates w ho reported perform ance anxiety
and were identified as anxious by the State-Trait A nxiety Inventory
(STAI; Spielberger, Gorsuch, & Lushene, 1970) participated in the
study. The su b jects were asked to increase their anxiety for 15
m inutes, tw ice each day, under one of the four rationale condi
tions: (1) no rationale given; (2) positive refram ing, w hich stressed
the positive characteristics of the anxiety sym ptom s; (3) a descrip
tion of the vicious cycle created w hen direct attem pts are m ade
to decrease anxiety; or (4) a double-bind condition, in w hich su b
jects were told that they may choose to defy the directive and that
the counselor expected them to change whether they followed the
directive or not. A no-attention condition was also included. The
procedures involved four individual interviews designed to differ
only in the nature of the rationale discussed.
The students were evaluated using the STAI and a M ood
Perception Inventory, w hich assessed their beliefs about the
nature, causes, and controllability of their anxiety. The data indi
240 CHAPTER 8
cated that the treatment conditions did not significantly differ from
the control condition. The authors interpreted these results as sug
gesting that therapeutic interventions that direct clients to increase
unwanted sym ptom s are an effective m eans of reducing perform
ance anxiety, regardless of the rationale given. Such a finding con
flicts w ith those of A scher and Turner (1980), w ho dem onstrated
that a rationale defining the directive as a vehicle for change (the
"perform ance an xiety " rationale in both studies) was superior to
a rationale that did not explicitly create this expectation (e.g., the
"aw areness" rationale in the A scher and Turner study, and the
"positive refram ing" rationale in the Boettcher and Dowd study).
T his discrepancy is possibly attributable to the differences
betw een analogue (Boettcher & Dowd, 1988) and clinical studies
(A scher & Turner, 1980) and to the different clinical diagnoses.
Robbins and M ilburn (1990) have also attem pted to account for
the effects of paradoxical interventions. These researchers inves
tigated reattribution, preparatory, reactance, and self-regulatory
reorientation models of paradoxically induced change. A reat
tribution model states that replacing dysfunctional attributions
with more adaptive ones (e.g., a refram ing) results in problem
relief as a result of decreased concern and worry. A preparatory
model postulates that paradoxical procedures serve to prepare the
recipients for arousal, thereby norm alizing the anxiety. The reac
tance model has been discussed in relation to Shoham -Salom on,
Avner, and Neeman (1989). In essence, it states that subjects resist
paradoxical techniques in an attem pt to assert behavioral freedom,
thereby causing abandonm ent of the target behavior. The final
m odel, self-regulatory reorientation, is said to act by disrupting
the client's use of m aladies arousal reduction strategies, thereby
disrupting the perform ance anxiety cycle.
In the first of two experim ents, the authors random ly assigned
122 su bjects reporting high and low test anxiety to one of four
experim ental conditions in a 2 x 2 matrix: positive or neutral
expectations w ith paradoxical or task-focusing instructions. The
prim ary dependent m easure was the num ber of correctly solved
anagram s in an analogue testing situation. A nalysis of the data
indicated that of the subjects who originally reported high anxiety,
those in the paradoxical condition reported lower levels of anxiety
and evidenced better performance than those subjects in the task-
focusing condition, regardless o f their expectancy of success. A n
analysis of subjects' attributions did not support a reattribution
explanation. The authors interpreted that these findings sup
ported a self-regulatory reorientation model rather than a reat
tribution explanation.
In the second experim ent, Robbins and M ilburn (1990) inves

tigated the role of preparatory and reactance m odels as explana
tions for paradoxical treatment success. The authors hypothesized
that preparatory inform ation would actually serve to direct stu
dents' attention to their arousal, thereby increasing anxiety and
reducing perform ance. They further hypothesized that an y reac
tance on the part of the subjects would cause them to give negative
evaluations of the experim enters on several bipolar ratings. The
results showed that for high, test-anxious subjects, preparatory
information did not decrease their performance; however, low, test-
anxious subjects did exhibit decrements under this condition. The
authors also found that high, test-anxious su bjects did not tend
to rate the experim enter more negatively following paradoxical
instructions, thereby failing to support the reactance model.
Overall, Robbins and M ilburn (1990) report little evidence for
preparatory, reactance, and attributional models of paradoxical
intention. In contrast, Shoham -Salom on, Avner, and N eem an
(1989) found reactance to be a significant m ediator in the results
obtained in paradoxical intention therapy. To some extent, the dif
ference may be due to the relatively arbitrary m anner in w hich
this construct has been m easured. A more reliable and objective
m ethod for assessin g this construct would seem to be a prerequi
site for additional investigation of this topic.
Paradoxical intention has also been investigated as a strategy for
pain m anagem ent. Gilligan, Ascher, Wolper, and Bochachevsky
(1984) studied 32 su bjects in a cold-pressor task. The su bjects
were random ly assigned to one of three treatm ent conditions
(paradoxical intention, rational self-statements, or self-observation)
or an expectancy control group. Paradoxical intention instructions
proposed that fear and anxiety would be reduced by "going w ith"
the discom fort rather than trying to fight it. Rational, self
statement instructions em phasized selection and focus on the pos
itive aspects of the experience. Self-observation su bjects were
inform ed that detailed analysis of the sensory aspects of the pro
cedure, excluding the discomfort (e.g., wetness, coldness, num b
ness), would decrease the discomfort and increase their tolerance.
Finally, it was suggested to expectancy control su bjects that the
second, comparative, cold-pressor experience would be less pain
ful than the first. Pain thresholds, tolerances, and discom fort rat
ings were gathered for each subject.
The data indicated that only the self-observation condition pro
duced significantly higher pain tolerance scores than the expect
ancy control group. In line w ith Blitz and D innerstein (1971), the
authors hypothesized that self-observation appears to be effective
242 CHAPTER 8
because it redirects the focus of attention away from the painful

aspects of the experience toward more neutral com ponents, thus
providing an effective strategy for enhancing tolerance. From this
perspective, the other conditions (paradoxical intention, rational
self-statem ents) are less helpful as they served to direct attention
toward painful aspects of the cold-pressor procedure.
Efran, Chorney, Ascher, and Lukens (1989) also studied the
effectiveness of paradox and coping styles in a cold-pressor task.
Ninety-two male undergraduate subjects were randomly assigned
to one of four conditions: self-observation, exaggeration (paradox
ical instructions), rational self-statem ents, or a control group. The
authors also studied the effect of individual coping styles on the
effectiveness of these conditions. "M o n ito rs" (individuals who
prefer having inform ation about stressors) were com pared with
"blunters" (individuals who avoid cues associated with stressors).
The results revealed each of the treatment conditions to be equally
effective com pared to the control condition. Paradoxical intention
was no more effective than the other treatm ent strategies on cold-
pressor task performance. Furthermore, subjects tended to do bet
ter w hen an instructional set of coping styles was consistent w ith
their preferred coping style. These results are in line w ith those
of G illigan, Ascher, Wolper, and Bochachevsky (1984) w ho also
failed to find any advantage in the use o f paradoxical intention
over other strategies in a cold-pressor task.
Concerns about the use of paradoxical interventions have been
voiced by several researchers (Cottone, 1981; Feldm an, Strong, &
Danser, 1982; G reenberg & Pies, 1983). To evaluate possible n eg
ative effects associated w ith their use, Perrin and Dowd (1986)
exam ined the effects o f counselor self-disclosure and paradoxical
homework directives on clients' perceptions o f counselor social
influence. Eight undergraduate students experiencing problem s
w ith test anxiety (as identified by the Test A nxiety Scale-Short
Form; Sarason, 1978) were random ly assigned treatm ents in a
2 x 2 x 2 factorial design under the following conditions: (a) a para
doxical intention directive for test anxiety or a nonparadoxical
directive; (b) a moderate level of counselor self-disclosure or reflec
tive statements only; or (c) a gender condition. Subjects were seen
for four 30-m inute videotaped interviews. These videotapes were
then viewed by 89 test-anxious undergraduates, w ho rated them
on the basis of perceived counselor willingness and ability to help,
level of trickiness, m anipulativeness, client anger toward therapist,
and level of confusion created by the homework directive. The
su bjects also rated the counselor using the Counselor Rating
Form-Short Version (CRF-S; Corrigan & Schm idt, 1983). This scale
m easures subjects' perceptions of the counselor's attractiveness,

expertness, and trustw orthiness.
The data indicated that subjects perceived the paradoxical direc
tives to be more tricky and/or m anipulative and confusing than
nonparadoxical directives, but that this did not influence their per
ceptions of the counselor's willingness or ability to help or increase
feelings of anger toward the therapist. No gender effects were
found. A s the authors point out, the su bjects' perceptions of the
paradoxical interventions as tricky and m anipulative may be an
objective perception of reality that is not necessarily value-laden.
Furtherm ore, counselor self-disclosure does not necessarily
enhance or lim it positive perceptions of the counselor's social
influence. These results point to the possibility that the use of
paradoxical techniques does not serve to degrade the therapist-
client relationship. In addition to problem s typical o f analogue
studies, several difficulties specific to this investigation further
limit the generalizability of its data. Significant am ong these is the
fact that results were obtained from a group of su bjects w ho were
not them selves clients, but w ho observed sessions w ith a third
person.
Notwithstanding the significant problems of internal and exter
nal validity inherent in clinical analogue studies, the results su g
gest that paradoxical directives are at least as effective for the
com plaints of the college sam ples studied as are nonparadoxical
directives. In addition to outcom e data, inform ation regarding
aspects of the process associated with the use of paradoxical direc
tives was also obtained. For exam ple, those w ho receive paradox
ical treatm ents are likely to attribute change to internal,
nonvolitional causes (Wright & Strong, 1982; Lopez & W am bach,
1982; Shoham -Salom on et al., 1989). However, it appears that
attem pting to m axim ize this attribution may be detrim ental
(Zodun et al., 1985). Furtherm ore, this research suggests that
paradoxical interpretations (e.g., refram ing) may produce more
lasting change than interpretations that im ply negative connota
tions of presenting sym ptom s (Beck & Strong, 1982).
Although paradoxical interpretations and directives need not
be consistent to be effective, consistency is likely to increase
the recipient's confidence in the provider (Feldman et al., 1982).
Also, conscious awareness of the nature and intent of paradox
ical techniques is not necessarily detrim ental to the effective
ness of these procedures. These findings suggest that the
m echanism for client change may involve more than sim ple,
therapeutic double bind (Z odun et al., 1985). A lso noteworthy,
prelim inary evidence of the use of paradoxical techniques and
244 CHAPTER 8
perceived counselor social influence indicates that the use of

these techniques is not detrim ental to the therapist-client rela
tionship (Perrin & Dowd, 1986).
Finally, w ith certain m ethodological flaws w hich m ake it diffi
cult to draw firm conclusions, these studies are generally well con
ceived and conducted. A s such, analogue studies may be viewed
as cost-effective pilots for the form ulation and initial testing of
process and outcome hypotheses. Those obtaining support at this
preliminary level can then be exposed to forms of clinical research
that require more extensive resources.
SUMMARY AND CONCLUSIONS

Since at least the 1950s, therapeutic paradoxical procedures have
gained prom inence w ithin a variety of therapeutic orientations.
However, despite num erous positive anecdotal claim s extolling
these procedures, until recently, there has been a paucity of con
trolled em pirical data supporting the efficacy of this approach.
Currently, there has been a grow th of em pirical research in the
form of controlled case, analogue, and clinical outcom e studies,
evaluating the efficacy of paradoxical interventions.
Although clinical outcom e data are available for a wide variety
of disorders, m ost of these problem s are the focus of few studies.
Thus, the present review has em phasized insom nia and agora
phobia, as these behavioral difficulties are associated w ith the
greatest depth of data. Based on the lack of available studies, a
definitive statem ent on the efficacy of paradoxical techniques
would be prem ature. The old joke about two women vacationing
at a Catskill resort sum s up the problem: O ne wom an says to the
other, "The food here is terrible." And, the other woman responds,
"Yes, and such small portions." There are m any criticism s that one
can lodge against even the best controlled studies and there are
not sufficient data, even in the case of insom nia and agoraphobia,
to resolve conflicts and support useful conclusions.
Generally, results are encouraging w ith m ost studies, suggest
ing that paradoxical intention is at least as effective as— and som e
tim es more effective th an — conventional behavioral procedures.
This is the case with both clinical and analogue studies. Providing
further corroboration are several m eta-analyses. Both Shoham -
Salomon and Rosenthal (1987) and Hill (1987) found paradoxical
procedures to be as effective as conventional com parison treat
m ent, and som etim es more effective than these treatm ents of
choice, for a num ber of behavioral problem s. A gain, however, the
conclusions of these m eta-analyses share a significant flaw— they

are based on insu fficient n u m bers of experim ents. T h u s, as data
from an increasing nu m ber of reasonably w ell-controlled studies
accum ulate, a reasonable conclusion is that paradoxical p ro ce
dures seem to hold potential as effective th erapeutic p rocedu res,
in som e cases treatm ents o f choice, and an cillary tech n iq u es in
others. Effort expended in experim ental dem onstration of this effi
cacy m ig h t very well b e rew arded.
REFERENCES
Agras, W. S., Syllvester, D., & Oliveau, D. The epidemiology of common
fear and phobia. Comprehensive Psychiatry, 1 9 6 9 ,1 0 ,151-156.
Anderson, L. T. Desensitization in vivo for men unable to urinate in a
public facility, journal of Behavior Therapy and Experimental
Psychiatry, 1977, 8, 105-106.
Ascher, L. M. Paradoxical intention as a component in the behavioral
treatment o f sleep onset insomnia: A case study. Paper presented at the
m eeting of the Association for the Advancement of Behavior
Therapy, Decem ber 1975, San Francisco, CA.
Ascher, L. M. Paradoxical intention in the treatment of urinary
retention. Behaviour Research and Therapy, 1 9 7 9 ,1 7 , 267.
Ascher, L. M. Em ploying paradoxical intention in the treatm ent of
agoraphobia. Behaviour Research and Therapy, 1 9 8 1 ,1 9 , 5 3 3 -5 4 2 .
Ascher, L. M. Enhanced paradoxical intention and the self-recursive anxiety
disorder. Paper presented at the m eeting of the Boston Behavior
Therapy Interest Group, Septem ber 1983, W heaton College.
Ascher, L. M . Therapeutic paradox: A primer. In L .M . A scher (E d .)
Therapeutic Paradox. New York: G uilford Press, 1989a.
Ascher, L. M . Paradoxical intention and recursive anxiety. In L. M .
A scher (E d .) Therapeutic Paradox. New York: G uilford P ress, 1989b.
Ascher, L. M., Bowers, M. R., and Schotte, D. E. A review of data from
controlled case studies and experiments evaluating the clinical
efficacy of paradoxical intention. In G. R. Weeks (Ed.) Prom oting
C han ge Through Paradoxical Therapy. Homewood, 111.: Dow Jones-
Irwin, 1985.
Ascher, L. M., & Efran, J. The use of paradoxical intention in cases of
delayed sleep onset insomnia. Journal o f C on su ltin g and C linical
Psychology, 1978, 8 , 547-550.
Ascher, L. M., & Schotte, D. E. The use of the single case design for
246 CHAPTER 8
research in paradoxical intention. In Proceedings of the 3rd World

Congress of Logotherapy, Anaelecta Frankliana III, 1983.
Ascher, L. M., Schotte, D. E., & Grayson, J. B. Enhancing effectiveness
of paradoxical intention in treating travel restriction in
agoraphobia. Behavior Therapy, 1986, 1 7 ,124-130.
Ascher, L. M., & Turner, R. M. Paradoxical intention and insomnia: An
experimental investigation. Behavior Research and Therapy, 1979a, 17,
408-411.
Ascher, L. M., & Turner, R. M. A comparison of two methods for the
administration of paradoxical intention. Behaviour Research and
Therapy, 1979b, 1 8 ,121-126.
Ascher, L. M., & Turner, R. M. A comparison of two methods for the
administration of paradoxical intention. Behaviour Research and
Therapy, 1980,1 8 ,121-126.
Barlow, D. H. Anxiety and its disorders: The nature and treatment of anxiety
and panic. New York: Guilford, 1988.
Beck, J. T., & Strong, S. R. Stimulating therapeutic change with
interpretations: A comparison of positive and negative
connotation. Journal of Counseling Psychology, 1982, 29, 551-559.
Blitz, B., & Dinnerstein, A. J. Role of attentional focus in pain
perception: Manipulation of response to noxious stimulation by
instruction. Journal of Abnormal Psychology, 1971, 77, 42-45.
Boeringa, J. A. Blushing: A modified intervention using paradoxical
intention. Psychotherapy: Theory, Research and Practice, 1983, 20,
441-444.
Boettcher, L. Lv & Dowd, E. T. Comparison of rationales in symptom
prescription. Journal of Cognitive Psychotherapy: An International
Quarterly, 1988.
Bornstein, E H., Sturm, C. Z., Retzlaff, R D., Kirby, K. L., & Chong, H.
Paradoxical instruction in the treatment of encopresis and chronic
constipation: An experimental analysis. Journal of Behavior Therapy
and Experimental Psychiatry, 1981,1 2 ,167-170.
Corrigan, J. D., & Schmidt, L. D. Development and validation of
revision in the counselor rating form. Journal of Counseling
Psychology, 1983, 30, 64-75.
Cottone, R. R. Ethical issues related to use of paradoxical techniques in
work adjustments. Vocational and Work Adjustment Bulletin, 1981,
1 4 ,167-170.
Dattilio. The use of paradoxical intention in the treatment of panic
attacks. Journal of Counseling and Development, 1987, 66,102-103.
DeBord, J. B. Paradoxical interventions: A review of the recent
literature. Journal of Counseling and Development, 1989, 67, 394-398.
Efran, J. S., Chorney, R. L., Ascher, L. M., & Lukens, M. D. Coping

styles, paradox, and the cold pressor task, journal of Behavioral
Medicine, 1989,12, 91-103.
Elliot, R. A case of inhibition of micturition: Unsystematic
desensitization. Psychological Record, 1967,17, 525-530.
Espie, C. A., & Lindsay, W. R. Paradoxical intention in the treatment of
chronic insomnia: Six case studies illustrating variability in
therapeutic response. Behavior Research and Therapy, 1985,14,
21-33.
Espie, C. A., Lindsay, W. R., Brooks, D. N., Hood, E. M., &Turvey, T. A
controlled comparative investigation of psychological treatments
for chronic sleep-onset insomnia. Behavior Research and Therapy,
1989, 27(1), 79-88.
Feldman, D. A., Strong, S. R., and Danser, D. B. A comparison of
paradoxical and nonparadoxical interpretations and directives.
journal of Counseling Psychology, 1982, 29, 572-579.
Fisher, S. G., Nietzel, M. T., & Lowery, C. R. The credibility of rationales for
paradoxical intention in the treatment of sleep-onset insomnia.
Unpublished manuscript, 1985.
Frankl, V. E. The doctor and the soul: From psychotherapy to logotherapy.
New York: Knopf, 1955.
Theory, research and practice, 1975,12, 226-237.
Frankl, V E. Logos, paradox, and the search for meaning. In M. J.
Mahoney & A. Freeman (Eds.) Cognition and psychotherapy. New
York: Plenum, 1985.
Gerz, H. Experience with the logotherapeutic technique of paradoxical
intention in the treatment of phobic and obsessive-compulsive
patients. American journal of Psychiatry, 1966,123, 548-553.
Gilligan, R. M., Ascher, L. M., Wolper, J., & Bochachevsky, C.
Comparison of three cognitive strategies in altering pain behaviors
on a cold pressor task. Perceptual and Motor Skills, 1984, 59,
235-240.
Goldstein, A. J. Case conference: The treatment of a case of agoraphobia
by a multifaceted treatment program. Journal of Behavior Therapy
and Experimental Psychiatry, 1978, 9, 45-51.
Greenberg, R. G., & Pies, R. Is paradoxical intention risk-free?: A
review and case report. Journal of Clinical Psychiatry, 1983, 44,
66-69.
Haley, J. Problem solving therapy. New York: Harper & Row, 1976.
Hersen, M., & Barlow, D. H. Single case experimental designs: Strategies for
studying behavior change. New York: Pergamon, 1976.
248 CHAPTER 8
Hill, K. A. Meta-analysis of paradoxical interventions. Psychotherapy,

1987, 24, 266-270.
Hsu, G. L. K., & Lieberman, S. Paradoxical intention in the treatment of
chronic anorexia nervosa. American Journal of Psychiatry, 1982,139,
650-653.
Jacob, R. G., & Moore, D. J. Paradoxical interventions in behavioral
medicine. Journal of Behavior Therapy and Experimental Psychiatry,
1984,15,205-213.
Kales, A., Bixler, E. O., Lee, I. A., Healy, S., & Slye, E. Incidence of
insomnia in the Los Angeles metropolitan area. Sleep Research,
1974, 4, 139.
Kelly, J. E., & Lichstein, K. L. (1980). A sleep assessment device.
Behavior Assessment, 1980, 2,135-146.
Kraft, R. G., Claiborn, C. D., & Dowd, E. T. Effects of positive
reframing and paradoxical directives in counseling for negative
emotions. Journal of Counseling Psychology, 1985, 32, 617-621.
Lacks, R, Bertelson, A. D., Gans, L., & Kunkel, J. The effectiveness of
three behavioral treatments for different degrees of sleep onset
insomnia. Behavior Therapy, 1983,14, 593-605.
Ladouceur, R., & Gros-Louis, Y. Paradoxical intention vs stimulus
control in the treatment of severe insomnia. Journal of Behavior
Therapy and Experimental Psychiatry, 1986, 17, 267-269.
Lamontagne, Y. Treatment of erythrophobia by paradoxical intention:
Single case study. Journal of Nervous and Mental Disease, 1978,266,
304-306.
Lamontagne, Y., & Marks, I. M. Psychogenic urinary retention:
Treatment by prolonged exposure. Behavior Therapy, 1973, 4,
581-585.
Last, C. G., Barlow, D. H., & O'Brien, G. T. A comparison of two
cognitive strategies in a patient with generalized anxiety disorder.
Psychological Reports, 1983, 53,19-26.
Lesage, A., & Lamontagne, Y. Paradoxical intention and exposure in
vivo in the treatment of psychogenic nausea: Report of two cases.
Behavioural Psychotherapy, 1985,13, 69-75.
Lopez, F. G., & Wambach, C. A. Effects of paradoxical and self-control
directives in counseling. Journal of Counseling Psychology, 1982, 29,
115-124.
Margolies, R., & Gilstein, K. W. A systems approach to the treatment of
chronic encopresis. International Journal of Psychiatry in Medicine,
1983-84,23,141-152.
Mavissakalian, M., Michelson, L., Greenwald, D., Komblith, S., &
Greenwald, M. Cognitive-behavioral treatment of agoraphobia:
Paradoxical intention vs. self-statement training. Behaviour Research

and Therapy, 1983, 21, 75-86.
Meichenbaum, D. Cognitive-behavior modification: An integrative
approach. New York: Plenum Press, 1977.
Michelson, L. Treatment consonance and response profiles in
agoraphobia: The role of individual differences in cognitive,
behavioral, physiological treatments. Behaviour Research and
Therapy, 1986, 24, 263-275.
Michelson, L., Mavissakalian, M., & Marchione, K. Cognitive and
behavioral treatments of agoraphobia: Clinical, behavioral, and
psychophysiological outcomes. Journal of Consulting and Clinical
Psychology, 1985, 53, 913-925.
Michelson, L., Mavissakalian, M., & Marchione, K. Cognitive,
behavioral, and psychophysiological treatments of agoraphobia:
A comparative outcome investigation. Behavior Therapy, 1988, 19,
97-120.
Michelson, L., M avissakalian, M. Marchione, K., Dancu, C., &
Greenwald, M. The role of self-directed in vivo exposure in
cognitive, behavioral, and psychophysiological treatments of
agoraphobia. Behavior Therapy, 1986, 27, 9 1 -1 0 8 .
M ilan, M. A., & Kolko, D. J. Paradoxical intention in the treatment of
obsessional flatulence rum inations, journal of Behaviour Therapy
and Experimental Psychiatry, 1982, 13(2), 167-172.
M ozdzierz, G. J. The use of hypnosis and paradox in the treatm ent
of a case of chronic urinary retention/"bashful bladder."
American Journal o f Clinical Hypnosis, 1985, 28, 4 3 -4 7 .
O tt, B. D., Levine, B. A., & Ascher, L. M . (1983). M anipulating the
explicit dem and of paradoxical intention instructions. Behavioural
Psychotherapy, 1983, 11, 2 5 -3 5 .
Perrin, D. K., & Dowd, E. T. Effect of paradoxical and
nonparadoxical self-disclosure on counselor social influence.
Journal o f C ounseling Psychology, 1986, 33, 2 0 7 -2 1 0 .
Propp, L. A self-control treatm ent for encopresis co m bin in g self
ch artin g w ith paradoxical in stru ction s: Two case exam ples.
Journal o f C hild and Adolescent Psychotherapy, 1985, 2, 2 6 -3 1 .
Ray, I., & M urphy, J. M etronom e conditioned u rin ary reten tion.
Psychosom atic M edicine, 1975, 25, 5 4 3 -5 5 5 .
Relinger, H ., & B o rn stein , P. H . Treatm ent of sleep o n set in so m n ia
by paradoxical in stru ction : A m ultiple baselin e d esig n . Behavior
M od ification , 1979, 3 , 2 0 3 -2 2 2 .
Relinger, H ., B o rn stein , P. H ., & M u n g as, D. M . T reatm ent of
in so m n ia by paradoxical in ten tio n : A tim e series a n a ly sis.
B ehavior Therapy, 1978, 9, 9 5 5 -9 5 9 .
250 CHAPTER 8
Robbins, B. P, & Milburn, T. W. The effect of paradoxical intervention on

the performance of test anxious college students. Unpublished
manuscript, 1990.
Sarason, I. G. The test anxiety scale: Concept and research. In C. D.
Spielberger & I. G. Sarason (Eds.), Stress and anxiety (Vol. 5).
Washington, DC: Hemisphere, 1978.
Seltzer, L. F. Paradoxical strategies in psychotherapy: A comprehensive
overview atul guidebook. New York: John Wiley and Sons, 1986.
Shoham-Salomon, V., Avner, R., & Neeman, R. You're changed if you
do and changed if you don't: Mechanisms underlying
paradoxical interventions, journal of Consulting and Clinical
Psychology, 1989, 57, 590-598.
Shoham-Salomon, V., & Rosenthal, R. Paradoxical interventions: A
meta-analysis. Journal of Consulting and Clinical Psychology, 1987,
55, 22-28.
Solyom, L., Garza-Perez, J., Ledwidge, B. L., & Solyom, C.
Paradoxical intention in the treatment of obsessive thoughts: A
pilot study. Comprehensive Psychiatry, 1972, 13, 291-297.
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). STAI
manual for the State-Trait Anxiety Inventory. Palo Alto, CA:
Consulting Psychologists Press, 1970.
Timms, M. W. The treatment of urinary frequency by paradoxical
intention. Behavioural Psychotherapy, 1985, 23, 76-82.
Turner, R. M., & Ascher, L. M. Controlled comparison of progressive
relaxation, stimulus control, and paradoxical intention therapies
for insomnia. Journal of Consulting and Clinical Psychology, 1979,
47, 500-508.
Turner, R. M., & Ascher, L. M. Therapist factor in the treatment of
insomnia. Behavior Research and Therapy, 1982, 20, 33-40.
Watzlawick, P, Beavin, J., & Jackson, D. D. Pragmatics of human
communication: A study of interactional patterns, pathologies and
paradoxes. New York: Norton, 1967.
Webb, W. B. Sleep: The gentle tyrant. Englewood Cliffs, NJ: Prentice-
Hall, 1975.
Wright, R. M., & Strong, S. R. Stimulating therapeutic change with
directives: An exploratory study. Journal of Counseling Psychology,
1982, 29, 199-202.
Zodun, H. I., Gruszkos, J. R., & Strong, S. R. Attribution and the
double bind in paradoxical interventions. Unpublished manuscript,
1985.
Part Four
Future Directions
9
The Mysterious Affair of Paradoxes
and Loops
by Steve d e Shazer a nd Elam Nunnally
FOREWORD
The purpose of this chapter is to examine the puzzles of paradox
and strange loops through the use of a double description of a
case example. The results of this twin explication suggest that we
use Occam's razor to cut through the complexity, and that a more
simple explanation is available.
In the classic murder mystery, the sleuth is surrounded by a
not so bright sidekick (the Watson figure) and not so bright police
(the Lestrade figure). During the course of the story the police
view the crime from a particular point of view which turns out to
be false; they are continually led astray by following red herrings
(clues which when followed too far lead an investigator in the
wrong direction). The sidekick too is led astray by viewing events
from the wrong perspective, although his com m ents sometimes
help to illuminate the problem for the sleuth. After a while, the
sleuth solves the puzzle by considering the same "fa cts " the po
lice and the sidekick did, but from a different perspective—usu
ally "upside-dow n."
The clues, red herrings, and use of the wrong perspective are
252
THE MYSTERIOUS AFFAIR O F PARADOXES A N D LOOPS 253
all part of the writer's craft, designed to mystify and entertain the
readers: The readers' game is to arrive at the proper solution be
fore the sleuth does. If readers are too quick, they spoil their own
pleasure. Therefore, the successful author needs to be very clever
indeed in the use of clues.
If the field of "m urder m ystery" had com m entators and con-
ceptualizers similar to those in the field of family therapy, the
story might be seen as follows: Each of the suspects and wit
nesses has a perspective (frame) that explains the m eaning of
their behavior and their perceptions, as do the police, sidekick,
and sleuth. Each of the clues, and red herrings is intended to pro
mote framing of the story in a "w rong way." At the end, the
sleuth reframes all prior frames, gets a new understanding (or
new perspective) of all the facts, and designs a counter red her
ring with this new point of view. In the classic final chapter,
which includes a m eeting of all the suspects, the detective uses
his counter red herring to break the old frames that frequently
causes the culprit to confess (some new behavior).
The sleuth then gives a post hoc explanation of how he arrived
at the solution. O ften this explanation is ju st a simple m atter of
redescribing the story from the point of view that inevitably led
him to the real culprit. If a clue pointed the police in one direc
tion, the sleuth saw it point in the exact opposite. O f course, dur
ing the development of the story, the sleuth too m ight have been
led astry now and then, but by the three quarter point, he knows
the proper frame to use to understand the problem.
Read in the spirit of murder m ysteries, the history and devel
opm ent of family therapy seem littered with clues as to the nature
and identity of the agent of change. W ho or what is the real agent
of change? For quite some time, many investigators followed the
clue of "in sig h t." (Som e still pursue it.) Discouraged with their
progress in pursuing that path, many investigators picked up on
the clue of "aw areness." Although awareness seemed more
promising than insight for a time, it soon became apparent to in
vestigators—especially the statistically m inded—that awareness
accounted for very little of the variance in observed change. This
is not to say that the clues offered by insight and awareness
added nothing to the search, for in figure/ground terms, the
study of how insight and awareness related to change provided
some detail for the ground portion of the picture.
Investigators were thus stimulated to seek a new clue to flesh
out the shadowy figure and discern the true identity of the
change agent. Soon this new clue was found—a clue so pervasive
that investigators kept stumbling over it everywhere—paradox. In
254 CHAPTER 9
a murder mystery, such a pervasive clue usually alerts the reader

to consider its possible falseness and wonder w hether it is a red
herring designed to lead him astray. If he then follows the clue to
try to find out what paradox m eans, he finds it leading in many
different and unrelated directions.
W hen a m urder m ystery reader is repeatedly confronted with
red herrings, the reader begins to wonder if the “ clu es" m ean
anything at all. A n apt analogy here is The Floating Admiral, the
com bined effort of various authors such as Christie, Sayers, and
Chesterton (1980), each of w hom contributed a chapter to the
story. Not knowing what the previous author had in mind or
what the clues m eant to him or her, the author of each subse
quent chapter tries to be more clever by half. At the end of The
Floating Admiral each author subm itted his or her solution (and
som e were too dumbfounded to figure it out at all). The reader
may be reminded of Agatha Christie's M urder on the Orient Express
in which all the various clues m eant nothing because they contra
dicted each other and led to each of the various suspects in turn
proving in the end that "everybody done it."
PARADOXES
The Oxford English Dictionary gives several definitions and ex
am ples of paradox. Am ong them are:
Paradox 1. "A statem ent or tenet contrary to received opinion

or belief, often with the implication that it is marvellous or
incredible; som etim es with unfavourable connotation as be
ing discordant with w hat is held to be established truth, and
hense absurd or fantastic, som etim es with favourable conno
tation, as a correction of vulgar error."
Translation, Burgersducem s' Logic 11, 1697, XV, p. 65. "A
Paradox is said to be a problem e true against the com m on
O p in io n .. .su ch as that viz, the Earth moves; which though
it be true, yet it may be so against the com m on O pinion, and
therefore a Paradox."
de Quincey Templars' Dial, 1854, W ks TV, p. 183. “A paradox
you know is simply that which contradicts the popular opin
ion—w hich in too m any cases is the false opinion."
Paradox 2. “A statem ent or proposition w hich on the face of it
seem s self-contradictory, absurd or at variance with com m on
sense, though on investigation or w hen explained, it may
prove to be well founded (or according to som e, though it is

essentially true)."
J. Norden Surv., Dial, 1607, IV, p. 195. " I can tell you a pretie
p arad o x e.. .Boggy and spungy g ro u n d .. .though it owne na
ture it be too moist, yet if it be overflowed with water it will
often settle and becom e firm e" (Warner, 1982, pp. 3 -4 ).
Historical Paradoxes
Zeno of Elea is well known for several paradoxes, am ong them
"Achilles and the Tortoise" and "T h e Flying Arrow." The flying
arrow conundrum goes like this: A n object is at rest w hen it occu
pies a space equal to its own dim ensions. An arrow in flight occu
pies, at any given m om ent, a space equal to its own dim ensions.
Therefore, an arrow in flight is at rest.
The logic here follows from the assum ption that tim e is com
posed of discrete m om ents. If this assum ption is not held, then
Zeno's conclusion will not follow. Zeno's assum ptions further in
cluded the idea that space is infinitely divisible and thus any finite
distance contains an infinite num ber of points: It is im possible to
reach the end of an infinite series in finite time. According to
Nahm,
Zeno's argument involves the assumption that the flying arow is at
rest at any point in its trajectory. But this can be said of every point
in the trajectory and what is at rest at every point does not move at
all. The solution to the paradox is impossible for philosophy until
mathematics, by the development of the differential calculus could
deal with the general problem of velocity at a point. Once the
mathematics is developed, the paradox of the flying arrow may be
considered as one involving a definite velocity for the arrow at
every point of its trajectory. (Nahm, 1964, p. 99)
The Achilles and the tortoise paradox follows: The tortoise,
given a head start, will never be overtaken in a race by the swifter
Achilles, for it is necessary that Achilles should first reach the
point from which the tortoise started, so that necessarily the tor
toise is always som ewhat ahead. By the time Achilles reaches
point two, the tortoise will be at point three and by the time
Achilles reaches point three, the tortoise will have moved on
again, and so on, ad infinitum.
According to Quine,
When we try to make this argument more explicit, the fallacy that
emerges is the mistaken notion that any infinite succession of inter
vals of time has to add up to all eternity. Actually when an infinite
256 CHAPTER 9
succession of intervals of time is so chosen that the succeeding in

tervals become shorter and shorter, the whole succession may take
either a finite or an infinite time. It is a question of a convergent
series. (Quine, 1976, p. 3)
Since Zeno's time, various conceptual schem es have developed

and shifted, each changing the understanding of this conundrum.
Quine points to a fallacy:
the notion that an infinite succession of intervals must add up to an
infinite interval. But surely this was part and parcel of the concep
tual scheme of Zeno's day. Our recognition of convergent series, in
which an infinite number of segments add up to a finite segment,
is from Zeno's vantage point an artificiality. (Quine, 1976, p. 9)
Eastern Paradoxes
Paradoxes of various types have long fascinated man, Eastern
as well as Western. W hile in the West paradoxes have been con
sidered intellectual, philosophical, mathematical puzzles to be
solved, the tradition in the East is different. Here paradoxes are
part of the path to enlightenm ent, particularly in the Zen Bud
dhist tradition. The paradox or koan is accepted as is; the Bud
dhist is mainly interested in the consequences of getting beyond
the conundrum. For instance, in the famous Zen illustration, a
Buddhist monk stood holding a stout stick over the head of his
student. He is reported to have said to his student: " I f you say
this is a stick, I will hit you with it. If you say this is not a stick, I
will hit you with it. Speak: Is it a stick or is it n ot?"
W ithin the Zen tradition, the student is very dependent on the
monk and their relationship is a highly valued one. Part of the
way a student becomes enlightened is by solving these riddles
posed by the monk. In a Western or any either/or context, it looks
as if the student is sure to be hit but in the Zen context, the stu
dent might get beyond the m onk's either/or koan by leaping up,
grabbing the stick, and screaming nonsense.
Perhaps this fascination with paradox is based on the idea that
people like to puzzle themselves (the murder mystery) and con
fuse themselves and each other, or that people like to make the
world problematic. W hile conundrum s can be frustrating, they
can also be fun. Be that as it may, in the field of family therapy
alone, paradox has been part of some of the best, most profound
thinking, and part of some of the most obtuse and confusing
thinking.
The Concept of Paradox in Family Therapy

In an effort to bring order out of confusion, one investigator
consulted the oracles. In A Delphi Study on Paradox in Therapy,
Christine Watson (1982)—no relation to Dr. John H. W atson—con
sulted a panel of expert therapists, pooling their opinions about
therapeutic paradoxes and searching for com m on them es and the
possibility of consensus. W hen asked "W h at elem ents are neces
sary for an intervention to be categorized as a paradoxical inter
vention?" responses from 18 oracles were distinguished as much
by their variety as by any com m onalities. Watson also invited the
experts to suggest criteria for using paradox and for not using
paradox in their interventions. As with the definition of paradoxi
cal intervention, "T h e criteria for use seem to be very in d iv id u a l-
each therapist using very different stylistic variables in deciding
when paradox is appropriate" (Watson, 1982, p. 101).
A third feature of the study was a request for the experts to tell
what they would consider doing as a strategic paradoxical inter
vention if involved with a case for which they were provided a
short summary. Watson (1982) found that the 19 responses re
ceived were "a s varied as the approaches to a definition" (p. 83).
Watson makes the point, however, that "T h ere exist an infinite
num ber of possible paradoxical interventions which could jolt or
surprise the client out of his/her current pattern" (p. 84). Further
more, "Regarding which particular m odus operandi to use in
therapeutic com m unication, almost every expert appeared to use
different conceptualizations" (p. 84). Com m enting on the variety
of responses to the case, Watson concludes that the results "w ar
rant em phasizing the seeming looseness of the term 'paradoxical
intervention.' It has come to mean 'all things to all th erap ists'" (p.
85). Nevertheless, Watson does not appear to be calling on thera
pists to set the term aside, for she adds that "T h e term 's defini
tion stands as a m etaphor for itself—encouraging variety, creativ
ity in problem solving, and a subtle sense of confusion about
what is happening" (p. 85).
Although the terms paradox and paradoxical intervention were
found to refer to slippery concepts, elusive of definition, a special
case of paradox—the double bind—has a widely accepted defini
tion. (Editor's N ote: The double bind hypothesis was first pro
posed by Gregory Bateson, Don Jackson, Jay Haley, and John
Weakland in a paper called "Toward a theory of schizophrenia."
This hypothesis has been interpreted in different ways, as shown
in Chapter 1.) Perhaps the double bind concept offers a useful
clue to the identity of the change agent.
258 CHAPTER 9
A DOUBLE BIND MAPPING

Watzlawick has described the double bind concept as follows:
1. Two or more persons are involved in an intense relationship
that has a high degree of physical and/or psychological sur
vival value for one, several, or all of them .
2. In such a context, a m essage is given which is so structured
that (a) it asserts som ething, (b) it asserts som ething about its
own assertion, and (c) these two assertions are mutually
exclusive.
3. Finally, the recipient of the m essage is prevented from step
ping outside the frame set by this m essage, either by com
m enting on it or withdrawing (Watzlawick et al., 1967, p. 212).
It is further assumed that these patterns are recurrent in the rela
tionship, and that the com plete set of ingredients is no longer
necessary w hen the people concerned have learned to perceive
the universe in double bind patterns.
It is im portant to recognize the distinction betw een contradic
tory and paradoxical injunctions. W ith a contradictory injunction,
the recipient is faced with a choice: He can decide on one alterna
tive, but then m ust suffer the consequences of not picking the
other. W ith a paradoxical injunction, the choice is an illusion: No
behavior is logical, and a self-perpetuating oscillation is set in mo
tion. According to W ilden,
If the double bind generates irresolvable oscillations between "yes"
and "no," it can do so only within a digitalized context of either/or;
the context of analytical logic. Such oscillations do not present a
problem for dialectical logic, which is of a higher logical type than
analytical logic and (paradoxically) subsumes it. In other words,
double binds are irresolvable only when metacommunication—in
logic or in life—is prevented through the way in which allowable
communication is framed or punctuated by those with the power to
do so. (Wilden, 1980, p. 123)
That is, if the either/or frame is not part of the context in which
the "paradoxical m essage" is sent and received, then the oscilla
tion or illusion of choice will not be "pathological."
Case Example
A couple was referred for therapy by a drug counselor who
viewed their marital problem s as precluding treatm ent for drug
abuse.1 (The couple was using cocaine three or more times per week; this
had been their pattern for over two years.) The wife Jane (not her real
name) described their situation to the team as one in which their joint
drug use was messing up their marriage; therefore, she wanted to stop the
drugs to save the marriage. The marital problems were, in her view,
symptoms of their drug problem.
Ralph (not his real name) did not see their use of drugs as the
real problem but rather their fights (some of which becam e physi
cal) and their argum ents (some about drugs) as the m ain concern.
He thought the fights and argum ents needed to stop to save their
marriage.
Up to this point, their dilemma m ight be described as a simple
contradiction, but their situation was not this either/or—even to
them . Interestingly, they also shared the notion that using drugs
prevented boredom (which neither of them handled well) and
that stopping the drugs m ight lead to the breakup of the marriage
they both valued highly because they would have less or maybe
even nothing in com m on.
Plotting this couple's situation onto a double bind map, we see
that this is a relationship in which
1. Use of drugs is m essing up their marriage; fights and argu

m ents—som e about drugs—are increasing.
2. Their use of drugs prevents boredom ; if they stop drugs, their
marriage m ight break up.
3. W ithdrawing from this bind m ight be accom plished by sepa
rating, but this is the very action they wish to avoid.
4. Increasing use of cocaine m ight provide a way to withdraw in
a sense by soothing them through some of the conflicts, but
then the argum ents and fights would probably increase, be
cause they are in conflict about their drug use, and the argu
m ents and fights are breaking up their marriage.
Unless they could find some way to step outside the frame, the
couple seem ed destined to remain in a self-perpetuating oscilla
tion, which could well becom e lethal.
The team developed the following intervention m essage, which
was delivered by the therapist (or conductor) at the close of the
session:
'The therapy described in this illustration was done by the team at the Brief
Family Therapy Center (BFTC) which includes, in addition to the authors, Insoo
Berg, Marilyn La Court, Eve Lipchik, and Alex Molnar.
260 CHAPTER 9
You've got a problem.

It seems to us, Ralph, that your marital problems are being ex
acerbated by the drugs, or fogged over by the drugs, or perhaps
even created by the drugs. Perhaps you need to stop the drugs,
just to see what is going on. But, on the other hand, we agree with
you, Jane, that if you two were to stop the drugs, then there might
be nothing there. And, you might not have time to create anything
before the marriage broke up. In short, we don't know what the
f— you are going to do.
I suggest you think about what I just said, and decide what ac
tions you are going to take.
Mapped as a counter double bind, the intervention contains

messages which state that the team views this as a relationship in
which
Stopping the drugs may be necessary to save the marriage.
Not stopping the drugs may be necessary to save the marriage.
Either of the above may break up the marriage.
Any alternatives you have thought of pose great risks to your
marriage.
You should take actions you have not thought of.
That is,
1. W ithin the context of therapy which has a high survival value
for their marriage, a m essage is sent.
2. The m essage (a) asserts that stopping the drugs is necessary
for saving the marriage, (b) asserts that this assertion is f a l s e -
stopping the drugs might break up the marriage, and (c) these
are mutually exclusive (either stop or not stop).
3. A nonspecific action is demanded (think about what you are
going to do and do it) which is designed to prom ote their get
ting outside their either/or frame.
By the next session one week later, Jane and Ralph had cut
their drug use by two-thirds, although their use followed the
same schedule. Furthermore, without talking about it, they had
initiated some new joint and separate activities. This time the
main thrust of the intervention m essage centered around the
team's worry about a relapse.
O ne week later, Jane and Ralph reported having eliminated the
drugs: They were continuing new activities—both together and
separately—and enjoying them, and arguing m uch less. In the in
tervention m essage, the team again worried about a relapse, spe
cifically about how soon a relapse might occur.
Follow-up contact at six m onths and one year indicated that

there had been no relapse (no drugs and only infrequent argu
ments). The couple also reported further improvements in their
life, both together and separately.
Notwithstanding that Jane and Ralph's dilemma and the intervention
can be plotted retrospectively2 onto double bind maps, we have to ask
ourselves: Is this particular kind of map useful in generating effective
interventions, and could other kinds of maps do the job better? Putting
the question in terms of our detective story metaphor: Are double binds
and counter double binds clues to the nature of persistence and change or
are they red herrings, distracting us from a better explanation? To help
answer this question, we turn our attention to alternative ways of
mapping.
A “STRANGE-LOOP” MAPPING
Cronen, Johnson, and Lannam ann (1982) have developed a new
theory of reflexivity in system s of social meaning and action. In
their view, Russell's Theory of Types (and, therefore, the double
bind) was based on an "inappropriate and largely outdated episte
m ology" (p. 91). Their theory considers reflexivity to be a natural
and necessary feature of hum an system s of meaning and rejects
the idea that reflexivity and paradox are coterm inus. Behavior, con
tent, episodes (interactions), relationships, life scripts, and cultural
patterns are all seen hierarchically, mutually defining each other:
Som e loops are problematic and some are not (de Shazer, 1982 [a]).
W hen the meaning of a situation cannot be determ ined by
moving through the hierarchical levels, the situation can be de
scribed as a "strange-loop." According to Hofstadter, a strange
loop can be mapped when, after moving through the hierarchical
levels "W e unexpectedly find ourselves right back where we
started" (1979, p. 10). A "charm ed-loop" can be distinguished
from a strange-loop w hen the natural and normal reflexivity is not
problematic. A mapping technique developed by Karl Tomm
(1982) will be used to illustrate the application of the strange-loop
description of the above case example.
If the couple's situation could be described as a charmed-loop,
then the meaning of the situation could be picked from any of the
2The double bind and counter double bind map was not used for the actual design
of this intervention. Over the past 6 or 7 years we at BFTC have found using this map in
designing interventions to be cumbersome and time consuming, that is, it takes longer
than the 10-12 minutes allotted for intervention design during the hour-long session.
262 CHAPTER 9
following, perhaps by moving up to another level of the hierar

chy: (1) stopping the drugs should lead to saving the marriage, or
(2) not stopping the drugs should lead to breaking up the mar
riage, or (3) stopping the drugs should lead to breaking up the
marriage, or (4) not stopping the drugs should lead to breaking
up the marriage.
But the situation is not that simple, not that charm ed. Both
people (and the observers) think that stopping the drugs might
either break up the marriage or save the marriage, and both peo
ple (and the observers) think that continuing the drugs might
either break up the marriage or save it. The couple's situation is
clearly a conundrum for the couple as well as the observers. Each
individual's position, which seem ingly should be distinct from
the other's includes its opposite, which should be logically ex
cluded—a strange-loop. W ithin the context of this marriage,
neither stopping nor not stopping the drugs can determ ine stay
ing or not staying together. The reflexivity of this couple's situa
tion is such that m eanings (and therefore actions) cannot be deter
mined through the context in which the behaviors and episodes
appear. The situation can be m apped by an observer in this way:
FIGURE 1
This map is an attempt to clarify the im passe and make under

standable the pragmatic effects of a strange-loop. It is im portant
to rem em ber that strange-loops, charm ed-loops, and double
binds do not exist. They are simply part of the descriptive tools an
observer brings to the observed situation, a way of organizing
data. Loops are part of the map, not part of the territory. Like any
mapping tool, they are either useful for the observer or they are
not. (The description either fits the observation or it does not.)
O ne reading of the map goes as follows: If one should w ant to
break up the marriage, then one should continue the drugs—but
that might save the marriage: Therefore, one should stop the
drugs. But stopping the drugs m ight break up the marriage;
therefore, one should continue the drugs to save the marriage.
Here is another reading of the map: If one should w ant to save
the marriage, then one should stop the drugs; but this m ight
break up the marriage; therefore, one should continue the drugs
to save the marriage.
The strange-loop map is a figure eight (on its side) which can
be read by starting at any of the four elem ents and then following
the arrows. Clearly this map fits the situation the couple described
and points out that the m eaning of their situation cannot be deter
mined, thus preventing the couple from making any decisions
and taking any actions since they are caught in confusion. There
is no way to resolve this interactional situation.
This strange-loop description of the couple's situation needs to
be placed within the context of therapy. This context includes the
therapist, the team behind the screen, and the videotaping equip
m ent (since the couple can see the camera and gave perm ission to
tape). Furtherm ore, the description also needs to include the
m eanings given to that context since the couple system is now a
subsystem w ithin the therapeutic suprasystem (de Shazer, 1982 [a]
& [b]). The wife came to therapy to stop the drugs in order to save
the marriage. The husband thinks that therapy is nothing but talk
and any advice will be useless. The therapist and the team by defi
nition think that therapy can be useful, but they need to qualify and
modify their views based on the couple's definitions and m ean
ings. That is, for this couple, effective therapy needs to be more than
just talk and needs to exclude useless advice.
The intervention m essage (repeated here for clarity) is an at
tem pt to give new m eanings to the couple's situation.
You've got a problem.

It seems to us, Ralph, that your marital problems are being ex
acerbated by the drugs, or fogged over by the drugs, or perhaps
even created by the drugs. Perhaps you need to stop the drugs,
just to see what is going on. But, on the other hand, we agree with
you, Jane, that if you two were to stop the drugs, then there might
be nothing there. And, you might not have time to create anything
before the marriage broke up. In short, we don't know what the
f— you are going to do.
264 CHAPTER 9
I suggest that you think about what I just said, and decide what
actions you are going to take.
The team attem pts to redefine the situation as one in which some
unspecified action is necessary, but that action is not stopping the
drugs or not stopping the drugs since neither can save the mar
riage. (The team does, however, imply the need to stop the drugs,
but not because that action will save the marriage.)
The team attempts to define the situation as one in which the
couple needs to "create som ething" in order to save the marriage;
furthermore, they deliberately make the presupposition (in the
last sentence) that the couple is going to take action. Clearly this
intervention can be seen as an attempt to introduce new criteria
about saving the marriage—taking action or doing som ething dif
ferent rather than fighting about stopping or not stopping the
drugs.
In short, the intervention can be seen as based on the same
strange-loop map. The reframing attem pts to change the meaning
of the arrows, or to disconnect them to break the recursive cycle.
There is some chance that the meaning is slippery enough to
prompt a different response from either husband, wife, or both.
The intervention introduced the possibility of some new behavior
which might make a "different enough difference."
O C C A M ’S RAZOR
William of Occam is best known for suggesting that we look for
the simplest explanation that fits. This advice is extremely perti
nent to therapists designing interventions. The strange-loop is no better
than the counter double bind map when designing interventions,
although both are useful retrospectively.3 In the everyday world of doing
therapy, there frequently is not enough time to use either map: The
criteria are too complex and both are more suited to post hoc
explanations.
Typically, w hen a client presents a complaint, some behavior
(i.e., not sleeping) will be seen as a "sym p tom ," and one facet of
the context (i.e., being depressed) will be seen as a "problem ."
The context and the behavior are recursive and inseparable, each
defining the other. W ithin the context of the particular marriage
discussed in this chapter, stopping or not stopping the drugs,
saving or not saving the marriage, are the problem. As O 'H anlon
3 As is the case with the double bind map, the strange loop map is too cumber
some and time consuming. See Footnote 2 and below.

puts it: " I f the loop generates itself again and again, we can speak
of the pattern being maintained by the redundant sequence (re
ally itself). At this point, as Bateson says, 'th e pattern is the
th in g '" (1982, p. 27).
If, as Bateson maintains, the breaching of the Theory of Types
is continual and inevitable in hum an com m unication (Bateson et
al., 1956) and if, as Cronen (1982) maintains, this sort of reflexiv
ity is normal and necessary, then we need Occam 's razor to sim
plify the clinical situation in order that effective interventions can
be designed within the usual clinical environment.
Implicit within the double bind and counter double bind expli
cation and the strange loop explication, is the notion that the in
tervention needs to be a mirror image o f the problem. The criteria for
a therapeutic double bind is simply the mirror image of the crite
ria for a pathogenic double bind: Like cures like. The central
premise implied by both the above explications is that therapeutic
interventions can be built on the same description (map) as that
used to describe the interaction.
It is the fit betw een the therapist's description of the pattern
and the form or map of the intervention that seem s central to the
process of initiating therapeutic changing. That is, the couple de
scribes the problematic pattern within a certain context/meaning/
frame, and then the therapeutic intervention is based on the same
pattern, but—im portantly—with a difference. The pattern upon
which the intervention is designed is a mirror image of the cou
ple's pattern. A mirror reflects what is placed in front of it with
one difference—the right-left reversal.
DOUBLE DESCRIPTION
Bateson referred to the double description (1979) of the same
pattern or sequence as the source of "id eas." If two descriptions
of the same sequence are not identical or are simply redundant,
then the com bined descriptions will include some news of differ
ence. Two identical descriptions without any difference are use
less because the com bination includes no news of difference;
therefore, no idea or bonus is possible. However, if two descrip
tions are not isomorphic to a high degree, the difference may be
too great to prompt ideas—almost like two descriptions of two dif
ferent sequences.
This descriptive process has been compared to the way our two
eyes work together to develop depth perception (Maruyama,
1977; Bateson, 1979; de Shazer, 1982 [a] & [b]). Similarly, it is the
266 CHAPTER 9
difference betw een w hat each eye sees that leads to the bonus of
depth perception, which is unavailable to one eye. If the individ
ual eyes were set too far apart, the degree of isom orphism would
be too low for the brain to com pute the difference and there
would be no depth perception bonus. However, if the views from
each eye were too similar, or we were to use only one eye, there
would be no news of difference and no depth perception.
The information from two descriptive processes with a high de
gree of isomorphism and yet some difference is of a different logi
cal type than that included in one description or m onocular vi
sion. The bonus is only available through the inform ation con
tained in the difference betw een the two descriptions. For
instance, there is more and different inform ation in two descrip
tions of two different chess gam es being played than in either
description by itself. The com parison inform s us of the difference
betw een the two games and the play options. This helps us de
velop an idea of the "gam e of ch ess."
The double bind maps and strange loop m aps of the case illus
tration presented here allow for the use o f O ccam 's razor. Both
descriptions involve m apping the same intervention pattern onto
the map, or describe the problematic pattern. Both include m any
of the same elem ents: There is a high degree of isom orphism be
tween double bind maps and strange-loop m aps. A bonus (or
idea) develops from this double description of the couple's situa
tion and the therapy situation: The intervention, regardless of the
principles behind its design, needs to fit with the client's pattern
in such a way that it becom es a mirror image. This is the sim pler
explanation that follows from the use of Occam 's razor.
Initiating Changing
Maruyama extends the m etaphor of binocular vision to system s
and suggests that system s use polyocular ways to know (1977).
"T h e Japanese think in poly-ocular v isio n . . . and they do not even
bother to find out 'objectivity,' because they can go m uch further
with cross-subjectivity" (Maruyama, 1977, p. 84). That is, the way
hum an system s know is based on the differences betw een the
various views of the members o f that system .
This line of thinking is at least implied in the various team ap
proaches to family therapy. Team approaches can be described as
using Bateson's double or multiple description (a polyocular view)
to get at the idea of system and understand w hat is going on in
the therapy room (Palazzoli et al., 1978; de Shazer, 1982 [b]). The
fam ily presents (shows) their polyocular description and simulta
neously each m em ber o f the team develops his or her own

description, all of w hich are placed in a collective (polyocular)
context during the consultation break in the therapy session.
Thus the team (a system ) evolves systemic notions about the
family's (a system ) problem , and more than one interactional
description is used to describe the interaction during the session.
Then "id e a s" about those patterns or sequences within a context
start to evolve. W hen the team presents their mirror image back
to the family in the form of a m essage, the family can receive a
bonus, w hich results from the ability to map the team 's interven
tion upon their own descriptions. Thus a difference can be noted. When
this is a different enough difference, the intended bonus develops and
changing is initiated.4
Since the couple (in the case example), like any hum an system ,
can be described as showing regularities or redundancies, chang
ing or noticeable difference cannot be initiated unless there is
som e source o f random ness that an intervention attem pts to pro
vide through the difference betw een the client's description and
the therapist's. A s Bateson put it,
A sequence of events is said to be random if there is no way of
predicting the next event of a given kind from the event or events
that have preceded and if the system obeys the regularities of prob
ability. Note that the events which we say are random are always
members of some limited set. The fall of an honest coin is said to be
random. At each throw, the probability of the next fall being heads
or tails remains unchanged. But the randomness is within the lim
ited set. It is heads or tails; no alternatives are to be considered.
(Bateson, 1979, p. 230)
Betw een the intervention and the following session, the couple
initiated som e new, joint activities and som e new, separate activi
ties, and reduced their drug use. At the tim e of the intervention,
neither the couple nor the team would have been able to predict
these changes, particularly in light of the im passe. Further
changes followed, including the elimination of the drug use and a
drastic reduction in the frequency of argum ents. These further
changes frequently follow once changing is initiated, a process
which Spiegel and Linn (1969) call the ripple effect.
4 A team approach (and thus the polyocular view) is the ideal research setting for
this type of comparative study. In addition to being useful for research tasks, a team
approach is also valuable for training purposes. We have found the concept of fit
(isomorphism, double description) to be more teachable and more readily usable for
designing interventions when working than either the double bind map or the strange
loop map because it is less time consuming and less cumbersome. See Footnotes 2 and
3.
268 CHAPTER 9
CONCLUSION
The "so lu tio n " to the paradox puzzle suggested by the double
description of the case example helps to clarify how changing is
initiated. Rather than paradox being the agent of change, this
view suggests instead that changing has som ething to do with the
fit between the couple's pattern and the pattern of the interven
tion. The fit between the two is not exact, even though both
descriptions can be mapped onto each other with a high degree of
isomorphism. The difference between the two patterns can be
seen to prompt a bonus, that is, the introduction of something
new or random into the couple's system.
In no way do we mean to imply that the concepts of paradox,
double bind, counter double bind, and strange/charmed-loops are
somehow "w rong ." Each is a good mapping tool, but it is impor
tant to remember the differences between maps and the territories
represented. Paradoxes and loops are part of our descriptive tools
and are useful only to the extent that they promote successful in
tervention. The difficulty arises w hen the concept becom es reified
and paradox is seen as the change agent. This reification can
prove to be a barrier that separates one subsystem from another
in the therapy situation, thus obscuring the interactive, systemic
nature of successful intervention.
We do suggest that the reified concept of paradox/paradoxical
intervention is similar to a red herring in a m urder mystery. Like
the manner in which any witness gives a description, observa
tions are grounded by "facts," and like any such story, observa
tions are made from the observer's point of view—which includes
biases. The paradox clue became a red herring as soon as it led
therapists to think that the concept was somehow real and true
(reification). Therefore, paradox was both the cause of human
problems and the agent of change. This leads to applying para
doxes (as interventions) to problems which can be mapped more
simply in other ways and therefore can use different interven
tions. This line of thinking can clear up the definitional muddle
that Watson (1982) confirmed.
Unlike Sherlock Holmes or any sleuth in a murder mystery, we
are not suggesting that our solution to the puzzle is "tr u e " or the
only one. Instead we suggest that the mirror image concept and
the polyocular view are useful in designing effective interventions
within the context of the usual therapeutic environment, and not
just post hoc explanations. Over the past six years, we have found
these concepts useful with a large num ber of clients with a wide
range of presenting problems or complaints. During this time, we
have found no need to describe what clients are doing or what we

are doing as "paradoxical."
POSTSCRIPT
Because the couple in our case example reported such radically
different behavior than that which they had dem onstrated prior
to our intervention and the follow-up session, we were struck by
the similarity between the intervention and a Zen koan. This con
nection is not a new one (Watts, 1961). In terms of the so-called
pragmatic effects of the intervention (the causal relationship prob
ably is not this simple), the couple did not try to figure out the
message in Western fashion. They responded more in an Eastern
way, similar to the Zen student's new behavior of leaping up,
grabbing the m onk's stick, and screaming "n o n sen se." O ur ex
planation is not very useful as it explains by using the unexplain
able koan. But this is the Zen way of doing things!
REFERENCES
Bateson, G., Jackson, D. D., Haley, J., & Weakland, J. H. Toward a
theory of schizophrenia. Behavioral Science, 1956, 1, (4), 251-264.
Bateson, G. Mind and nature: A necessary unity. New York: Dutton,
1979.
Christie, A., Sayers, D., Chesterton, G. K., & Certain Other Members
of the Detection Club. The floating admiral. New York: Charter,
1980.
Cronen, V., Johnson, K., & Lannamann, J. Paradoxes, double binds,
and reflexive loops: An alternative theoretical perspective, Family
process, 1982, 21, (1).
de Shazer, S. Some conceptual distinctions are more useful than
others. Family Process, 1982, 21, (1), 71-84. (a)
de Shazer, S. Patterns of brief family therapy. New York: Guilford,
1982. (b)
Hofstadter, D. R. Godel, Escher, Bach: An eternal golden braid. New York:
Basic Books, 1979.
Maruyama, M. Heterogenistics: An epistemological restructuring of
biological and social sciences, Cybemetica, 1977, 20, 69-86.
Nahm, M. C. Selections from early Greek philosophy. New York:
Appleton-Century-Crofts, 1964.
270 CHAPTER 9
O'Hanlon, B. Strategic pattern intervention. Journal of Strategic and

Systemic Therapies, 1982, 1, (4), 26-33.
The Oxford English Dictionary, Vol. VII. Oxford: Clarendon Press,
1961, 450.
Quine, W. V. The Ways of Paradox and Other Essays (rev. ed.).
Cambridge: Harvard, 1976.
Selvini-Palazzoli, M., Cecchin, G., Prata, G., & Boscolo, L. Paradox and
Counterparadox. New York: Aronson, 1978.
Spiegel, H., & Linn, L. The "ripple effect" following adjunct hypnosis
in analytic psychotherapy. American Journal of Psychiatry, 1969, 126,
53-58.
Tomm, K. Clinical Applications of Strange Loop Theory, Paper
presented at the Brief Family Therapy Center's Distinguished Speakers'
Program, 1982.
Warner, J. Paradox and the eye of the beholder. Underground Railroad,
1982, 3, (4).
Watson, C. A Delphi study on paradox in therapy, 1982, unpublished
manuscript.
Watts, A. Psychotherapy east & west, New York: Pantheon, 1961.
Watzlawick, P., Beavin, J., & Jackson, D. D. Pragmatics of human
communication. New York: Norton, 1967.
Wilden, A. System and structure (2nd ed.). London: Tavistock, 1980.
10
Contradiction and Its Resolution
among the Psychotherapies:
Results of a Preliminary
Investigation*
b y Michael J. Bopp, Ph.D.
INTRODUCTION
Psychotherapy has generated a vast array of sophisticated m eth
ods and constructs. Currently, m uch is know n about the prom o
tion of hum an change. Yet pervading the field, there persists a
marked conceptual confusion. Am idst the m otley assortm ent of
ideas and m ethods, there has been little hope for unification. De
spite the current popularity o f "eclecticism ," there are still dog
matic, insular claims by proponents of various theories that their
own model is more "tr u e " than the others. There is also the sense
that what we do as psychotherapists is more an art than a science,
and that we have not yet reached the status of "real doctors"
(Berne, 1966).
This lack of coherence reflects the preparadigmatic phase of sci
ence, as evidenced by: (1) the presence of num erous com peting
perspectives; (2) the sense that the field, like that of physical op
tics before Newton, can be characterized as "som eth in g less than
scien ce" (Kuhn, 1970, p. 13); (3) the continued need to justify
prem ises each tim e a new claim is made rather than take them for
granted (Kuhn, 1970).
*The author wishes to thank Sam Kirschner, Dierdre Kramer, and Karyn Scher
for suggestions and constructive criticisms.
271
272 CHAPTER 10
Earlier in the development of psychology in general and psy

chotherapy in particular, efforts were made to fit facts to the New
tonian framework. Such attem pts have met with quite limited
success (Buss, 1975, 1979; M atson, 1964; Overton & Reese, 1973;
Ratner, 1971; Reese & Overton, 1970; Rychlak, 1968, 1976). Subse
quently, the general systems outlook becam e influential (e.g., Ber-
talanffy, 1968; Miller, 1975; Watzlawick et al., 1967; Watzlawick et
al., 1974). At least partially an outgrowth of the influence of the
systems approach, there has been a recent proliferation of philo
sophical activity, particularly within family therapy (e.g., Keeney,
1982; Keeney & Sprenkle, 1982; Liddle, 1982). W ith such work
still in the divergent stage of concept generation, there is if any
thing less conceptual consolidation than before (Coyne, 1982).
A premise of this chapter is that while still far from consoli
dated, a paradigm is in fact em erging within social science. This
paradigm is based on dialectical philosophy. It is not new, but
dates back to ancient Greece. According to Rychlak (1968), an on
going oscillation has existed betw een demonstrative (empirical,
reductionistic, analytic) and dialectical (holistic, synthetic) world
views over the course of Western scientific history. Particularly
with the conceptual shifts of the 20th century, physics and the
other physical sciences have been moving away from Newtonian
demonstrative m etaphors toward ones more closely affiliated with
dialectical notions such as contextualism, holism , relativity,
pluralism, and paradox (Koplowitz, 1978; Oppenheim er, 1956;
Riegel, 1973; Schrodinger, 1967; Sinnott, 1981; Zukav, 1979).
In social science in general, num erous developments mark a
similar progression (Buss, 1979; Riegel, 1973, 1979). W ithin the
particular branch of social science called psychotherapy, a crucial
development has been the increasingly active interest in phenom
ena that fall under the rubric of contradiction and paradox. O f
course paradox is im portant in the context of this book for its
technical relevance to the promotion of therapeutic change. Yet it
has far-reaching consequences on a metatheoretical level.
Paradox implies a type of illogic, a contradiction. It pertains to
two (or more) apparently opposed statem ents regarded as simul
taneously true. Taken together the two statem ents seem absurd;
what makes them a paradox is that they also seem true. A para
dox is experienced as unsettling and provocative because it
breaches a fundamental assum ption of our understanding. This
assum ption is so deeply ingrained as to be taken for granted; it is
a facet of "com m on sense."
To grasp why a paradox is so unsettling requires that one strug
gle to reveal just what assumption is being violated. That assum p
tion is the Aristotelian premise of noncontradiction, namely that
C O N T R A D IC TIO N A N D ITS RESOLUTION 273
an assertion and its negation cannot both be true (Aristotle, 1952;

Riegel, 1973). In Western culture this is a perceptual invariant: A
desk is just that, a desk. Those who would view it otherwise we
pronounce to have lapsed from "reality testing" and to merit the
terms "delusional or hallucinatory" (Arieti, 1955).
Returning to metatheory, that which is so interesting about
paradox in social science is that its pervasive relevance is forcing
us to explore some conceptual unsettlem ents (anomalies, in
Kuhn's terms). Indeed paradox has drawn us to reflect upon
deeply embedded assum ptions.
Outside the laboratory or therapy room we would regard the
desk as just that. O ur conceptual inclination in professional prac
tice is to apply the same logic and think of an em otion, a stimu
lus, a behavior, or a thought as a singular, unipolar phenom enon.
Yet the further the psychotherapeutic field develops in its under
standing of such phenom ena, the more it encounters bipolar or
multipolar constructs and the less able it is to rely on implicit Aris
totelian categories. Consequently, making sense of double binds,
paradoxical injunctions, the contradictions of ambivalence, id-
superego oppositions, hypnosis, and primary process thought
generates the same unsettled feeling that would derive from
viewing a desk as both itself and som ething else simultaneously.
The more contradiction-laden concepts we reveal, the more pres
sure we feel to scrutinize our assum ptions.
In a way, the psychotherapeutic field can be said to be becom
ing delusional, seeing double or multiple m eanings where singu
larities had previously resided. Yet the current break with existing
cognitive structures does not indicate a regression, but rather a
step forward. In life-span developmental psychology, there is ac
cumulating experimental evidence that the hypothetico-deduc-
tive, Aristotelian logic of formal operations is not, as presumed by
Piaget (1967), the culmination of intellectual development. Instead
there is much to support the existence of a more mature, "p ost-
formal operations" stage, central to which is the capacity to lo
cate, appreciate, and integrate contradiction (Basseches, 1978,
1980; Clayton, 1975; Kegan, 1979; Koplowitz, 1978; Kramer, 1983;
Riegel, 1973, 1976; Sinnott, 1981). Analogously, psychotherapy's
current concern with paradox and contradiction suggests a reor
ganization to a higher stage of understanding.
That this reorganization is occurring is strengthened w hen we
consider that paradoxical formulations have been appearing not
just in one branch of psychotherapy, but across all of the major
ones (Andolfi, 1974; Ascher, 1980; Ascher & Efran, 1978; Bopp,
1983; Esterson, 1970; Frankl, 1975; Haley, 1963; Mozdzierz et al.,
1976; Raskin & Klein, 1976; Selvini-Palazzoli et al., 1978; Watzla-
274 CHAPTER 10
wick et al., 1974; Weeks & L'Abate, 1979. See Weeks & L'Abate,
1982, for overview and integration). This reorganization is quite
significant given the field's historical factionalism . Moreover, that
such convergence culm inates in a construct anom alous to conven
tional scientific logic suggests not only the possible em ergence of
a new order of understanding, but also that such a new order
m ay have the potential to integrate diverse theoretical and practi
cal perspectives. This claim gains additional support in light of
the fact that the logic of contradiction, which a dialectical outlook
applies to ontological problem s, is also relevant to epistemological
ones. That is, rather than trying to achieve a singular theoretical
account, the dialectical approach attem pts to synthesize contra
dictions am ong diverse perspectives at a higher level. Theoretical
oppositions persist while a more integrative "tr u th " is forged.
Paradox is not the only concept signaling the increasing rele
vance of dialectical m etatheory in psychotherapy. The following is
an alignment of dialectical concepts with theoretical contributions
in the literature that illustrate their application: (1) the premise of
bidirectional (interactive) or multiple causation (e.g ., Bandura, 1978;
Bertalanffy, 1968; Jung, 1961; Koplowitz, 1978; Langs, 1976, 1978;
Progoff, 1973; Riegel, 1973; Rogers, 1980 [b]; Stanton, 1980); (2)
the concern with organization and related notions o f system, context,
and structure (Boszormenyi-Nagy & Spark, 1973; M inuchin, 1974;
Speck & Attneave, 1973; Stanton, 1980); (3) the prem ise of the
ontological primacy o f directed (i.e., developmental) motion and change
(Boszormenyi-Nagy & Spark, 1973; Haley, 1973; Rogers, 1980 [b];
Weeks & Wright, 1979; (4) the premise of the ontological primacy o f
relations (Boszormenyi-Nagy & Spark, 1973; Esterson, 1970;
Langs, 1976; Rogers, 1980 [b]); and (5) the attention to the integra
tion o f change and stability, coined in dialectical term s, transforma
tion or "movement through form s" (Basseches, 1978, 1980; Watzla
wick et al., 1974; Weeks & L'Abate, 1982).
The goal of this chapter, like that of the study (Bopp, 1983)
from which it has been abstracted, is to attem pt a metatheoretical
clarification in psychotherapy. Its intention is to illustrate how the
psychotherapies m anifest a movement toward an overarching dia
lectical world view. The claim is not that this world view is fully
formed, or that there is a perfect alignm ent betw een theory and
underlying m etatheory in psychotherapy. Rather, this is a pre
liminary effort to achieve some congruence betw een what we do
as therapists and an appropriate, corresponding model of science.
W hile the Bopp study's concern was the presence of an overall
dialectical m etatheory among psychotherapies, the present chap
ter will focus only on those aspects of dialectical m etatheory ger
m ane to paradox, the concern o f this book. Thus we will exclude
consideration of other central categories in th e dialectic world

view, such as m otion and relationship. Instead, we will focus on
contradiction, the m etatheoretical underpinning of applied para
dox (Esterson, 1970; Weeks & L'Abate, 1982).
METHOD
The data for this study included a lengthy interview (1 to 2 XA
hours) with a prom inent psychotherapist/theorist from each of
the following four schools: hum anistic, cognitive-behavioral, fam
ily therapy, and psychoanalytic. The interviewees were Carl Rog
ers, P h .D ., Arnold Lazarus, P h .D ., Ivan Boszormenyi-Nagy,
M .D ., and Robert Langs, M .D ., respectively.
The interview transcripts were coded using the Dialectical
Schem ata framework (Basseches, 1978, 1980; Bopp & Basseches,
1981). This framework consists o f an inventory of moves in
thought or conceptual strategies which characterize various as
pects of dialectical thinking. It was compiled through a survey of
the types of thinking em ployed by philosophers like Hegel and
Marx, who are clearly associated with the dialectical framework.
Like the experimental approach of Piaget (1967) and Kohlberg
(1971), the procedure involved dem onstrating the logical organi
zation of the subjects' thinking. Like the transform ational gram
mar m ethod (Chomsky, 1968), its intention was to illuminate
"d eep structure" m odes of understanding as they interact with an
array of specific thought contents. It was also similar to the work
of Bandler and Grinder (1975) as it involved generating a model
that cuts across schools of psychotherapeutic practice despite
marked differences at the level of theory. In short, this study con
stituted a preliminary exploratory effort to dem onstrate the pres
ence of a com m on style of understanding based on dialectics.
Each interview was sem istructured and open ended insofar as
it revolved around five core questions chosen to generate discus
sion rather than lead to definitive "an sw ers." The questions were:
1. How do you explain how people change—in general, and by
m eans of psychotherapy?
2. W hat would you say is the goal of psychotherapy? How do
you establish that "c u re " or "su ccessful outcom e" has taken
place?
3. How would you explain psychopathology in nontheoretical,
com m on sense term s? Similarly, how do you understand the
etiology of psychopathology?
The following questions pertained to case material (either hy
276 CHAPTER 10
pothetical or actual) which the subject had presented as indicative

of successfully conducted therapy within his particular model:
4. W hat were the m ost im portant changes engendered in the cli
e n ts ) over the course of therapy? W hat defined success in the
treatment of this case?
5. W hat features of treatment were m ost responsible for its effi
cacy? How do you understand the cause of the successful re
sults obtained?
The questions were sent to subjects prior to the meeting.
In addition, the interview was interactive. The interviewer was
an active participant in dialogue in the form of posing questions,
introducing alternative viewpoints, and pursuing digressions
from the core topics that appeared useful for exploring implicit
cognitive and philosphical underpinnings. Finally, the interviews
were in depth in that the probing was thorough and com prehen
sive, intending to document not only the substantive content of
the interviewee's thinking, but also the general forms of cognitive
organization (Basseches, 1978).
As defined by Basseches (1978, p. 46), dialectic is "develop
mental transformation (i.e., developmental movement through
forms) which occurs via constitutive and interactive relation
ships." To think dialectically is to approach the process of under
standing from the vantage point of this definition. Basseches has
generated 24 schem ata representative of this approach. O ne such
schema involves the location of contradictions, while another per
tains to assertion of the existence of relations, the limits of separa
tion, and the value of relatedness. The 24 schem ata operate as an
organized whole of understanding, a world view. In the following
text, we will consider only those schem ata relevant to contradic
tion, the central concern of this chapter. The reader should bear in
mind that what follows represents only part of the type of think
ing employed by dialectical thinkers. Elaboration is contained in
Bopp (1983).
Definitions1
O ne approach to contradiction in dialectical m etatheory is to
recognize and describe thesis-antithesis-synthesis (T-A-S) movement. In
these terms, a thesis can be an idea, elem ent, or force and its an
tithesis is that which is opposite, excluded from, outside of, apart
from , or contrary to the thesis. The synthesis is an integration of
1From Basseches, 1978; Bopp and Basseches, 1981.

thesis and antithesis. Synthesis is usually more complex than

either of its predecessors because it includes both of them , brings
them into relation, and binds them . T-A-S movement may be cy
clical in that the synthesis becom es the thesis for a new antithesis.
That is, to the extent that a synthesis is not all-inclusive, that
which has been left out of it constitutes a second antithesis.
Relatedly, theses and antitheses can be regarded as correlative.
Applying this mode of understanding, one recognizes the correlativ-
ity o f a thing and its other. To view things as correlative, a thinker
sees them as relative to each other. Each is in some sense depen
dent on the other. This leads to the notion o f the interdependence o f
opposites. To assert that the concept of figure depends on that of
ground, and vice versa, is to appeal to this idea. The recognition o f
composition by interpenetrating opposites is another corollary of the
correlativity construct. To apply this strategy is to regard whatever
can be seen as a synthesis in term s of the thesis and antithesis
which com pose it.
W hen the "th in g s" under consideration are understood in for
mal terms (i.e., as structures or system s) then sources of contra
diction can be found either within or outside the form. Such
thinking is termed the location (or description o f the process o f emer
gence) o f contradictions or sources o f disequilibrium within a system
(form) or between a system (form) and external forces and elements anti
thetical to the system's (form's) structure. In other words, through
the application of this schem a, one attends to forms and seeks to
locate sources of disequilibrium either internal or external to
them. The understanding o f the resolution o f such contradictions in
volves a notion o f transformation in developmental direction. This
transformation is a "m etaform al" movement through form s, con
sistent with the definition of dialectic above.
The italicized phrases above are the nam es of Dialectical Sch e
mata. They represent facets of dialectical thought. In the follow
ing section, we will consider how these schem ata apply to think
ing about psychotherapy.
RESULTS
The Humanistic Tradition: Carl Rogers

Earlier in his career, Rogers (e.g ., 1959) tried to establish a con
ceptual foundation for his phenom enological orientation. At the
time, the empiricist perspective held prom inence in American
psychology and there was not sufficient conceptual coherence to
278 CHAPTER 10
hold forth an alternative position supportive of phenom enology

(Rogers, 1980 [b]). As discussed in the interview (Bopp, 1983) and
in Rogers (1980 [b]), num erous developments have taken place re
cently, particularly in the natural sciences, that have given sup
port to such an alternative position. In the interview, Rogers re
fers to findings from chemistry, biology, and physics implicative
of dialectical concepts such as pervasive complexity, ontological
interconnectedness, the interdependence of change and form ,
and the constructive role of consciousness in the shaping of scien
tific knowledge. A s he states in the interview, Rogers experiences
a sense of satisfaction now because he views these advances as
facilitating the metatheoretical fram e for w hich he had searched.
The following excerpt dem onstrates the application of the con
tradiction m etaconstruct, that is, it involves the location of contra
diction or sources of disequilibrium between a form and elements external
to the form's structure. Herein, psychopathology is described in
term s of an external contradiction betw een the self-concept (the
form ) and organism ic experience.
Rogers: I think that we build [defensiveness] in2 socially, that the child
.. .gradually builds a self-concept which is not matched by what the or
ganism is experiencing. And then, since they are trying to live by this
self-concept, that means they must defend themselves against some of their
organic experiencing. (pp. 9-10)3
Here, the self-concept is construed in terms of a structure that the
individual “builds."4 This internal form comes to be at odds with "organic
experiencing." In other words, the self-concept can integrate only certain
aspects of the person. These aspects therefore exist in contradiction to the
self-system. The threat to the equilibrium of the self-concept implied by
the organic experience is conveyed in the notion that there must be a
defense against that experience. That is, the defense is an aspect of the
self-concept's functioning whose purpose is to preserve the current level
of structural organization in the self-concept.
Rogers expands this line of thinking subsequently:
2italicized phrases in this and the following quoted passages represent

clear manifestations of particular dialectical schemata.
3Page numbers for this and the following quoted passages (unless
otherwise identified) refer to locations of the quote within the interview
transcript (Bopp, 1983)
4The formal aspects of the self-concept had been discussed earlier in the
interview and in published sources (Rogers, 1951,1959,1980 [a]). When
viewed from an external frame of reference, the self-concept is termed the
self-structure (Rogers,1959).
Rogers: I want to say in advance, I don't like the term "psychopathol

ogy," but I don't know of any other general area that covers it. Any
way, disharmony within the self, maladjustment, or psychopathology, is,
in my estimation, a fairly extreme discrepancy between the self-concept,
on the one hand, and what is actually being experienced. (p. 10)
In this passage, psychopathology is being defined in term s of dis
equilibrium (disharmony, m aladjustm ent). The source of this dis
equilibrium is the "d iscrep an cy " betw een the self-concept and the
experience which is antithetical to it.
The understanding of the resolution of disequilibrium or contradiction
in terms of developmental transformation is a related facet of dialecti
cal thinking. In effect, it provides the thinker with a w ay to grasp
the resolution of a problem located by attention to contradiction.
The following illustration of this m etaconstruct pertains to the
process of growth in psychotherapy.
Rogers: . . . as one becomes more integrated, more whole, one also de
mands more of relationships, demands more of life, and so there are
new crises at a higher level of complexity I've been very much fascinated
by Prigogine’s work—the Belgian scientist, chemist—and I think that his
notion of the process of change is quite fitting to psychotherapy. That, in a
chemical molecule or in an organism, as the perturbations increase, change
is becoming necessary, it's forced. And that change is to a higher degree of
complexity where such complex situations are the most liable to change,
.. .change is more an expectation,. . . It means that, yes, you reach a
new level of integration. And so does that mean that you can stop
growing? No, it means that the new challenges will be of a different order
and that you will in time produce new perturbations which will produce
more change. (pp. 15-16)
This passage is an outgrowth of an earlier discussion about recon
ciling discrepant aspects of self. Rogers is stating here that the
resolution of an earlier contradiction ("a s one becom es more inte
grated, more w h o le") generates its own em ergent contradictions
("n e w crises at a higher level of com plexity"). In other words, the
solution to one problem is the basis of a new one. This resolution
process is synthetic. The discrepancies from earlier contradictions
are not solved, in the sense of rejecting thesis in favor of antithe
sis (or vice versa), but are integrated within a new, higher level of
organization. This movement to the next level is a movement
through form s, or dialectical transform ation. Moreover, in Rogers'
view and consistent with dialectical form ulations, this process is
quite developmental, as suggested by these phrases: "m ore inte
grated," "m o re w hole," "h ig h er level of complexity," "n e w chal
lenges of a different order." Particularly interesting in terms of
280 CHAPTER 10
this developmental aspect of dialectical thinking is the notion that

the transformation process, while providing greater stability
("m ore integrated," "m ore w h o le"), paradoxically generates
greater movement ("ch ange is becom ing necessary, it's forced").
Let us turn now to another aspect of dialectical thought rele
vant to the contradiction heuristic, namely the assertion o f the corre
lativity o f a thing and its other. Applied to the hum an condition, this
metaconstruct has important implications. Rogers has consis
tently maintained the position that one's self becom es known as a
function of other (Rogers, 1961, 1980 [b]). This is the priority of
em pathy in human development. By being able to temporarily let
go of one's self and experience the other's reality (i.e., the process
of empathy), it becom es possible for each person to realize that
that which at first appears "oth er," that is, not self, is also self.
That is, it becom es possible to experience how one's self and an
other's self are the same.
This is one aspect of correlativity, namely, the notion of inter
penetrating opposites: Though A and not-A from one perspective
seem antithetical, from another A is not-A as not-A is A.
Rogers: . . . it depends on empathy. You could tell me about your reality.
Unless I am willing to try to sort of let myself go and understand it from
inside, your telling me about it may not produce much of any effect.
It may have a little usefulness, but to really experience a shared reality, I
have to try to understand yours from the inside and you have to understand
mine from the inside. (p. 18)
.. .it means letting myself go, knowing that I can come back to myself.
(P- 19)
Interviewer: . . . So that to the extent that I can empathize with you,
then I am you and I am yet myself.
Rogers: That's right, uh-hm. Buber's I-Thou relationship. (p. 20)
Starting from self, one becom es other ("to try to understand
[your reality] from the in side") and vice versa, and returns to self
with a clarified and enlarged reality. Self and other, in the context
of empathy, become the same while retaining their respective
uniquenesses ("know ing that I can come back to m yself"). Further,
self and other paradoxically becom e more defined as self and other
by virtue of realizing their sam eness. This is the notion of the inter
dependence o f opposites, another aspect of correlativity. Referring to
change in psychotherapy, Rogers states:
Rogers: .. .as I understand you from inside, you become more able to
understand yourself (p. 22)
In other words, an outgrowth of em pathic com m unication is a

clearer delineation of self (which of course applies also to other).
Self could not exist without other, nor other without self.
Pairs such as thing and its other, self and not-self, and self and
other are instances of antithesis. Turning now to the conceptual
strategy, the recognition and description o f thesis-antithesis-synthesis
movement, we can derive a clearer understanding of the synthetic
nature of dialectical change. The following passage occurred in
the context of the discussion from which the correlativity excerpts
above were drawn:
Rogers: Suppose you hold some values that are very different from my own.
Our tendency is to say, "Well, you're bad then, or you're different,"
or something. If I can let go of my values for a moment so as to really see
what your values seem like inside, then that will enlarge my reality and I
can return also to the values that I hold. I probably will return a little
bit changed from having seen someone else's values differently be
ing willing to step out of myself for the moment out of knowing that
I can return within that self. That's why it needs, ideally, it needs a
fairly secure individual to do that, to let go of self and really see what
it seems like from the other side. (p. 19)
In the opening line, Rogers establishes a thesis-antithesis relation
ship in terms of values: "Su p p o se you hold some values [antithe
sis] that are different from my own [thesis]." The relationship be
tween the two sets of values is antithetical in the sense of an ap
parent mutual exclusivity.
Rogers then points to the tendency to invalidate the other's val
ues, consistent with the logical formulation, A or not-A: "O u r
tendency is to say, 'Well, you're bad then, or you're different,' or
som ething." His thinking then attends to the notion of correlativ
ity ("se e what your values seem like in sid e"), which leads to inte
gration ("w ill enlarge my reality").
This resolution is not linear. It does not consist of an exclusion of
one set of values as "b a d " in favor of the other, presum ed good;
nor is it a compromise or alloy of the two values. The assertion that
this process requires a "se cu re " individual implies that the em
pathizing person remains well defined unto him self, though none
theless "ch an g ed ." O n the contrary, the resolution entails a syn
thetic integration at a higher level. The two sets of values remain
polarized (rather than com promised), but now under the rubric of
a more integrated and differentiated "en larg e[d ]. . . reality." Appar
ently unrelated prior to synthesis, they are now brought into rela
tion in terms of their correlativity. Note also the ongoing presence
of contradiction in the resolution as if: I remain the same while yet I
have changed. Finally, from the words change and enlarge my reality,
282 CHAPTER 10
we can also infer that the movement in this thesis-antithesis-syn-

thesis process is developmental.
The Behavioral Tradition: Arnold Lazarus

The use of a logic of contradiction in the Lazarus interview was
prom inent in Lazarus's thinking about the case of M s. A, a client
whom he regarded as successfully treated by m eans of multi
modal therapy. Before considering this material, an im portant
caveat m ust be discussed. M ade quite clear in Lazarus's recent
writings (1981) and in the interview from which the following pas
sages were extracted, the premise of technical pluralism is crucial
in this model. That is, Lazarus does not advocate using the same
psychotherapeutic approach with each client. Rather, based on a
multimodal assessm ent across seven dim ensions (behavior, affect,
sensation, imagery, cognition, interpersonal functioning, biol
ogy), the therapist selects treatm ent m ethods specifically tailored
to the needs of each client. Thus while his thinking and practice
with M s. A may be characterized as representing features of
dialectic metatheory, he could proceed m uch differently with an
other client.
Let us turn now to the first passage.
Lazarus: ...s h e couldn't tolerate ambiguity was really one th in g ...so
that there's much more of a toleration-of-differences view. We call it a
kind of relativity notion, you see, that she has acquired instead of an
absolutistic idea of rightI wrong, good/bad. The dichotomy has now become a
continuum. That's the major kind of cognitive restructuring that I would
talk about. (pp. 47-48)
This passage illustrates that m ode of dialectical thinking which

follows along thesis-antithesis-synthesis lines. Lazarus describes the
client as originally unable to "tolerate ambiguity." This absolutis
tic approach was based on thesis-antithesis "d ich otom [ies]" such
as right/wrong and good/bad. The implication is that M s. A's in
ability to tolerate ambiguity was based on an either-or type of
thinking through which antitheses (right/wrong and good/bad)
were irreconcilable ("abso lu tistic").
Lazarus asserts that M s. A made progress in therapy w hen she
acquired a "toleration-of-differences view," such that she could in
tegrate dichotomous perspectives within a synthetic w hole. The
idea of bringing into relation apparently mutually exclusive op
tions (antitheses) is conveyed in the words: "T h e dichotomy has
now becom e a continuum ." That is, the fact that they can now
coexist along a com m on dim ension, a continuum , m eans that
they have been brought into relation whereas previously they had
been distinct. Such integration is also transform ational. If we re
gard cognitive structure as a form and its “ restructuring" in a pos
itive direction as a movement through form s, then we have an
instance of the metaformal aspect of dialectical thinking through
which the thinker seeks to com prehend transform ation.
The preceding excerpt dem onstrated an attention to contradic
tion and its resolution in Lazarus's thinking about the client's
cognitions. Turning now to the next excerpt, we can observe a
similar m ode of understanding applied to the therapeutic interac
tion. Continuing his discussion of the material cited above, Laz
arus stated that "th ere was an all-or-none that ju st changed into a
whole continuum of time and effort and m eaning and valu e" (p.
49). In response to the interviewer's question of how he as M s. A's
therapist effectuated such progress, Lazarus states:
Lazarus: Challenging all the time, pointing out "Look what you are do
ing again and again and again. What can you do? Let's get the op
tions, let's get the alternatives. There is not one way of viewing it. It is not
either right or wrong. This idea of you are either for me or against me
is ridiculous. There are many, many possibilities. Let's get you to look
at them. What are they?" This was a very important part of the ther
apy, the constructive altemativism, if you will. (p. 49)
In this passage, Lazarus is em ploying the dialectical concern for

interaction as a source of movement. The m ode of exchange be
tween him self and M s. A entails a drawing to the client's atten
tion of "op tion s," "alternatives," and "possibilities." A s the thera
pist, Lazarus establishes him self as an antithesis to M s. A . In
other words, he contradicts her contradictions or negates the ne
gation (Engels, 1940). In contrast to her outlook of "o n e way of
viewing it," and it's "eith er right or w rong," he is continuously
challenging. That is, he repeatedly poses the antithetical position;
for example, "T h ere are many, m any possibilities" "le t's get the
alternatives."
W hereas the first Lazarus passage cited above pertained partic
ularly to the cognitive dim ension, Lazarus subsequently asserts
(p. 50) that this m ethod of "constructive altem ativism " also ap
plied with regard to each dim ension of hum an functioning listed
above (behavior, affect, sensation, etc.).
The course of thinking described above is particularly interesting
metatheoretically. It provides a clear dem onstration of transform a
tional dialectical ideas by a theorist w hose roots are in the behavior-
ist tradition. This tradition has strong affiliations with reductionis-
tic, demonstrative (Rychlak, 1968) and m echanistic (Overton &
284 CHAPTER 10
Reese, 1981) assum ptions. Lazarus's line of thinking indicates a

shift from an exclusively linear, unidirectional, environmentalist
outlook toward one which can accommodate, at least in part, or-
ganismic, interactive, and paradoxical features. Such a progression
seem s to correspond to the increasing theoretical appreciation
among behaviorists of cognitive parameters in the regulation of
behavior (e.g., Bandura, 1978; M ischel, 1973). That is, by attribut
ing a measure of inner control to the organism, one establishes
metatheoretically an interplay between person and environm ent
where previously person had been regarded as passive and reactive
and environment as unidirectionally causative.
Moreover, with the introduction of cognitive events as explana
tory constructs, one is challenged to account for the num erous
contradictions inherent in subjective phenom enology (ironically,
these were among the conceptual problems that early behavior
ism had tried to circumvent by its staunch objectivism ). The
sources of these developments are of course complex. For a fuller
discussion, see Berman, 1978; Bandura, 1978; M ischel, 1973;
Staats, 1981.
In any event, Lazarus's thinking in term s of M s. A's therapeutic
gains and his role in promoting them underscores the presence of
attention to contradiction (and the broader transform ational con
structs of the dialectical world view) in at least one mode of cogni
tive behavior therapy. In particular, we have seen an effort to un
derstand a particular client's "psychopathology" in term s of unre
solved contradiction and an effort to prom ote therapeutic change
through a playing out of interpersonal contradiction betw een cli
ent and therapist. A fuller account of Lazarus's use of this mode
of understanding (Bopp, 1983) would illustrate his efforts to syn
thesize the contradictions am ong theories and therapeutic tech
niques which at the level of theory appear mutually exclusive
(Lazarus, 1967, 1976, 1981).
The Family Therapy Tradition:

Ivan Boszormenyi-Nagy
Family therapy, perhaps more than other schools of psycho
therapy, has moved a considerable distance toward the articula
tion of an elaborated dialectical m etatheory (Bopp & Weeks,
1984). In particular, Boszormenyi-Nagy has been quite explicit in
his appeal to dialectical m etaconstructs in the construction of his
family model (Boszormenyi-Nagy, 1965 [a], 1965 [b]), giving rise to
"dialectical intergenerational family therapy" (Boszormenyi-Nagy
& Spark, 1973), a predecessor of the current "contextual family
therapy" (Boszormenyi-Nagy & Ulrich, 1981). Thus the interpre

tation of the Boszorm enyi-Nagy interview was unique insofar as
less translation was required to get from theory to underlying
metatheory: W ith Boszorm enyi-Nagy the dialectical m etatheory is
quite explicit.
Leading into the following passage, the interviewer had asked
Boszormenyi-Nagy to explain the logic he had used earlier in re
ferring to an "internal contradiction."
Boszormenyi-Nagy: Well, Hegel's dialectical thinking is the most ad

vanced model as far as I can think of for this and I try to make some
connections there in one of the chapters of Invisible Loyalties. So then
it's inevitable that every thesis can be considered from its opposite, so the self
from the side of the nonself or the other. And then it has to be placed in some
kind of synthetic transcendence of the antithesis. That's the point. So there’s
nothing tragic about built-in antithesis and deal with it. How to transcend
it, that's the problem. That's the way. That's the way. Not just the problem,
but that's the way. (pp. 11-12)
Here we have a clear instance of attention to contradiction. In the

sentence, " S o then, it's inevitable that every thesis can be consid
ered from its opposite, so the self from the side of the nonself or
the other," Boszorm enyi-Nagy is pointing out that corollary of at
tention to contradiction termed the correlativity o f a thing and its
other. That is, he is explaining the constitution of a contradiction
or antithetical relationship by asserting that a particular thesis can
be understood from a consideration not just of itself but also of its
opposite, the nonself or other.
This formulation is a central facet of the dialectical world view.
Contrary to the Aristotelian prem ise of "identity," by which a
thing is seen as having a singular, intrinsic nature, this formula
tion asserts that things are constituted as a function of their rela
tions to other things. That is, self does not have an independent,
monadic, ontological status, but becom es known as a function of
its relation to not-self or other. W hen this metatheoretical as
sumption is applied on a theoretical level, it has m ajor utility in
the understanding of personality development.
Now one is led to understand psychic functioning not only on
the basis of internal dynamics, but also in term s of how these dy
namics are played out interpersonally. W hile this insight may ap
pear obvious currently, it is useful to recall that the shift from an
instinct-based personality theory to a more interpersonal outlook
occurred only late in Freud's career and marked a conceptual ad
vance in psychoanalytic theory w hen elaborated by members of
the object relations school (e.g., Guntrip, 1956, 1971). Such think
286 CHAPTER 10
ing is also at the core of both Boszorm enyi-Nagy's model of family

therapy (as discussed presently) and Rogers's model of client-cen-
tered therapy (discussed earlier).
Returning to the interview passage above, we have ju st consid
ered Boszorm enyi-Nagy's establishm ent of the interdependence of
a thesis and antithesis. Following this, his thinking attends to the
process of movement toward resolution of antithesis. In other
words, he applies an intellectual strategy called the recognition and
description o f thesis-antithesis-synthesis movement. This sequence of
thoughts requires little interpretation as Boszorm enyi-N agy asserts
the need for a "synth etic transcendence of the antithesis." M ore
over, consistent with dialectical assum ptions, the contradictoriness
of an antithetical relationship is regarded as valuable, though prob
lematic: " S o there's nothing tragic about built-in antithesis-----
How to transcend it, that's the problem . That's the way." N ote the
persisting contradictory complexity of the transcendence's being
"N o t just the problem, b u t. . . the way." W hile a resolution to the
earlier antithesis, the "synth etic transcendence" still contains the
problem; it is sim ultaneously a problem and a solution.
The mode of thinking has direct applicability to the process of
contextual therapy. The following excerpt em anates from a
discussion of the treatm ent of a delinquent. The author had asked
Boszormenyi-Nagy to explain how in his model such a client
would have "en titlem en t" while being blam ew orthy for destruc
tive behavior.
Boszormenyi-Nagy: That’s the internal contradiction. Therefore, the thera

pist, the contextual therapist, has a dual approach to it. In one sense he
certainly doesn't like destructive relating in the current behavior. On the
other hand, he is willing to [trace]? partiality, which means empathy plus
crediting, to where it belongs. That means the past relationships of this
person as a victimized child. And on that level there is a great deal of
willingness to be partial without reservation from that point of view.
Now that I have been partial to you on that basis, now I invite you to
look at how you can be unjust to others, and at that point, the person
can hear me much better because I have been partial to his side. (p. 10)
This passage refers to the "revolving slate" notion (Boszormenyi-

Nagy & Spark, 1973). That is, the individual's currently destruc
tive behavior is an outgrowth of an injustice earlier inflicted on
him . Prior to this passage, Boszorm enyi-Nagy had criticized the
traditional individual therapy model for focusing exclusively on
the pathological, that is, negative aspects of the individual.
In contrast, the contextual m odel seeks to reveal the internal
contradiction. W hile there is a need for the delinquent to be held
accountable for m isdeeds ("blam ew orth y"), there is also a need

for partiality toward his side; that is, this individual is "en titled "
to "em pathy plus crediting." By thus addressing the original vic
timization, the subm erged contradiction is drawn out, whereas in
a traditional therapy model only one of its poles (the pathology)
presumably would be addressed. Such a "d u al approach" in con
textual therapy rests explicitly on the heuristic category o f contra
diction.
A further im plication of Boszorm enyi-N agy's thinking for the
m etaconstruct of contradiction pertains expressly to epistemology.
Given that interpersonal system s are com posed of two or more
individuals, contradictions will be apparent in their construction
of reality since the individual's perspectives will never be identi
cal. W ith respect to the dyadic context, pressure is exerted on the
theorist to devise a way to grasp a shared construction of "real
ity " despite contradiction betw een the two parties. This is dem on
strated in Langs's concept of the "bipersonal field " (Langs, 1976)
and Rogers's concept of em pathy as "sh ared reality" (Rogers,
1980 [b]).
W hen multiple parties com prise the social context, as in the
case of a family, the conceptual dem and is greater. A n outgrowth
of this problem is a recent proliferation of work in the family ther
apy literature to devise a sophisticated epistem ological model (see
1982 editions of Family Process). Boszorm enyi-Nagy addresses this
issue in the form of a "m ultilaterality of fairness" (interview,
Bopp, 1983, p. 25; Boszorm enyi-Nagy & Ulrich, 1981, p. 25).
Analogous to Rogers's effort to achieve a shared reality with the
client, Boszormenyi-Nagy, through the application of this con
cept, strives to be "p artial" to the perspective of each m ember of
the social system involved. To the extent that each person has a
unique and valid perspective, the family therapist encounters an
array of epistemological contradictions. W ith the integration of
each successive perspective into a broader multilateral under
standing, there has been a resolution of contradiction and a corre
sponding epistemological transform ation. This m ode o f under
standing is far removed from unipolar, reductionistic epistem olog
ical prem ises.
The Psychoanalytic Tradition: Robert Langs

The presence of dialectical m etatheory in psychoanalysis has
been detailed in several sources (Fenichel, 1967; Rychlak, 1968,
1976; Stierlin, 1969). Yet the connections betw een the two intellec
tual traditions (psychoanalysis and dialectics) are uneasy, due at
288 CHAPTER 10
least partially to a confusion in psychoanalytic theory vis-^-vis

goals of an ultimate reductionism (Rychlak, 1976). As touched on
briefly above, however, it is in Freud's and his followers' work
that invaluable contributions to the understanding of hum an con
tradictions were developed. Among such phenom ena are inter
plays between: id and reality, unconscious and conscious, pri
mary process and secondary process, eros and thanatos, reality
and pleasure principles, and good and bad object representations.
Through psychoanalysis, we have learned of bipolar crises of de
velopment (Erikson, 1963) and of the "sy n th etic" functions of the
ego (Nunberg, 1960).
The metatheoretical strategy underlying such constructs is il
lustrated in the words of Freud:
We find, then, that certain among the impulses to perversion occur
regularly as pairs of opposites and this,. . . has a high theoretical
significance.
An especially prominent part is played as factors in the formation

of symptoms in psychoneuroses by the component instincts which
emerge for the most part as pairs of opposites----
(Freud, 1962, pp. 50, 58)
Robert Langs has contributed a great deal toward clarifying and
extending the Freudian mode of treatm ent. From his work in this
area a num ber of constructs pertinent to the present discussion of
contradiction in psychoanalysis can be brought to light. In partic
ular, through discussion of the following excerpts, we will con
sider the applicability of the contradiction m etaconstruct to: (1)
the constitution of psychopathology as "m ad n ess" (Langs, in
preparation); (2) the functioning of the psychotherapeutic frame;
and (3) the nature of therapeutic com munication.
The first instance of an appeal to the notion o f contradiction is quite
explicit. In the interview from which the following passages were
extracted, Langs repeatedly associated his concept of m adness
with the term contradiction. W hen asked to elaborate he states:
Langs: I think subjective madness is to experience contradiction, an unresolva-
ble contradiction: I'm alive but I'm gonna be dead, I was bom but I
was bom to die, I'm thinking this thought but now I'm thinking an
other thought, I'm thinking an opposite thought,. . . bearable contra
dictions lead to madness.. . . Unbearable contradictions create craziness and
madness. Subjectively experienced unbearable contradiction is subjectively
experienced madness. It really does drive you crazy when somebody gives you
contradictory messages and then when you internalize them you begin to feel
crazy. (p. 55)
This is a particularly straightforward specification of contradic

tion. Langs states that bearable contradiction "lead s to " m adness
while unbearable contradiction "crea tes" m adness. He also gives
examples of unresolvable contradictions such as, " I was born, but
I was bo m to die." He further asserts that contradiction can occur
interpersonally, as in the case of "som ebody giv[ing] you contra
dictory m essages" or intrapersonally, as w hen such m essages be
come internalized, leading one to "begin to feel crazy."
A second instance of appeal to contradiction occurs in regard to
Langs's concept of the therapeutic frame, defined as " a metaphor
for the implicit and explicit ground rules of psychotherapy or psy
choanalysis [through which to] create a basic hold for the thera
peutic interaction" (Langs, 1982, p. 726). The m anagem ent of the
frame, in term s of set time, set place, set fee, confidentiality, rela
tive anonymity of the therapist, and so on is considered indis
pensable to the proper conduct of effective psychotherapy (Langs,
1976, 1978, 1982).
In dialectical terms, the frame is a form construct that describes
a set of physical and psychological boundaries and contains in
ternal contradictions. W hen the frame is secure, it facilitates the
experience of m adness by both patient and therapist, defined
above as internal contradictions such as being born to die. It is
imperative in such a context that the therapist manage his own
m adness, that is, successfully tolerate contradictions so as not to
impose them onto the therapeutic field. To the extent that this is
accomplished, suitable conditions are thus provided for the thera
pist to interpret and synthesize and for the patient to achieve
higher levels of integration.
The excerpt below illustrates the location of internal sources of
disequilibrium within the frame.
Langs: What happens when you secure the frame is that they react now
to the meanings of the secure fram e.. . . What are the problems of the
secure frame? That's what is so fascinating. This is where Freud's use
of the couch came out so beautifully. The secure frame says you have
to be here every time at the appointed hour. So the secure frame en
traps you. The secure frame, which I believe includes the couch says,
"You cannot look at me." So there are interpersonal deprivation and sepa
ration issues. The secure frame creates depressive anxieties because you
have an object loss in not being able to look at the therapist. It creates para
noid and phobic anxiety because you're entrapped and you are restricted and
you’re, again, with someone who is also capable of securing the frame. Pa
tients recognize that. It's a very powerful capacity, and it makes you
very threatening. If you're that strong, you know, will you then turn
against me? (p. 35)
290 CHAPTER 10
This passage originated in the interviewer's asking w hether there

would still be issues for a patient to bring into the field if the
frame were m anaged adequately; or, if there were no frame devia
tions evoking conflict in the therapeutic dyad. Langs's response is
to specify the various internal contradictions w ithin the secure
frame. That is, with frame m anagem ent issues out of the way, the
patient is left to deal with sources of disequilibrium that are in
trinsic to being within the frame. These include the experience of
"entrapm ent," "interpersonal deprivation and separation issues,"
"paranoid and phobic anxiety," and the threat of the therapist's
"pow erful capacities."
Further, consider the paradoxical qualities of the fram e. First,
the frame sim ultaneously provides gratification (through en
couraging symbiosis, self-reflection, interpersonal security) and
frustration (e.g ., "interpersonal deprivation and separation").
The tension betw een these two poles opens the interactive field
for the patient and therapist to engage in the exchange of symbols
requisite to insight, understanding, and growth. To shift too far in
the direction of frustration would be to prom ote disillusionm ent
in the patient (discussed elsewhere in the interview), while to
shift too far toward gratification would im pede the patient's toler
ation for ambiguity, conflict, and pain, without which growth
could not occur.
A second, related paradox follows. W hile serving as the sine
qua non of "cu re," the secure frame also activates the patient's
m ost frightening anxieties and experiences of m adness. In this
sense, a solution is sim ultaneously a problem . W hile the experi
ence and expression of m adness are the chief requirem ents for
"cu re," they contain the threat of severe disorganization.
Finally, consider Langs's concept (1978, 1982) of the me/not-me
interface. Recall that according to the Aristotelian law o f noncon
tradiction, a "th in g " cannot be both itself and som ething else si
multaneously. Recent advances in the physical sciences have re
peatedly violated this tenet, as in accounts of physical phenom
ena that cannot be characterized in unipolar term s (e.g ., as either
m atter or energy). In Freud's original work, this m ode of under
standing appeared clearly in the concept o f overdetermination
through which dreams and sym ptom s are viewed as a con
vergence of associations, m eanings, and im pulses. The signifi
cance of such a heuristic strategy lies in the fact that it transcends
the quest for a monadic understanding of basic scientific phenom
ena in favor of a view of dual or multiple constitution.
The me/not-me interface construct illustrates this m ode of
thinking particularly well.
C O N T R A D IC T IO N A N D ITS RESOLUTION 291
Langs: AH I mean by that is that when a patient is in therapy, every associa

tion, everything they do and say, every communication has some unconscious
connection to a perception of the therapist, to an image of the therapist based
on the therapist's intervention. That's the not-me part of the interface. On the
other side, the same communications are also self-perceptions. And often
unconscious self-perception. I'm talking usually about encoded mes
sages. The same communication will encode a perception of the therapist and
of the patient. Himself or herself. So the me/not-me interface just simply
meant that every communication faces both ways. (pp. 8-9)
This is a particularly cogent example of the interpenetration o f oppo
sites aspect of dialectical thinking. In this sense, Langs states that
all of a patient's activity in therapy ("every association, everything
they do and say," etc.) constitutes an intermixture o f perceptions,
of the therapist and of the patient. That is, the patient's activity
represents a synthetic integration. Langs traces this out in his
thinking. Encoded in "every com m unication" is a perception of
the therapist, the not-m e, and a self-perception, the m e. Com m u
nications therefore are certainly not simple and direct but rather
quite complex, consistent with the definition of synthesis. Given
this synthetic quality, therapeutic com m unications consist of in
terpenetrating opposites. Stated differently, the "w h o le " of the
patient is represented in the singularity of the m om ent. M eta-
theoretically, this m ode o f understanding illustrates an apprecia
tion of epistemological contradiction and of the dialectical concept
"m an y-in -one" (Rychlak, 1976).
CONCLUSIONS AND DISCUSSION

The line of argum ent developed in this chapter is summarized
concisely in the following passages from Invisible Loyalties (Boszor-
m enyi-Nagy & Spark, 1973).
We propose that the understanding of the structure of a relational
world requires a dialectical rather than absolute or monothetical
way of thinking. The essence of the dialectical approach is a libera
tion of the mind from absolute concepts which in themselves claim
to explain phenomena as though the opposite point of view did not
exist. According to dialectical thought, a positive concept is always
viewed in contrast with its opposite, in the hope that their joint
consideration will yield a resolution through a more thorough and
productive understanding. The principles of relativity and indeter
minancy in physics and the concept of homeostatic regulations of
living things are examples of increasingly dialectical orientation in
natural sciences, (p. 18)
292 CHAPTER 10
* * * * *
Psychology, psychotherapy and psychopathology have also been in

a gradual transition toward a more dialectical viewpoint, (p. 19)
In this chapter, we have considered evidence of this position in

the form of interview data with four prom inent psychotherapists
of diverse traditions. The most im portant observation was their
common appeal to contradiction in expressing their com prehen
sion of major psychotherapeutic issues. W ith Langs and Boszor
menyi-Nagy, the reference to contradiction was explicit. In the
Rogers and Lazarus transcripts, the m etaconstruct appeared not
at the level of theory, but in the form of their logic. The salient fact
for each was that the contradiction category was applied with ref
erence to both intra- and interpersonal features of psychotherapy.
The intrapersonal contradictions were: the self-system and unas
similated experience (Rogers); dichotomous thinking, perception,
and so on, leading to failure to tolerate ambiguity (Lazarus); the
internal contradiction of blam eworthiness and entitlem ent (Bos
zormenyi-Nagy); and the patient's experiences of m adness (de
fined as contradiction) intrinsic to the secure analytic relationship
(Langs).
Interpersonally there was: the correlativity of self and other in
the process of em pathy (Rogers); the contradicting of M s. A's in
ternal contradictions—the negating of negation—via the therapist-
client interaction (Lazarus); the contextual therapist addressing
not only the delinquent client's accountability for misdeeds but
also his unmet entitlem ent to restitution for wrongdoings perpet
uated on him, the second pole of contradiction (Boszormenyi-
Nagy); and the analyst's curtailment of personal "m ad n ess" to
promote synthesis of the patient's m adness via the bipersonal
field (Langs).
The concern with epistemological contradiction was also salient.
With Rogers, particularly careful attention is given to the achieve
ment of a "shared reality" with a client. He is strongly concerned
with stepping outside of his own perspective to adopt that of his
client, the purpose being to differentiate and further define each
party's knowledge (defined in experiential terms) of self, other, and
the bonds between the two. In the Lazarus material, the client's
problem was construed in the explicitly epistemological terms of a
propensity for "absolu tistic" and "d ichotom [ous]" ideas. Using
his constructive alternativism m ethod, which consisted of "ch al
lenging," and presenting " o p tio n s .. .altern ativ es.. .possibilities,"
Lazarus strove to interrupt his client's usual mode of understand
ing and provoke growth to a more integrative level. In contrast to
"absolutistic ideas," this new level entailed a "toleration of differ

ences view."
Attention to epistemological contradiction and its resolution
was evident in Boszorm enyi-Nagy's concept, multilaterality of
fairness, through which the contextual therapist endeavors to
achieve partiality to the perspectives of each m ember of a family
system . This process entails transcendence of the therapist's own
biases and need-based predelictions in order to generate an inclu
sive, multifaceted, synthetic "tru th ." In Langs's work, this mode
of thought was represented clearly in the me/not-me interface
construct. That is, the analysand's activity at a given m om ent in
treatm ent represents perceptions of not only the self but also the
therapist. Included are contributions from a num ber of psycho
logical sources such as associations, affects, thoughts, images,
and so on, all culminating in the dynamic m om ent. Also relevant
to epistem ology in Langs's system is the analyst's effort to synthe
size the num erous contradictions that com prise the patient's expe
rience of "m ad n ess."
We noted in the Introduction that num erous psychotherapeutic
concepts are provoking an exploration of underlying conceptual
assum ptions. W ith respect to the present discussion of episte
mology, it is interesting to note how the utilization of an interac
tive or interpersonal approach to therapy leads to epistemological
considerations.
In the early days of psychotherapy the therapist, presumed
"cu re d " by a training analysis, was considered a blank screen for
the patient's projections (Langs, 1976, 1978, 1982). In that context,
linear, unipolar m etaconstructs are apropos. W ith the adoption of
an interpersonal model, however, the therapist enters the inter
active field as another hum an being with conflicts and limitations.
No longer an objective, mostly detached figure, the therapist has
a construction of reality (though presumably more adaptive and
managed constructively) which operates as an active epistem olog
ical force in the relationship. Now, rather than merely interpret
ing, the interpersonally oriented therapist m ust align his/her own
reality with that of the patient in order to prom ote the latter's
epistemological reconstructions.
W ith the development of the family therapy approaches, this
alignment and reconstruction process becom es more challenging.
Here it is no longer that one individual, the "identified patient,"
has lapsed from "reality testin g "; rather, a social system has
becom e dysfunctional, producing a scapegoat to em body the
family's internal contradictions. Moreover, because the family
therapist has refused to isolate one family m ember as "psycho-
294 CHAPTER 10
pathological," s/he now has the challenging philosophical tasks of

(1) taking seriously the identified patient's perspective, loss of
reality testing notwithstanding; (2) taking seriously the perspec
tive of each family m ember and the fam ily's aggregate under
standing of the current crisis; (3) synthesizing an overall under
standing that incorporates each family m em ber's perspective, in
cluding that of the ostracized individual; and (4) perform ing the
additional synthesis of integrating the previous synthesis with the
theoretical constructs of the therapy model and the assum ptions
and biases of the therapist as hum an being in process. Such chal
lenges, worthy of Socrates or Hegel, are an integral part of a ther
apist's day-to-day practice and provide insight into the currently
prom inent concern for epistemology, particularly am ong family
therapists. Moreover, given how deeply ingrained assum ptions of
noncontradiction are, it is no surprise that so m uch of what thera
pists think and practice is term ed paradoxical.
The study partially reported in this chapter was preliminary.
Obviously it suffers from the problem s inherent in the use of in
terview data and small sample size. Yet as an exploratory study, it
is useful in illuminating connections betw een theory and m etathe
ory in psychotherapy; further research and theory are clearly in
order. In closing, we will turn to the im plications that could fol
low from further empirical and conceptual support of this line of
inquiry.
Implications for Psychotherapy
M etatheory. O n a metatheoretical level, one implication of this

work is the suggestion that psychotherapy may be moving toward
an overarching integration. This im pending integration derives
from an exam ination and critique of philosophical presupposi
tions, especially the Aristotelian law of noncontradiction, and a
corresponding shift to reliance upon the relevance of contradic
tion. The data in this study, in addition to conceptual argum ents
from various sources (Andolfi, 1974; Ascher, 1980; Berm an, 1978;
Boszorm enyi-Nagy & Spark, 1973; Esterson, 1970; Fenichel, 1967;
Koplowitz, 1978; Mozdzierz et al., 1976; Raskin & Klein, 1976;
Rychlak, 1968; Staats, 1981; Stierlin, 1969; Weeks & L'Abate, 1982)
have indicated that that which is vital in both the theory and prac
tice of psychotherapy—across schools—is paradoxical. That such
convergence centers on a construct anomalous to traditional sci
entific m ethods yet basic to dialectical ones suggests that the over
arching integration is best accom m odated through a dialectical
approach.
C O N T R A D IC T IO N A N D ITS RESOLUTION 295
Although m uch therapeutic work in the past utilized contradic

tion and paradox, efforts were m ade to force explanations to fit
m echanistic, linear, and reductionistic term s. M ore recently, in the
face of severe difficulties with such endeavors, a retreat from
striving for conceptual coherence occurred by appealing to such
notions as eclecticism .
W ith ongoing attention being given to paradoxical approaches
and a corresponding articulation o f dialectic metatheory, however,
the m ethods and concepts of psychotherapy may be brought in
creasingly into phase. In other words, if psychotherapy has been
largely based on paradoxical m ethods, and if the appropriate sci
entific framework for such m ethods is a dialectical one, then the
upcoming era in psychotherapy may w itness to a newfound con
ceptual order deriving from the field's current variegation. Such
consolidation could in turn relieve the "a n x iety " experienced by
social scientists uncertain of the status of their fields relative to
the natural sciences, w hich have been becom ing considerably dia
lectical in outlook over the 20th century (e.g ., Sinnott, 1981).
O ne way in which this reordering could happen would be to
reframe the dogm atism and often contentious appeals to truth
that have occurred within psychotherapy. Rather than perpetuat
ing the view that only one theory can be correct, a dialectical out
look would prom ote the view that each theory is "tru e ," that is,
reflects a valid explanation ventured from a particular level of ob
servation. Moreover, the contradictions am ong the theories would
themselves be valued as they reveal the relative limitations of the
individual theories as well as those of the theory construction
process itself (a central problem being the historical failure to
achieve an encom passing, internally consistent model). Interthe
ory contradictions also signal the potential for new understanding
em erging from their syntheses.
Finally, this scientific approach would strive to multiply (rather
than minimize) theoretical perspectives as a way to achieve an in
clusive theoretical outlook (Basseches, 1978; Bopp & Basseches,
1981; Boszorm enyi-Nagy & Ulrich, 1981). Viewed in this way, the
development of psychotherapy as a whole, with all its variega
tion, represents the unfolding of a large-scale, conceptual enter
prise. W hat has been lacking, however, has been a frame within
which to consolidate this m osaic. The fram e best suited to this
end is a dialectical one.
Practice. At the practice level, there are also im portant implica

tions. First, em anating from the presupposition of contradiction
as central is the view that effectively conducted psychotherapy
would prom ote rather than obviate paradox within clients. The
296 CHAPTER 10
goal of psychotherapy would no longer be to achieve a state of

unconflicted "h e a lth " or "cu re." Instead, the therapy process
would consist of efforts to locate and polarize contradictions be
neath what m ight appear to be singularities. Further, such ther
apy would prom ote an ongoing state of dynamic tension formed
precisely out of contradictory experience. A n effort would be
made not only to foster toleration but resilience and valuing of
contradiction by the client (and certainly by the therapist as well).
This notion has been advanced by Riegel (1976, 1979) as a crite
rion of successful hum an development.
A second implication for clinical practice is to broaden our view
of paradox. In early work on this topic (e.g ., Frankl, 1975) para
dox was understood as a specific therapeutic technique. More re
cently, however, there has been a growing sense that many psy
chotherapeutic practices actually constitute expressions of implicit
intra- and interpersonal paradoxes—despite the name a particular
intervention has carried. In fact, the results of this study support
this view through the finding of implicit or explicit reliance on
contradiction m etaconstructs in diverse areas of psychotherapy
not associated with recent discussions of paradox. Extending this
reasoning then, one m ight postulate that far from being a new
technique, paradox has been in use all along. Further, there is a
need for more explicit articulation of the paradoxes already occur
ring under different nam es throughout the psychotherapeutic
field.
REFERENCES
Andolfi, M. Paradox in psychotherapy. American Journal of
Psychoanalysis, 1974, 34, 221-228.
Arieti, S. Interpretation of schizophrenia. New York: Robert Brunner,
1955.
Aristotle. Metaphysics. In R. M. Hutchins (Ed.), Great books of the
western world. Chicago: Encyclopedia Britannica, 1952.
Ascher, L. M. Paradoxical intention. In A. Goldstein & E. B. Foa
(Eds.), Handbook of behavioral interventions: A clinical guide. New
York: Wiley, 1980.
Ascher, L. M., & Efran, J. S. Use of paradoxical intention in a
behavioral program for sleep onset insomnia. Journal of Consulting
and Clinical Psychology, 1978, 46, 547-550.
Bandler, R., & Grinder, J. The structure of magic I: A book about language
and therapy. Palo Alto: Science and Behavior Books, 1975.
Bandura, A. The self-system in reciprocal determinism. American

Psychologist, 1978, 33, 334-358.
Basseches, M. A. Beyond closed-system problem-solving: A study of meta-
systematic aspects of mature thought. Unpublished doctoral
dissertation, Harvard University, 1978. (University Microfilms
International, 1979).
Basseches, M. A. Dialectical schemata: A framework for the empirical
study of the development of dialectical thinking. Human
Development, 1980, 23, 400-421.
Berman, D. S. Cognitive-behaviorism as a dialectical contradiction: The
unity of opposites. Human Development, 1978, 21, 248-254.
Berne, E. Principles of group treatment. New York: Grover, 1966.
Bertalanffy, L. General system theory. New York: Braziller, 1968.
Bopp, M. J. A study of dialectical metatheory in psychotherapy.
Unpublished doctoral dissertation, Temple University, 1983.
Bopp, M. J., & Basseches, M. A coding manual for the dialectical schemata
framework. Unpublished manuscript, 1981.
Bopp, M. J., & Weeks, G. R. Dialectical metatheory in family therapy.
Boszormenyi-Nagy, I. A theory of relationships: Experience and
transaction. In I. Boszormenyi-Nagy & J. L. Framo (Eds.), Intensive
family therapy. New York: Harper & Row, 1965. (a)
Boszormenyi-Nagy, I. Intensive family therapy as process. In
I. Boszormenyi-Nagy & J. L. Framo (Eds.), Intensive family therapy.
New York: Harper & Row, 1965. (b)
Boszormenyi-Nagy, I., & Spark, G. M. Invisible loyalties: Reciprocity in
intergenerational family therapy. Hagerstown, M d.: Harper & Row,
1973.
Boszormenyi-Nagy, I., & Ulrich, D. N. Contextual family therapy. In
A. S. Gurman & D. R Kniskern (Eds.), Handbook of family therapy.
New York: Brunner/Mazel, 1981.
Buss, A. R. The emerging field of the sociology of psychology.
American psychologist, 1975, 30, 988-1002.
Buss, A. R. A dialectical psychology. New York: Irvington Publishers,
1979.
Chomsky, N. Language and mind. New York: Harcourt, Brace & World,
1968.
Clayton, V. Erikson's theory of human development as it applies to the
aged: Wisdom as contradictive cognition. Human Development, 1975,
18, 119-128.
Coyne, J. A brief introduction to epistobabble. Family therapy networker,
1982, 6, 27-28.
298 CHAPTER 10
Engels, F. Dialectics of nature. New York: International Publishers, 1940.

Erikson, E. H. Childhood and society (2nd ed.). New York: Norton, 1963.
Esterson, A. The leaves of spring: A study in the dialectics of madness.
Middlesex, England: Penguin, 1970.
Fenichel, O. Psychoanalysis as the nucleus of a future dialectical-
materialistic psychology. American Imago, 1967, 24, 290-311.
Theory, Research and Practice, 1975, 12, 226-237.
Freud, S. Three essays on the theory of sexuality. New York: Avon Books,
1962.
Guntrip, H. Recent developments in psychoanalytical theory. British
Journal of Medical Psychology, 1956, 29, 82-99.
Guntrip, H. Psychoanalytic theory, therapy and the self. New York: Basic
Books, 1971.
Jung, C. G. Memories, dreams, reflections. New York: Vintage Books,
1961.
Kegan, R. G. The evolving self: A process conception for ego
psychology. The counseling psychologist, 1979, 8, 5-34.
Keeney, B. What is an epistemology of family therapy? Family Process,
1982, 21, 153-168.
Keeney, B., & Sprenkle, D. Ecosystemic epistemology: Critical
implications for the aesthetics and pragmatics of family therapy.
Kohlberg, L. From is to ought: How to commit the naturalistic fallacy
and get away with it in the study of moral development. In
T. Mischel (Ed.), Cognitive development and epistemology. New York:
Academic Press, 1971.
Koplowitz, H. Unitary thought: A projection beyond Piaget's formal
operations stage. Unpublished manuscript, 1978.
Kramer, D. A. Post-formal operations? A need for further
conceptualization. Human Development, 1983, 26, 91-105.
Kuhn, T. S. The structure of scientific revolutions (2nd Ed.). Chicago:
University of Chicago Press, 1970. (Originally published 1962)
Langs, R. The bipersonal field. New York: Jason Aronson, 1976.
Langs, R. The listening process. New York: Jason Aronson, 1978.
Langs, R. Psychotherapy: A basic text. New York: Jason Aronson,
1982. (a)
Langs, R. Cure through madness. In preparation.
Lazarus, A. A. In support of technical eclecticism. Psychological Reports,

1967, 21, 415-416.
Lazarus, A. A. (Ed.). Multimodal behavioral therapy. New York: Springer,
1976.
Lazarus, A. A. The practice of multimodal therapy: Systematic,
comprehensive and effective psychotherapy. New York: McGraw-Hill,
1981.
Liddle, H. On the problem of eclecticism: A call for epistemological
clarification and human-scale theories. Family Process, 1982, 21,
243-250.
Matson, F. W. The broken image: Man, science and society. New York:
Braziller, 1964.
Miller, J. G. General systems theory. In A. M. Freedman, H. I. Kaplan,
& B. J. Sadock (Eds.), Comprehensive textbook of psychiatry—II.
Baltimore: Williams & Wilkins, 1975.
Minuchin, S. Families and family therapy. Cambridge, Mass.: Harvard
University Press, 1974.
Mischel, W. Toward a cognitive social learning reconceptualization of
personality. Psychological Review, 1973, 80, 252-283.
Mozdzierz, G., Macchitelli, F., & Lisieki, J. The paradox in
psychotherapy: An Adlerian perspective. Journal of Individual
Psychology, 1976, 32, 169-184.
Nunberg, H. The synthetic function of the ego. Practice and theory of
psychoanalysis. New York: International Universities Press, 1960.
Oppenheimer, R. Analogy in science. American Psychologist, 1956, 11,
127-135.
Overton, W. F., & Reese, H. W. Models of development:
Methodological implications. In J. R. Nesselroade & H. W. Reese
(Eds.), Life-span developmental psychology: Methodological issues. New
York: Academic Press, 1973.
Overton, W. F., & Reese, H. W. Conceptual prerequisites for an
understanding of stability-change and continuity-discontinuity.
International Journal of Behavioral Development, 1981, 4, 99-123.
Piaget, J. Six psychological studies. New York: Random House, 1967.
Progoff, I. Jung, synchronicity and human destiny: Non-causal dimensions of
human experience. New York: Julian Press, 1973.
Raskin, D. E., & Klein, Z. E. Losing a symptom through keeping it: A
review of paradoxical treatment techniques and rationale. Archives
of general psychiatry, 1976, 33, 548-555.
Ratner, C. Principles of dialectical psychology. Telos, 1971, 9, 83-109.
300 CHAPTER 10
Reese, H. W., & Overton, W. R Models of development and theories of

development. In L. R. Goulet & R B. Baltes (Eds.), Life-span
developmental psychology: Theory and research. New York: Academic
Press, 1970.
Riegel, K. E Dialectical operations: The final period of cognitive
development. Human Development, 1973, 16, 346-370.
Riegel, K. E The dialectics of human development. American
Psychologist, 1976, 31, 689-700.
Riegel, K. F. (Ed.). Foundations of dialectical psychology. New York:
Academic Press, 1979.
Rogers, C. R. Client-centered therapy: Its current practice, implications, and
theory. Boston: Houghton Mifflin, 1951.
Rogers, C. R. A theory of therapy, personality, and interpersonal
relationships, as developed in the client-centered framework. In S.
Koch (Ed.), Psychology: A study of a science, 3. Formulations of the
person and the social context. New York: McGraw-Hill, 1959, 184-256.
Rogers, C. R. On becoming a person: A therapist's view of psychotherapy.
Boston: Houghton Mifflin, 1961.
Rogers, C. R. Client-centered psychotherapy. In H. I. Kaplan, B. J.
Sadock, & A. M. Freedman (Eds.), The comprehensive textbook of
psychiatry—III. Baltimore: Williams & Wilkins, 1980. (a)
Rogers, C. R. A way of being. Boston: Houghton Mifflin, 1980. (b)
Rychlak, J. F. A philosophy of science for personality theory. Boston:
Houghton Mifflin, 1968.
Rychlak, J. F. (Ed.). Dialectics: Humanistic rationale for behavior and
development. Basel, Switzerland: Karger, 1976.
Schrodinger, E. What is life? The physical aspects of the living cell.
Cambridge, Eng.: Cambridge University Press, 1967.
Selvini-Palazzoli, M., Boscolo, L., Cecchin, G., & Prata, G. Paradox and
counterparadox: A new model in the therapy of the family in schizophrenic
transaction. New York: Jason Aronson, 1978.
Sinnott, J. D. The theory of relativity: A metatheory for development?
Human Development, 1981, 24, 293-311.
Speck, R., & Attneave, C. Family networks. New York: Pantheon Books,
1973.
Stanton, M. D. Family therapy: Systems approaches. In G. P. Sholevar,
R. M. Benson, & B. J. Blinda (Eds.), Treatment of emotional disorders
in children and adolescents. Jamaica, N.Y.: Spectrum, 1980.
Staats, A. W. Paradigmatic behaviorism, unified theory, unified theory
construction methods, and the Zeitgeist of separatism. American
psychologist, 1981, 36, 239-256.
Stierlin, H. Conflict and reconciliation. New York: Science House, 1969.

Watzlawick, P., Beavin, J., & Jackson, D. Pragmatics of human
Watzlawick, P., Weakland, J., & Fisch, R. Change: Principles of problem
formation and problem resolution. New York: Norton, 1974.
Weeks, G. R., & L'Abate, L. A compilation of paradoxical methods.
Mazel, 1982.
Weeks, G. R., & Wright, L. Dialectics of the family life cycle. American
Journal of Family Therapy, 1979, 7, 85-91.
Zukav, G. The dancing Wu Li masters: An overview of the new physics.
New York: Bantam, 1979.
11
A Metatheory of Paradox
by Gerald R. Weeks, Ph.D.
The purpose of this volume is to help define the term therapeutic

paradox and to describe ways of working paradoxically. W hile all
the contributors appear to work in a sim ilar manner, their expla
nations or theories about how they work differ. The aim of this
chapter is to create a theory of paradox that transcends any par
ticular theory of paradox and any particular approach to psycho
therapy. A n attem pt is m ade to develop a m etatheory of paradox
in psychotherapy.
PARADOX AS A UNIVERSAL ASPECT OF

THERAPY
W eeks (1977) and W eeks and LAbate (1982) proposed a dialec
tical theory of change for psychotherapy. This theory provided the
foundation for the application of paradoxial strategies in therapy.
A s early as 1977, W eeks suggested that the com m on elem ent of
all psychotherapies was paradox. In their 1982 book, W eeks and
LAbate compared and analyzed the Adlerian, Behavioral, Gestalt,
Logotherapeutic, D irect A nalysis, Provocative, and H ypnothera
peutic schools to dem onstrate the com m on elem ent of paradox.
302
A METATHEORY OF PARADOX 303
Seltzer (1986) has significantly extended W eeks and L'Abate's

(1982) work in this area. He presents a thoroughgoing analysis of
paradox in both Eastern and W estern approaches to therapy.
Seltzer gives us an idea of ju st how prevalent paradoxical thinking
is in other schools of therapy by considering the labels used to
describe paradoxical techniques w ithin a few:
From the psychoanalytic perspective, which includes the work of

paradigmatic psychotherapists, we have inherited the descriptors
"antisuggestion," "going with the resistance," "joining the resist
ance," "reflecting" (or "mirroring") the resistance," "siding with
the resistance," "paradigmatic exaggeration," "supporting the
defenses," "reductio ad adsurdum," "reenacting an aspect of the
psychosis," "mirroring the patient's distortions," "participating in
the patient's fantasies," "outcrazying the patient," and "the use of
the patient as consultant." From the vantage point of behavior ther
apy, we may appreciate paradoxical elements in such procedures
as "blowup," "implosion," "flooding," "instructed helplessness,"
"massed practice," "negative practice," "paradoxical intention,"
"stimulus satiation," and "symptom scheduling." In gestalt ther
apy, an approach where the term paradox is rarely employed, the
attempt to foster change paradoxically may be recognized in the
therapist's cruel-to-be-kind suggestions to "stay with the (negative)
experience," or to "exaggerate the feeling" (sensation, experience,
speech, movement, etc.). (p. 20)
Seltzer's (1986) theoretical analysis led him to develop a m eta

theory of paradox in psychotherapy. His m etatheory can be further
distilled to a theory based on social psychology developed by
Strong and Claiborn (1982). Seltzer (1986) devotes considerable
attention to the issue of how different schools of therapy are
defined. He believed, like Weeks (1977), that the apparently sharp,
theoretical differences am ong approaches begin to disappear
when the clinician is viewed in the here-and-now context of seeing
client(s). This idea is one that has been heard in clinical circles for
m any years. He argues that paradoxial strategies are present in
all system s of therapy. These strategies share the common element
of defying the clients' expectations and involve som e form of
refram ing and/or sym ptom prescription.
A m etatheory is required to bind together the various
approaches. O m er (1981) was am ong the first theorists to proffer
a unified concept of paradox. In an effort to arrive at the single
common denominator of therapeutic efficacy, he proposed the con
cept of sym ptom decontextualization. T his idea con sists of the
therapist m odifying both the form and the context of the sym p
304 CHAPTER 11
tom. Form, in his definition, refers to the request to continue the

sym ptom under different conditions (e.g., scheduling, exaggera
tion, different time or place). Context refers to the fact that what
one had been trying to stop is now allowed expression. In Omer's
view, the second change is one of m eaning. The client m ust alter
his or her attitude toward a behavior if it is allowed expression.
Omer's (1981) use of the term decontextualization is not as
descriptive as Deissler's (1985) use of the term recontextualization.
The m eanings of these two term s are similar. D eissler states that
in sym ptom decontextualization the sym ptom is changed by alter
ing the "recursive context" (i.e., the num ber of people involved
in the problem and/or the tem poral and spatial contexts). It seem s
more descriptive to say that the sym ptom is recontextualized.
W hen a client is able to change the context of the sym ptom , the
m eaning inevitably changes only because the client is able to dem
onstrate som e control over the uncontrollable, som e volition over
the involitional, and som e m indfulness over the spontaneous
(m indless or automatic behavior).
W hat do all therapies have in com m on w hich is inherent in the
paradoxial view of therapy?
Seltzer (1986) states, "all therapies can be perceived as endeavoring
to assist clients in com prehending the voluntariness— and their
controllability— of behaviors w hich have com e to appear nonvoli-
tional" (p. 164). Seltzer concludes that all paradoxical strategies
decontextualize sym ptom s. In short, our metatheory is that paradox
ical strategies change the m eaning of the sym ptom from that w hich
is uncontrollable to that which is controllable. This view is consistent
w ith the core of Weeks and L'Abate's (1982) approach, w hich offers
five principles for paradoxical intervention. These principles deal
with both form and context. The third principle requires that the ther
apist change the direction of control of a sym ptom . In an individual,
the sym ptom is placed under voluntary control, and for system s, the
sym ptom is placed under the control of those w ho had allowed the
sym ptom to heretofore control them.
This m etatheory of paradox can now be used to exam ine the
universal aspects of therapy. Seltzer (1986) discussed this problem
from several different perspectives. First, the nature of the ther
apeutic relationship is such that the therapist m ust be in control.
Establishing a therapeutic relationship is synonymous with taking
control. However, the control the therapist takes is not to eliminate
the problem, but to help the client eliminate the problem. The ther
apist is then taking charge by placing the client in charge. For
exam ple, in the client-centered school of Rogers, the client is
placed in charge through the therapist's attitude of unconditional
positive regard. A behaviorist would place the client in charge by

being directive in using techniques and stating that these are only
tools and the client m ust supply the rest. All therapy is built on
the paradox of taking control by giving it away. It is only under
this condition that the client can develop self-control.
Second, Seltzer (1986) argues that the therapist should m aintain
a positive view of sym ptom s. In other words, in all system s of
therapy, the sym ptom is never directly attacked. The sym ptom is
seen as serving som e need of the client. It is im portant to under
stand, not attack, the sym ptom . This attitude is translated into
action via a perm issiveness toward sym ptom s. Even behaviorists
give clients perm ission to have their sym ptom s during the base
line period. W eeks and LAbate (1982) found that sym ptom s
always represented som ething positive. Sym ptom s have been
defined as allies, com m unications, existential statem ents, and
vehicles of change. In fact, they found that the newer schools of
therapy, especially the strategic school, started to define sym p
toms in more explicit, positive ways. Weeks and LAbate (1982) and
countless other therapists have proposed that sym ptom s be
viewed as "frien d s" and not "th e en em y " as m any clients believe.
The third universal factor is the therapist's view o f not only the
sym ptom , but the person. The client enters therapy expecting to
be changed and fearing the therapist may disapprove because of
his or her sym ptom atic behavior. The client discovers that the ther
apist is attentive, supportive, empathetic, friendly, and receptive.
The therapist m aintains his or her attitude of acceptance, includ
ing acceptance of resistance; and the client's effort to resist not only
fails, but is somehow redefined in a way that indirectly brings con
trol back to the therapist.
The therapeutic relationship is m arked by a num ber of charac
teristics that run counter to the client's expectations. It is a unique
social relationship in several paradoxical ways. The therapist
shows an attitude of detached concern or uninvolved involvement
and takes control by giving it away. This process is accom plished
by being indirectly direct (Seltzer, 1986). The goal is to help the
client change spontaneously (Strong & Claiborn, 1982; W eeks &
L'Abate, 1982). C han ge is not directly attributed to som ething the
therapist did. C hange is defined through the interaction betw een
the therapist and client, w ith the client attributing change to him-
or herself.
The fourth unifying theme is the paradoxical nature of the ther
apeutic process. Seltzer (1986) sum m arizes this idea in term s of
the client(s) working through, rather than around. In every system
of therapy, the client is encouraged to move toward the sym ptom
306 CHAPTER 11
rather than flee from it. Otherw ise, the therapy would simply sup
port the avoidant behavior of the client. In the psychodynam ic lit
erature, the term for this process is literally called working
through (Singer, 1970).
Through som e type of conditioning procedure, the behaviorist
has the client focus attention on the sym ptom in order to reduce
the anxiety associated with it. Gestalt therapists encourage the cli
ent to stay w ith the feeling. In fact, B eisser (1970) states the heart
of G estalt therapy is a paradoxical theory of change. He asserts
that change occurs w hen one becom es w hat he is, not w hen he
tries to becom e w hat he is not. In long-term analytic therapy, the
therapist assum es that change will occur very slowly and only
after a therapeutic relationship involving transference has
occurred, precipitating greater insight. The underlying m essage
to the client is "D o n 't change quickly and be w ho you a re ."
Seltzer's (1986) two final points deal w ith the paradox of how
the therapist takes responsibility for the client taking responsibil
ity for his or her problem s. In short, the com m on denom inator for
different therapies is that the client learns to exercise self-control.
The therapist m ust convince the client that she or he can learn to
help him - or herself. This task may be accom plished directly or
indirectly. The responsibility for change is always put back on the
client. The therapist provides a framework for this task— not
ready-m ade solutions.
W hen clients accept that they can do som ething to alleviate the
sym ptom atic behavior, they are also forced to accept the fact that
their sym ptom s m ust be under their control. By definition, a
sym ptom is defined by the client as a behavior that is uncontrol
lable, involuntary, and spontaneous. Every system of therapy
seeks to teach the client that sym ptom s are behaviors that are con
trollable, voluntary, and volitional. Every system of therapy rec
ognizes and has techniques to deal w ith clients w ho deny,
disqualify, and externalize responsibility for sym ptom atic behav
ior. O n the contrary, every system has a set of rules w hich allows
the therapist to deny, disqualify, and externalize any responsibil
ity for the occurrence of change. This set of contradictory condi
tions form s a context in w hich the attribution for change m ust
belong to the client.
Weeks and L'Abate (1982) fully recognize this framework. They
note:
The meta-level structuring of paradoxical therapy is not one of

restriction. On the contrary, the intervention allows the client the
fullest expression of freedom in developing a new frame of refer
ence. From the outset of therapy, the meta-framework has been one
of initiating changes. The fact that clients respond to paradox by
attributing change to themselves serves as evidence of this view,
(p. 246)
Seltzer (1986), W eeks and L'Abate (1982), and Stanton (1984)

have shown how paradoxical strategies are deeply em bedded in
m any system s of psychotherapy. Seltzer (1986) extends this work
through his theoretical analysis to show that paradox is a unifying
concept w hich is universal in the various system s of therapy. In
spite of the fact that contents differ, the therapeutic process at the
m etatheoretical level rem ains constant. This m etatheoretical
understanding provides a new framework for integrating differ
ential theories o f therapy.
A n attem pt at defining a m eta- or unified theory that incorpo
rates both paradoxical and other m ethods of therapeutic change
is the "geod ynam ic balan ce" theory of Stanton (1984). Stanton
places sym ptom s w ithin their interpersonal (e.g., fam ilial) context,
a context that both generates and m aintains them , and also one
that can be utilized by the therapist for sym ptom elimination. The
sym ptom is seen as one elem ent in an interpersonal sequ ence— a
series of behavioral events am ong a group of people, such as a
family. A s each m em ber of the group chim es in w ith his or her
contribution to the sym ptom sequence, he or she is regarded as
moving, at least symbolically in a kind of "orbit" toward some peo
ple and away from others; i.e., "in the service" o f som e, and not
of others. The whole sequence thus takes on the im age of a cho
reographed dance, in w hich the sym ptom is only one move in the
dance. Such dances can be identified w hen one observes the
sym ptom as it actually occurs in the real world, that is, w ithin its
norm al interpersonal context.
In order to bring about change, the therapist has two general
options. The "d iversio n " approach attem pts to directly block or
redirect the sequence, stopp in g it and perhaps offering an alter
native, whereas the "com pression" option (paradoxically) exagger
ates a person's orbit by encouraging or pushing the person further
in the direction he or she is already m oving. A n exam ple would
be to try to get the teenaged son in an enm eshed mother-son rela
tionship to actually sit on the mother's lap, and also to have the
m other accom pany the son to school and hold h is hand en route.
Com pression moves usually result in a rebound effect, in w hich
the m em bers counteract the intervention and stop doing w hat
they were doing. In contrast, diversion, in order to be successful,
usually requires energy that is equal to or greater than that of the
308 CHAPTER 11
behavior being blocked. Stanton (1984) notes eight therapeutic

approaches that predom inantly utilize diversion techniques and
15 therapies that em phasize compression (e.g., paradoxical) m eth
ods. In its delineation of a two-sided or opponent-process feature
of therapy— going w ith the system or sym ptom versus counter
acting it— good dynam ic balance theory is consonant w ith both
Taoist philosophy and Bopp and Weeks' (1984) formulations about
the therapeutic dialectic. Perhaps, a distinctive feature of Stanton's
theory is that it synthesizes alm ost all therapeutic tech niqu es—
both paradoxical and non-paradoxical— under one conceptual
um brella, m aking room for either approach depending on the
nature, difficulty, and chronicity o f the problem .
The bridge betw een the m etatheoretical work of the authors
reviewed thus far and the integration of differential therapies has
been developed by Strong and Claiborn (1982). These theorists
proposed a theory of change based on the principles and research
of social psychology and psychotherapy.
The Paradox of Spontaneous Compliance

Strong and Claiborn (1982) identified two types of change pro
cesses. The first type is forced change or compliance. Forced com
pliance refers to a change in oneself that is attributed to another
person. The experience of change is of doing som ething different,
rather than being different. Every parent, spouse, and therapist
who has ever tried to force another person to be (act) different
know s this effort is futile. If com pliance does occur, it is only
under duress, and will persist only as long as is needed to placate
the other person. This understanding prevents therapists from
telling clients w hat to do or giving advice.
Therapeutic change is best defined as the therapist's ability to
create the context for spontaneous com pliance. The term itself is
contradictory. How does the therapist get the client to do w hat the
therapist wants, w ithout getting the client to do w hat he or she
wants? The therapist must take control without appearing to take
control. Therapy is a paradox.
The idea of spontaneous com pliance may seem unusual; how
ever, it is a universal h um an experience. W hen we want som eone
to go to lunch w ith us, we do not say, "You will go to lunch w ith
m e at 12 o 'clo ck ." You m ight say, "W ould you like to go to lunch
w ith m e at 12 o'clock?" In m arriage, spouses do not give each
other orders. If they do, they find them selves in an untenable sit
uation. Spouses attem pt to influence each other indirectly by m ak
ing statem ents such as "I want to be w ith y o u ," or "W ould you
like to talk ?" W hen spouses attem pt to control through forced

compliance, dysfunctionality results. The therapist working with
couples m ust help each m em ber of the couple learn how to take
control indirectly by taking control of the couple w ithout appear
ing to take control.
The "affirm ation " and "n eg atio n " paradoxes are used to create
spontaneous com pliance (Strong & C laiborn, 1982). The affirm a
tion paradox (implicitly paradoxical) is m ade up of three elements:
1. The therapist presents the desired behavior and insists that the
behavior be ad opted as part of the d efinition of the
relationship.
2. The therapist com m unicates that change is a result of proc
esses internal to the client and is not in com pliance w ith the
therapist.
3. The therapist identifies an agent responsible for change that
acts beyond the client's volitional control, (p. 145)
The therapist w ants to create a context in w hich the client spon

taneously com plies by attributing change to self. The com m uni
cations therapist may teach the couple new ways of talking. Such
techniques are useless unless the clients want to learn to use them.
Thus, the couple can attribute success to their w anting to work
together in com m unicating more effectively.
The therapist using the negation paradox (explicitly paradoxical
strategies) encourages the client to change by encouraging the
client not to change. In fact, that behavior w hich is not to be
changed has been defined as involuntary by the client. By encour
aging the involuntary behavior, the therapist brings it under the
control of the individual, w hich ch an ges its m eaning. In couples
and fam ilies, the direction of control is changed by placing the
person being controlled in charge. This task is done by refram ing
the behavior in such a way that it is good for the other person
or it is perceived in such a way that the other has control over
the sym ptom (W eeks & L'Abate, 1982). In both cases, the clients
must change their behavior vis-a-vis the therapist in order to gain
control.
The thesis of this chapter is that all psychotherapy has a para
doxical com ponent. It has been shown that paradoxical strategies
and techniques are em bedded in m any system s of therapy— a fact
w hich has not been recognized or adm itted because of practition
ers' adherence to their own theoretical biases. The metatheoretical
analysis of psychotherapy has shown how change is essentially
facilitated in the sam e way across therapies. The therapist adopts
310 CHAPTER 11
a num ber of paradoxical stances vis-a-vis the client. This

metatheoretical description has been reduced to an even more par
sim onious description by show ing that therapy is som e com bina
tion of affirm ation and negation paradoxes (Strong & C laiborn,
1982).
The goal of therapy is to en han ce its paradoxical nature— to
work toward creating conditions of spontaneous com pliance.
There are two basic therapeutic processes w hich help to achieve
this end. These processes are refram ing and prescriptions.
R efram ing. Refram ing is the m ost common strategy used in the
system s therapies as well as in psychotherapy in general. It is fun
damental for psychotherapy because it helps the client change per
spective in such a way that ch an ge is easier to m ake. System s
therapists have referred to this concept as: relabeling (Haley 1973);
refram ing (Watzlawick et al., 1974); content refram ing (Bandler &
Grinder, 1982); redefinition (Andolfi, 1979); seeing the good
(LAbate, 1975); positive connotation (Palazzoli, et al. 1978); ascrib
ing noble intentions (Stanton, Todd, & A ssociates, 1982); non
blam ing (Alexander & Parsons, 1982); and context m arkers
(Bateson, 1979; LAbate, G anahl, & H an sen , 1986; Viaro, 1980).
Watzlawick, W eakland, and Fisch (1984) were am ong the first
to discuss the central role of refram ing in therapy. They defined
refram ing as changing the conceptual and/or em otional m eaning
attributed to a situation. The behavior that is refram ed is the
behavior that has been defined or fram ed as bein g sym ptom atic
by the client.
A refram ing statement is quite different from an interpretation.
A n interpretation carries som e truth value. The therapist actually
believes the statement represents some aspect of reality. Refram ing
statem ents are not intended to have the sam e validity. The theory
of truth w hich is used by the therapist is pragm atic. In the prag
matic theory of truth, that which works is considered true (James,
1907). The therapist attem pts to construct a view of reality that
is more conducive to change, rather than replace the client's faulty
world view w ith one that is correct (Kelly, 1955). Tennen, Eron,
and Rohrbaugh (1985) have stated:
Adaptive functioning does not require being in touch with reality

and having an accurate view of the world. Rather, any world view
that interdicts problem sequences is considered in planning inter
ventions. This position is nested not only in "functional theory,"
but in growing empirical literature pointing to the healing effects
of illusion (Lazarus, 1983), the illusion of control (Alloy &
Abramson, 1979; Tiger, 1979) and the misperception of one's suc
cess and its causes (Greenwald, 1980; Muller & Ross, 1975). High
self-esteem (Tennen & Herzberger, submitted for publication;
Tennen et al., submitted for publication), lack of depression
(Lewinsohn et al., 1980; Abramson & Alloy, 1981) and better adjust
ment to illness and injury (Bulman & Wortman, 1977; Taylor, 1983;
Tennen et al., in press) have been associated with >ionveridical and
illogical perceptions and beliefs. The shift from psyche to system
and from reality testing to useful illusions may represent the most
significant contributions of the strategic therapist in his/her use of
paradox, (pp. 199-200)
The use of refram ing in the literature usually has two m eanings
(W eeks & L'Abate, 1982). O ne is to change the way in w hich a
sym ptom is defined in term s of som e polarization (such as good
versus bad and crazy versus sane). This use of refram ing stem s
from the various m odels of psychotherapy, including m edical,
moral, psychological, statistical, and personal discom fort. A
sym ptom defined in term s of the moral m odel is w rong, sinful,
or bad. Individuals frequently attribute bad intent to behavior.
This attribution leads to the moral perspective that they are okay
and the behavior is bad. R efram ing can be used to change the
value attributed to the problem . For exam ple, couples usually
believe their fights are exclusively negative behaviors. The fights
have been framed as destructive, negative, and so on and are seen
as representing negative intent. In refram ing the fighting behavior,
the therapist wants to change the attributed m eaning from bad
to good. For instance, the therapist m ight say,
The two of you must care a great deal about each other and yourself,
because you invest so much of your energy in fighting. Couples
who don't care or are indifferent don't fight. Your fighting shows
there is something worth fighting for in spite of the fact that you
may end up appearing to fight against each other.
The second use o f refram ing is to move the focus from the indi
vidual to the system . W hen couples or fam ilies present problems,
there is usually a sym ptom bearer who is carryin g the problem
for the rest of the family. The other m em ber(s) do not see the con
nection betw een their behavior and the behavior of the "sick " one.
In couples, the "h ea lth y " spouse externalizes and/or denies any
responsibility for the problem . The attributional strategy in the
couple is linear, not circular. O ne of the therapeutic tasks is to get
them to see how the problem stem s from their interaction.
Reframing is the method whereby the therapist can move the cou-
312 CHAPTER 11
pie from a linear attributional strategy to an interactional or cir

cular attributional strategy.
The two uses of refram ing described above are usually com
bined in a statement given to the client system . Palazzoli, Cecchin,
Prata, and Boscolo (1978) call this technique positive connotation.
The sym ptom is given a positive m eaning and all the m em bers
of the system are linked together. R efram ing as defined by
Palazzoli and colleagues (1978) addresses all the factors com pris
ing Strong and Claiborn's (1982) m odel. R efram ing ch an ges the
definition of the relationship, changes the m eaning of the behavior
by altering or disrupting the interpretative framework, and dis
rupts one's ability to predict another's behavior. It also puts the par
ticipants on the sam e level so that one cannot take advantage of
the other, thereby creating greater congruence.
O ne goal of the refram e is to prescribe a shared m eaning in the
relationship (e.g., fighting m eans you care). It also increases inter
dependence by suggesting that both want the sam e goal for the
relationship. Finally, the m ost significant effects are to change the
attributions from linear to circular and from negative to positive.
If the refram e is successful and the couple believe they both
deserve the sam e thing, then they m ust ask them selves if there
aren 't better ways to achieve it, rather than engaging in the sam e
endless gam e. The result should be new efforts to change one
spouse's impressions of the other by changing his or her behavior.
The use of refram ing in psychotherapy and m arital therapy
deserves considerable attention. This discussion centers mostly on
the theoretical aspects of refram ing. Several recent reviews offer
greater understanding: Weeks and LAbate (1982); L'Abate, H ansen
and G anahl (1986); and Jones (1986).
In summary, reframing the sym ptom s has multiple effects. The
m ost im portant effects in system therapy are to change the linear
attributional strategy of the couple to a circular attributional strat
egy and to change the attribution o f m eaning given to the sym p
tom to one w hich is positive. Positive actually refers to some
aspect o f the relationship w hich helps the couple bond. The pos
itive dynam ic (e.g., protection) helps to create greater congruence
by defining som ething both desire. O nce both partners see how
they participate in the system , they are ready to work coopera
tively rather than competitively. Additionally, by attributing pos
itive intention(s), each partner is invited to try different strategies
to m anage the other, w hich m eans each partner m ust change.
Although refram ing is an essential part of therapy, it is usually
not sufficient to bring about change by itself. W eeks and L'Abate
(1982) present cases in w hich refram ing was enough to effect
change, but these are the exceptions. Refram ing sets the stage for
the second phase of therapy, w hich is prescriptive in nature.
Prescriptions. The second m ajor task of the therapist is to pro

vide prescriptions for change. Refram ing was used to create a con
text for change by altering the m eaning of the problem. Refram ing
operates at the cognitive/perceptual level. The client is not asked
to change behavior, although it's implied. The therapist must now
proceed w ith a strategy designed to bring about ch an ge in the
behavioral/affective areas. T his task is accom plished through the
use of "d irectives" (Haley, 1976) or "p rescrip tio n s" (L'Abate et al.,
1986; Weeks & L'Abate, 1982). These two term s may be used inter
changeably. The author has chosen to use the term prescription.
A prescription is a set of instructions or in ju nctions the client is
requested to follow. Unlike refram ing, the prescription possesses
som e dem and characteristics for change. The prescription may
take a variety of forms and may be given in session or extrasession
(homework).
It may seem too sim plistic to reduce all the various approaches
of therapy to refram ing and prescriptions. However, by view ing
psychotherapy as a coherent or system atic set of prescriptions, it
is easier to break the bond of adherence to a particular theoretical
system . Nichols (1984) recognizes the use of directives as one of
the m ost widespread techniques in fam ily therapy. He states:
Experientialists use directives to promote affective experiences in

therapy sessions; behaviorists use directives to teach parents new
ways of disciplining their children; structural family therapists use
directives in the form of tasks between sessions; Bowenian thera
pists use directives to advise patients how to improve relations with
their parents; and strategic therapists use directives to outwit resist
ance and provoke change, (p. 87)
Prescriptions fall w ithin the two categories of affirm ation and

negation paradoxes. In either case, all prescriptions are essentially
paradoxical w ithin the context o f therapy because the therapeutic
encounter is essentially paradoxical. W hatever prescriptions are
used m ust be congruent w ith the paradoxical context of therapy
(Seltzer, 1986). The reader will recall that Seltzer discusses (a) tak
ing control by giving it away; (b) m aintaining a positive view of
the sym ptom and the person; (c) not attem pting to force change;
(d) allowing the client to have the sym ptom ; (e) helping the client
attribute change to self rather than outside forces; and (f) placing
314 CHAPTER 11
the client in charge of the sym ptom so that it is seen as control

lable, voluntary, and volitional.
Ultimately, change results from the therapist's ability to change
the direction of control in a relationship. A prescription is required
w hich will change the control strategy used in the relationship
w ithout leading to one's suprem acy or surrender. A sense of self-
control and m utual control is the aim . Sym ptom atic behavior
involves the denial of both self-control (I can 't help m y behavior)
and m utual control (I am not trying to tell you how to behave).
W hen more appropriate control strategies are found, there is no
longer a need for sym ptom atic behavior.
REFERENCES
Alexander, J., & Parsons, B. Functional family therapy. Monterey, CA:
Brooks, Cole, 1982.
Andolfi, M. Redefinition in family therapy. American Journal of Family
Therapy, 1979, 7, 5-15.
Ascher, M., Bowers, M., & Schotte, D. A review of data from controlled
case studies and experiments indicating the clinical offering of
paradoxical intention. In G. Weeks (Ed.), Promoting change through
paradoxical therapy (pp. 216-251). Homewood, IL: Dow Jones-Irwin,
1985.
Bandler, R., & Grinder, J. Reframing: Neurolinguistic programming and the
transformation of meaning. Moab, UT: Real People Press, 1982.
Bateson, G. Mind and nature: A necessary unity. NY: Bantam Books, 1979.
Beisser, A. The paradoxical theory of change. In J. Fagan & R. Shepard
(Eds.), Gestalt therapy now (pp. 77-80). NY: Harper & Row, 1970.
Bopp, M., & Weeks, G. Dialectic metatheory in family therapy. Family
Process, 1984, 23, 49-61.
Deissler, K. Beyond paradox and counterparadox. In G. Weeks (Ed.),
Promoting change through paradoxical therapy (pp. 60-99).
Homewood, IL: Dow Jones-Irwin, 1985.
Erickson. New York: Ballantine, 1973.
Haley, J. Problem-solving therapy. San Francisco: Jossey-Bass, 1976.
James, W. Pragmatism. New York: World Publishing, 1907.
Jones, W. Frame cultivation: Helping new meaning take root in families.
American Journal of Family Therapy, 1986,14, 57-68.
Kelly, G. The psychology of personal constructs, Vol. 1, A history of

personality. New York: W. W. Norton, 1955.
L'Abate, L. A positive approach to marital and family intervention. In
L. Wolberg & M. Aronson (Eds.), Group therapy 1975— an overview.
New York: Stratton Intercontinental Medical Books, 1975.
L'Abate, L., Ganahl, G., & Hansen, J. Methods of family therapy.
Englewood Cliffs, NJ: Prentice Hall, 1986.
Nichols, M. Family therapy: Concepts and methods. New York: Gardner
Press, 1984.
Omer, H. Paradoxical treatments: A unified concept. Psychotherapy:
Theory, Research, and Practice, 1981, 18, 320-324.
Palazzoli, M., Cecchin, M., Prata, G., & Boscolo, L. Paradox and
Seltzer, L. Paradoxical strategies in psychotherapy: A comprehensive overview
and guidebook. New York: Wiley, 1986.
Singer, E. Key concepts in psychotherapy. New York: Basic Books, 1970.
Stanton, M. Fusion, compression, diversion, and the workings of
paradox: A theory of therapeutic/systemic change. Family Process,
1984,23,135-167.
Stanton, M., Todd, T., & Associates. The family therapy of drug addiction.
New York: Guilford, 1982.
Strong, S., & Claiborn, C. Change through interaction. Social psychological
processes of counseling and psychotherapy. New York: John Wiley, 1982.
Tennen, H., Eron, J., & Rohrbaugh, M. Paradox in context. In G. Weeks
(Ed.), Promoting change through paradoxical therapy (pp. 187-215).
Homewood, IL: Dow Jones-Irwin, 1985.
Viaro, M. Case report: Smuggling family therapy through. Family
Process, 1980,19, 253-271.
Watzlawick, P., Weakland, J., & Fisch, R. Change: Principles of problem
formation and problem resolution. New York: W. W. Norton, 1974.
Weeks, G. Toward a dialectical approach to intervention. Human
Development, 1977, 20, 277-292.
Weeks, G., & L'Abate, L. Paradoxical psychotherapy: Theory arid practice
with individuals, couples, and families. New York: Brunner/Mazel,
1982.

Paradoxical Intention

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Paradoxical Intention

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Paradoxical Intention

Uploaded by

Copyright:

Available Formats

Promoting Change

All Rights Reserved

This e-book contains material protected under International and

Created in the United States of America

For information regarding this book, contact the publisher:

International Psychotherapy Institute E-Books

Dr. Weeks is a Professor at the University of Nevada-Las Vegas and

Dr. Weeks is among a handful of individuals to be awarded the

In 1982, he published Paradoxical Psychotherapy: Theory and Practice

For additional information about Dr. Weeks please google, “Gerald R.

1. A D elp h i Study o f Paradox in

Part Two A pproaches to Paradoxical Intervention

2. U n p red ictab ility and C hange: A H olographic

3. Beyond Paradox and C ou n terp arad ox...................................... 60

4. Paradoxical Intention .......................................................................99

5. Paradoxical Techniques: O ne Level o f

6. E ricksonian Styles o f Paradoxical T reatm en t..................... 134

7. Paradox in C o n te x t.......................................................................... 187

Part T h ree R esearch on Paradoxical Techniques

Part Four Future D irections

10. C ontradiction and Its R esolu tion am ong

11. A M etatheory o f Paradox ............................................................ 302

L. Michael Ascher, Ph.D. Professor and Director of the Behavior

Psychology Program at Georgia State University. Dr. L'Abate has

A few years ago paradox in psychotherapy was used by only a few

goes over technique. In the third section o f the book M ichael

Since the publication of this book in 1985, articles on paradox

Gerald R. Weeks, Ph.D.

The purpose of this study was to provide an operational defini­

Since the concept of paradox is still in the formative stages, a

An interesting sidelight on the Delphi is found in the m eanings

Step 1. First, each respondent's com plete answers to the ques­

them as feedback), again looking for com m on m eaning units. Here

Step 5. Based on the initial and feedback responses, three litera­

Step 1 . The participants' answers to the first question in this

Step 2: Summary of Theme Substance. The request for an

changing the epistemological frame of reference. O ne respondent

Theme 1. Paradoxical intervention is a term for w hat the thera­

Theme 2. Paradoxical intervention can be viewed in term s of the

a. It is paradoxical w hen a client/family sees the therapist's inter­

Step 4. Nine com posite responses to the Round O ne question

Summary Results from Round Two Questionnaire

Part A : Literature Definitions. The literature definitions fa­

’ References to the literature definitions were not provided in the study.

Part B. Operational Definitions from Panel M embers. The

A paradoxical intervention is one that has an element of surprise

Summary of Results from Round Three

A. A paradoxical intervention can be defined as follows: Within a

onstrated tendency to discount, criticize, forget, postpone, or mis­

In addition to the above three choices, there was one distinc­

with valuing and giving directions. Instead of saying in prose,

m eant by a paradoxical intervention. They are listed below and a

ess. The therapist might handle a therapeutic situation differently

There exists an interpretation or instruction to the family that re­

diction is set up. This contradiction is often called a "paradoxical

This system s-oriented view puts m uch more em phasis on the

a. Paradoxical intervention as a term which specifies som ething

The purpose of this study was to provide an operational defini

Step 1. First, each respondent's com plete answers to the ques

Step 5. Based on the initial and feedback responses, three litera

Theme 1. Paradoxical intervention is a term for w hat the thera

a. It is paradoxical w hen a client/family sees the therapist's inter

Part A : Literature Definitions. The literature definitions fa

onstrated tendency to discount, criticize, forget, postpone, or mis

In addition to the above three choices, there was one distinc

There exists an interpretation or instruction to the family that re

b. Paradoxical intervention as a term for defining the client's per

The second, more concrete approach to a definition em pha

pecting, or apparently w anting us to be, and then taking a posi

how they wish them to be different. They maintain a "proper" dis

Gradually the implications began to appear broader than in the ear

theme, which individual or subgroup on which to focus an inter

do not exist. My team has left me totally confused. One of my col

been hospitalized. Though under great stress, Enid regularly vis

At the following session, regardless of how effective the interven

2. The Need to Consider "T im e." I will restrict m yself to re

3. The Need to Consider Self-reference, Recursiveness, or Au

however, that there is a vital difference in speaking about individ